entry_point
stringlengths 1
65
| original_triton_code
stringlengths 4.5k
619k
| python_code
stringlengths 208
60.9k
| triton_code
stringlengths 1.15k
275k
| repo_name
stringlengths 7
115
| module_name
stringlengths 1
65
| synthetic
bool 1
class | uuid
int64 0
18.5k
| licenses
sequencelengths 1
6
| stars
int64 0
19.8k
| sha
stringlengths 40
40
| repo_link
stringlengths 72
180
| pytorch_code
stringlengths 200
4.05k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
CosineSimilarity_custom | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/3v/c3vyztzgw3iiqfspecgmyuijbvk6r675ogzi4x5mnjsc4kdyk3gp.py
# Topologically Sorted Source Nodes: [cosine_similarity], Original ATen: [aten.linalg_vector_norm, aten.clamp_min, aten.div, aten.mul]
# Source node to ATen node mapping:
# cosine_similarity => clamp_min, clamp_min_1, div, div_1, mul, pow_1, pow_2, pow_3, pow_4, sum_1, sum_2
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%arg1_1, 2), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [1], True), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 0.5), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%pow_2, 1e-08), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%arg1_1, %clamp_min), kwargs = {})
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%arg0_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, [1], True), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %clamp_min_1 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%pow_4, 1e-08), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%arg0_1, %clamp_min_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %div), kwargs = {})
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0 = async_compile.triton('triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 10, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (x3), xmask)
tmp17 = tl.load(in_ptr1 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr1 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-08
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tmp18 = tmp17 * tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = libdevice.sqrt(tmp27)
tmp29 = triton_helpers.maximum(tmp28, tmp13)
tmp30 = tmp16 / tmp29
tmp31 = tmp15 * tmp30
tl.store(out_ptr0 + (x3), tmp31, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/25/c25qpwkdkb7p2sqprzwstqk3uxuf6n63kmah5qbi67gtdiiylms4.py
# Topologically Sorted Source Nodes: [cosine_similarity, sub], Original ATen: [aten.sum, aten.rsub]
# Source node to ATen node mapping:
# cosine_similarity => sum_3
# sub => sub
# Graph fragment:
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sum_3), kwargs = {})
triton_poi_fused_rsub_sum_1 = async_compile.triton('triton_poi_fused_rsub_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_rsub_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_rsub_sum_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = (xindex // 16)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (64*x1)), xmask)
tmp1 = tl.load(in_ptr0 + (16 + x0 + (64*x1)), xmask)
tmp3 = tl.load(in_ptr0 + (32 + x0 + (64*x1)), xmask)
tmp5 = tl.load(in_ptr0 + (48 + x0 + (64*x1)), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 1.0
tmp8 = tmp7 - tmp6
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cosine_similarity], Original ATen: [aten.linalg_vector_norm, aten.clamp_min, aten.div, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0.run(arg1_1, arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cosine_similarity, sub], Original ATen: [aten.sum, aten.rsub]
triton_poi_fused_rsub_sum_1.run(buf0, buf1, 64, grid=grid(64), stream=stream0)
del buf0
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class CosineSimilarity_custom(nn.Module):
def __init__(self, dim: 'int'=1, eps: 'float'=1e-08):
super(CosineSimilarity_custom, self).__init__()
self.dim = dim
self.eps = eps
def forward(self, x1, x2):
return 1 - F.cosine_similarity(x1, x2, self.dim, self.eps)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0(in_ptr0,
in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp16 = tl.load(in_ptr1 + x3, xmask)
tmp17 = tl.load(in_ptr1 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp19 = tl.load(in_ptr1 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr1 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-08
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tmp18 = tmp17 * tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = libdevice.sqrt(tmp27)
tmp29 = triton_helpers.maximum(tmp28, tmp13)
tmp30 = tmp16 / tmp29
tmp31 = tmp15 * tmp30
tl.store(out_ptr0 + x3, tmp31, xmask)
@triton.jit
def triton_poi_fused_rsub_sum_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = xindex // 16
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 64 * x1), xmask)
tmp1 = tl.load(in_ptr0 + (16 + x0 + 64 * x1), xmask)
tmp3 = tl.load(in_ptr0 + (32 + x0 + 64 * x1), xmask)
tmp5 = tl.load(in_ptr0 + (48 + x0 + 64 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 1.0
tmp8 = tmp7 - tmp6
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_0[grid(256)](
arg1_1, arg0_1, buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_rsub_sum_1[grid(64)](buf0, buf1, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf0
return buf1,
class CosineSimilarity_customNew(nn.Module):
def __init__(self, dim: 'int'=1, eps: 'float'=1e-08):
super(CosineSimilarity_customNew, self).__init__()
self.dim = dim
self.eps = eps
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| Lhx94As/PHO-LID | CosineSimilarity_custom | false | 5,518 | [
"MIT"
] | 1 | 44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | https://github.com/Lhx94As/PHO-LID/tree/44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, dim: 'int'=1, eps: 'float'=1e-08):
super().__init__()
self.dim = dim
self.eps = eps
def forward(self, x1, x2):
return 1 - F.cosine_similarity(x1, x2, self.dim, self.eps)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Symmetric | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/4f/c4fekrlplou6jqs7d22hwo63cvbbodj3arj7a5rnr62q3dnsdcpp.py
# Topologically Sorted Source Nodes: [triu, add], Original ATen: [aten.triu, aten.add]
# Source node to ATen node mapping:
# add => add
# triu => full_default, ge, sub, where
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%unsqueeze, %unsqueeze_1), kwargs = {})
# %ge : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%sub, 0), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%ge, %arg0_1, %full_default), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%where, %permute), kwargs = {})
triton_poi_fused_add_triu_0 = async_compile.triton('triton_poi_fused_add_triu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_triu_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_triu_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y3 = yindex
y1 = (yindex // 4)
tmp3 = tl.load(in_ptr0 + (x2 + (4*y3)), xmask & ymask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp0 = x2 + ((-1)*y0)
tmp1 = tl.full([1, 1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = y0 + ((-1)*x2)
tmp7 = tl.full([1, 1], 1, tl.int64)
tmp8 = tmp6 >= tmp7
tmp10 = tl.where(tmp8, tmp9, tmp4)
tmp11 = tmp5 + tmp10
tl.store(out_ptr0 + (x2 + (4*y3)), tmp11, xmask & ymask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [triu, add], Original ATen: [aten.triu, aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_triu_0.run(arg0_1, buf0, 64, 4, grid=grid(64, 4), stream=stream0)
del arg0_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
class Symmetric(nn.Module):
def forward(self, X):
return X.triu() + X.triu(1).transpose(-1, -2)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_triu_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK:
tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y3 = yindex
y1 = yindex // 4
tmp3 = tl.load(in_ptr0 + (x2 + 4 * y3), xmask & ymask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp0 = x2 + -1 * y0
tmp1 = tl.full([1, 1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = y0 + -1 * x2
tmp7 = tl.full([1, 1], 1, tl.int64)
tmp8 = tmp6 >= tmp7
tmp10 = tl.where(tmp8, tmp9, tmp4)
tmp11 = tmp5 + tmp10
tl.store(out_ptr0 + (x2 + 4 * y3), tmp11, xmask & ymask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_triu_0[grid(64, 4)](arg0_1, buf0, 64, 4,
XBLOCK=4, YBLOCK=32, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class SymmetricNew(nn.Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| Lezcano/tutorials | Symmetric | false | 5,519 | [
"BSD-3-Clause"
] | 1 | 24946b2e6d3d825afed6b35c1c4d618a70a88be8 | https://github.com/Lezcano/tutorials/tree/24946b2e6d3d825afed6b35c1c4d618a70a88be8 | import torch
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
class Model(nn.Module):
def forward(self, X):
return X.triu() + X.triu(1).transpose(-1, -2)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Readout | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ig/cig776c5vnkodl5ylq5rpdauwikblftbceldk4vqgy3ytpdyzgbf.py
# Topologically Sorted Source Nodes: [x, y], Original ATen: [aten.mul, aten.sum]
# Source node to ATen node mapping:
# x => mul
# y => sum_1
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, %unsqueeze), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
triton_poi_fused_mul_sum_0 = async_compile.triton('triton_poi_fused_mul_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex % 64
x1 = (xindex // 4) % 16
x2 = (xindex // 64)
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x1 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (64 + x3), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (16 + x1 + (64*x2)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (128 + x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (32 + x1 + (64*x2)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (192 + x3), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (48 + x1 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tl.store(out_ptr0 + (x4), tmp14, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x, y], Original ATen: [aten.mul, aten.sum]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_sum_0.run(arg0_1, arg1_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
del arg1_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from torch.nn import Module
import torch
import torch.utils.data
def aggregate(x, dim, aggr='add', mask=None, keepdim=False):
"""
Args:
x: (..., A, ..., F), Features to be aggregated.
mask: (..., A, ...)
Returns:
(..., , ..., F), if keepdim == False
(..., 1, ..., F), if keepdim == True
"""
assert aggr in ('add', 'mean')
if mask is not None:
x = x * mask.unsqueeze(-1)
y = torch.sum(x, dim=dim, keepdim=keepdim)
if aggr == 'mean':
if mask is not None:
n = torch.sum(mask, dim=dim, keepdim=keepdim)
n = torch.max(n, other=torch.ones_like(n))
else:
n = x.size(dim)
y = y / n
return y
def readout(x, mask, aggr='add'):
"""
Args:
x: (B, N_max, F)
mask: (B, N_max)
Returns:
(B, F)
"""
return aggregate(x=x, dim=1, aggr=aggr, mask=mask, keepdim=False)
class Readout(Module):
def __init__(self, aggr='add'):
super().__init__()
assert aggr in ('add', 'mean')
self.aggr = aggr
def forward(self, x, mask):
return readout(x, mask=mask, aggr=self.aggr)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch.nn import Module
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex % 64
x1 = xindex // 4 % 16
x2 = xindex // 64
x4 = xindex
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (64 + x3), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (16 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp7 = tl.load(in_ptr0 + (128 + x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (32 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (192 + x3), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (48 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tl.store(out_ptr0 + x4, tmp14, xmask)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_sum_0[grid(256)](arg0_1, arg1_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
del arg1_1
return buf0,
def aggregate(x, dim, aggr='add', mask=None, keepdim=False):
"""
Args:
x: (..., A, ..., F), Features to be aggregated.
mask: (..., A, ...)
Returns:
(..., , ..., F), if keepdim == False
(..., 1, ..., F), if keepdim == True
"""
assert aggr in ('add', 'mean')
if mask is not None:
x = x * mask.unsqueeze(-1)
y = torch.sum(x, dim=dim, keepdim=keepdim)
if aggr == 'mean':
if mask is not None:
n = torch.sum(mask, dim=dim, keepdim=keepdim)
n = torch.max(n, other=torch.ones_like(n))
else:
n = x.size(dim)
y = y / n
return y
def readout(x, mask, aggr='add'):
"""
Args:
x: (B, N_max, F)
mask: (B, N_max)
Returns:
(B, F)
"""
return aggregate(x=x, dim=1, aggr=aggr, mask=mask, keepdim=False)
class ReadoutNew(Module):
def __init__(self, aggr='add'):
super().__init__()
assert aggr in ('add', 'mean')
self.aggr = aggr
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| LichenYang-Jeffrey/GAT-for-COVID-19 | Readout | false | 5,520 | [
"MIT"
] | 1 | 91cc6048f14856f3ef9dfebf2db45e2a36975159 | https://github.com/LichenYang-Jeffrey/GAT-for-COVID-19/tree/91cc6048f14856f3ef9dfebf2db45e2a36975159 | from torch.nn import Module
import torch
import torch.utils.data
def aggregate(x, dim, aggr='add', mask=None, keepdim=False):
"""
Args:
x: (..., A, ..., F), Features to be aggregated.
mask: (..., A, ...)
Returns:
(..., , ..., F), if keepdim == False
(..., 1, ..., F), if keepdim == True
"""
assert aggr in ('add', 'mean')
if mask is not None:
x = x * mask.unsqueeze(-1)
y = torch.sum(x, dim=dim, keepdim=keepdim)
if aggr == 'mean':
if mask is not None:
n = torch.sum(mask, dim=dim, keepdim=keepdim)
n = torch.max(n, other=torch.ones_like(n))
else:
n = x.size(dim)
y = y / n
return y
def readout(x, mask, aggr='add'):
"""
Args:
x: (B, N_max, F)
mask: (B, N_max)
Returns:
(B, F)
"""
return aggregate(x=x, dim=1, aggr=aggr, mask=mask, keepdim=False)
class Model(Module):
def __init__(self, aggr='add'):
super().__init__()
assert aggr in ('add', 'mean')
self.aggr = aggr
def forward(self, x, mask):
return readout(x, mask=mask, aggr=self.aggr)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
AdaINConv2dLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/wi/cwixe2mvtbslnlduj6iu3ijkrf5snrdwvhhqsvnunr7zp3pbtb6h.py
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add
# mv => mul, sum_1
# norm => pow_1, pow_2, sum_2
# truediv => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute, %primals_2), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, None), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 1e-12), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_0 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {5: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 6), equal_to_1=(5,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_0(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp25, None)
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp26, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/zj/czj3q7hwubmlpr6euena2hgafqa2zyahzg2ovq7jfcmwd7h7pead.py
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
# Source node to ATen node mapping:
# mv_1 => mul_1, sum_3
# truediv => div
# Graph fragment:
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %div), kwargs = {})
# %sum_3 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [1]), kwargs = {})
triton_per_fused_div_mv_1 = async_compile.triton('triton_per_fused_div_mv_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_mv_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_mv_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (64*x0)), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/vy/cvykqyjomwxei4fyropo4jqebs4lj5nohnzcebbnfyn5byyue7bz.py
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_1 => add_1
# norm_1 => pow_3, pow_4, sum_4
# truediv_1 => div_1
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, None), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_4, 0.5), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_4, 1e-12), kwargs = {})
# %div_1 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, %add_1), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_2 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_2(in_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp8, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/px/cpxcgb5q54ixp24egg5wrifmrdsxkmow6sly6c226q3lm2gvyou5.py
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
# Source node to ATen node mapping:
# sigma => mul_3, sum_6
# Graph fragment:
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %sum_3), kwargs = {})
# %sum_6 : [num_users=2] = call_function[target=torch.ops.aten.sum.default](args = (%mul_3,), kwargs = {})
triton_per_fused_dot_3 = async_compile.triton('triton_per_fused_dot_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_dot_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_dot_3(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/op/cop3qrnpelmxzprgtvtmh4w3aypjouw7zyvmylhriwz2v7g6bddm.py
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_2 => div_2
# Graph fragment:
# %div_2 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_4, %expand), kwargs = {})
triton_poi_fused_div_4 = async_compile.triton('triton_poi_fused_div_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ia/ciaqbxaapwdirvjneo7orzmp3nsd523g2ea3cyzdaow6bnehgvuj.py
# Topologically Sorted Source Nodes: [x_1, x_2, add_2, var_2, x_3, mul_2, content, x_4], Original ATen: [aten.convolution, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# add_2 => add_2
# content => add_3
# mul_2 => mul_6
# var_2 => rsqrt
# x_1 => convolution
# x_2 => sub
# x_3 => mul_5
# x_4 => gt, mul_7, where
# Graph fragment:
# %convolution : [num_users=3] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %div_2, %primals_5, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%convolution, %view_4), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_6, 1e-08), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_5, %select), kwargs = {})
# %add_3 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_6, %select_1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%add_3, 0), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_3, 0.2), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %add_3, %mul_7), kwargs = {})
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where, 0), kwargs = {})
triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5 = async_compile.triton('triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*i1', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
x1 = (xindex // 4)
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (x0 + (8*x1)), xmask)
tmp14 = tl.load(in_ptr1 + (4 + x0 + (8*x1)), xmask)
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 / tmp3
tmp5 = tmp2 - tmp4
tmp6 = tmp5 * tmp5
tmp7 = tmp6 / tmp3
tmp8 = 1e-08
tmp9 = tmp7 + tmp8
tmp10 = libdevice.rsqrt(tmp9)
tmp11 = tmp5 * tmp10
tmp13 = tmp11 * tmp12
tmp15 = tmp13 + tmp14
tmp16 = 0.0
tmp17 = tmp15 > tmp16
tmp18 = 0.2
tmp19 = tmp15 * tmp18
tmp20 = tl.where(tmp17, tmp15, tmp19)
tmp21 = tmp20 > tmp16
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
tl.store(out_ptr0 + (x2), tmp20, xmask)
tl.store(out_ptr1 + (x2), tmp21, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (64, ), (1, ))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, ), (1, ), torch.float32)
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1; del buf1 # reuse
buf17 = empty_strided_cuda((64, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
stream0 = get_raw_stream(0)
triton_per_fused_add_div_linalg_vector_norm_mv_0.run(buf2, primals_4, primals_2, buf0, buf17, 1, 64, grid=grid(1), stream=stream0)
buf3 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
triton_per_fused_div_mv_1.run(primals_4, buf0, buf2, buf3, 4, 64, grid=grid(4), stream=stream0)
buf5 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_2.run(buf3, buf5, 1, 4, grid=grid(1), stream=stream0)
buf6 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
triton_per_fused_dot_3.run(buf5, buf3, buf6, 1, 4, grid=grid(1), stream=stream0)
del buf3
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
triton_poi_fused_div_4.run(primals_4, buf6, buf7, 256, grid=grid(256), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(primals_1, buf7, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (256, 4, 1, 1), (4, 1, 1, 1))
buf9 = buf8; del buf8 # reuse
buf10 = empty_strided_cuda((256, 4, 1, 1), (4, 1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((256, 4, 1, 1), (4, 1, 1, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_1, x_2, add_2, var_2, x_3, mul_2, content, x_4], Original ATen: [aten.convolution, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5.run(buf9, primals_5, primals_6, buf10, buf11, 1024, grid=grid(1024), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [], Original ATen: []
buf12 = torch.ops.aten.set_.source_Tensor(primals_2, buf5)
assert_size_stride(buf12, (4, ), (1, ))
del buf0
# Topologically Sorted Source Nodes: [truediv], Original ATen: [aten.div]
buf18 = torch.ops.aten.set_.source_Tensor(primals_3, buf17)
assert_size_stride(buf18, (64, ), (1, ))
del primals_3
return (buf10, buf7, primals_1, primals_2, primals_4, buf2, buf5, buf6, buf7, buf9, reinterpret_tensor(primals_6, (256, 4, 1, 1), (8, 1, 1, 1), 0), buf11, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((256, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((32, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, x, style):
x = self.pad(x)
x = self.conv2d(x)
x = self.norm(x, style)
if self.activation:
x = self.activation(x)
return x
def get_inputs():
return [torch.rand([256, 4, 4, 4]), torch.rand([32, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_0(in_out_ptr0, in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp25, None)
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp26, None)
@triton.jit
def triton_per_fused_div_mv_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 4
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 64 * x0), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_2(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp8, None)
@triton.jit
def triton_per_fused_dot_3(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
@triton.jit
def triton_poi_fused_div_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5(
in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK: tl.
constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
x1 = xindex // 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (x0 + 8 * x1), xmask)
tmp14 = tl.load(in_ptr1 + (4 + x0 + 8 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 / tmp3
tmp5 = tmp2 - tmp4
tmp6 = tmp5 * tmp5
tmp7 = tmp6 / tmp3
tmp8 = 1e-08
tmp9 = tmp7 + tmp8
tmp10 = libdevice.rsqrt(tmp9)
tmp11 = tmp5 * tmp10
tmp13 = tmp11 * tmp12
tmp15 = tmp13 + tmp14
tmp16 = 0.0
tmp17 = tmp15 > tmp16
tmp18 = 0.2
tmp19 = tmp15 * tmp18
tmp20 = tl.where(tmp17, tmp15, tmp19)
tmp21 = tmp20 > tmp16
tl.store(in_out_ptr0 + x2, tmp2, xmask)
tl.store(out_ptr0 + x2, tmp20, xmask)
tl.store(out_ptr1 + x2, tmp21, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (64,), (1,))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64,), (1,), torch.float32)
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1
del buf1
buf17 = empty_strided_cuda((64,), (1,), torch.float32)
get_raw_stream(0)
triton_per_fused_add_div_linalg_vector_norm_mv_0[grid(1)](buf2,
primals_4, primals_2, buf0, buf17, 1, 64, XBLOCK=1, num_warps=2,
num_stages=1)
buf3 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_div_mv_1[grid(4)](primals_4, buf0, buf2, buf3, 4,
64, XBLOCK=1, num_warps=2, num_stages=1)
buf5 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_2[grid(1)](buf3, buf5,
1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf6 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_dot_3[grid(1)](buf5, buf3, buf6, 1, 4, XBLOCK=1,
num_warps=2, num_stages=1)
del buf3
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_div_4[grid(256)](primals_4, buf6, buf7, 256,
XBLOCK=128, num_warps=4, num_stages=1)
buf8 = extern_kernels.convolution(primals_1, buf7, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (256, 4, 1, 1), (4, 1, 1, 1))
buf9 = buf8
del buf8
buf10 = empty_strided_cuda((256, 4, 1, 1), (4, 1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((256, 4, 1, 1), (4, 1, 1, 1), torch.bool)
triton_poi_fused_add_convolution_leaky_relu_leaky_relu_backward_mul_rsqrt_sub_5[
grid(1024)](buf9, primals_5, primals_6, buf10, buf11, 1024,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf12 = torch.ops.aten.set_.source_Tensor(primals_2, buf5)
assert_size_stride(buf12, (4,), (1,))
del buf0
buf18 = torch.ops.aten.set_.source_Tensor(primals_3, buf17)
assert_size_stride(buf18, (64,), (1,))
del primals_3
return (buf10, buf7, primals_1, primals_2, primals_4, buf2, buf5, buf6,
buf7, buf9, reinterpret_tensor(primals_6, (256, 4, 1, 1), (8, 1, 1,
1), 0), buf11)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayerNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayerNew, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, input_0, input_1):
primals_2 = self.conv2d.module.bias
primals_5 = self.conv2d.module.weight_u
primals_3 = self.conv2d.module.weight_v
primals_4 = self.conv2d.module.weight_bar
primals_1 = input_0
primals_6 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| LenKerr/Semantic-Colorization-GAN | AdaINConv2dLayer | false | 5,521 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super().__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class Model(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigm
# ... truncated (>4000 chars) for memory efficiency |
TokenEmbedding | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/wi/cwibqvrnbfx7xhnfzzckhfwxbmmaeepyx4l2irzdxw23feqjr3lp.py
# Topologically Sorted Source Nodes: [long], Original ATen: [aten._to_copy]
# Source node to ATen node mapping:
# long => convert_element_type
# Graph fragment:
# %convert_element_type : [num_users=2] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%primals_1, torch.int64), kwargs = {})
triton_poi_fused__to_copy_0 = async_compile.triton('triton_poi_fused__to_copy_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__to_copy_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tmp0.to(tl.int64)
tl.store(out_ptr0 + (x0), tmp1, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ia/ciafxdq32uqzkpbws275y3bp3mee3juggqv7sqnd2mxb3zrxr2oq.py
# Topologically Sorted Source Nodes: [embedding, mul], Original ATen: [aten.embedding, aten.mul]
# Source node to ATen node mapping:
# embedding => embedding
# mul => mul
# Graph fragment:
# %embedding : [num_users=1] = call_function[target=torch.ops.aten.embedding.default](args = (%primals_2, %convert_element_type), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%embedding, 2.0), kwargs = {})
triton_poi_fused_embedding_mul_1 = async_compile.triton('triton_poi_fused_embedding_mul_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*i64', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_embedding_mul_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_embedding_mul_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x0 = xindex % 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 4, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tl.device_assert(((0 <= tmp4) & (tmp4 < 4)) | ~(xmask), "index out of bounds: 0 <= tmp4 < 4")
tmp6 = tl.load(in_ptr1 + (x0 + (4*tmp4)), xmask)
tmp7 = 2.0
tmp8 = tmp6 * tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.int64)
# Topologically Sorted Source Nodes: [long], Original ATen: [aten._to_copy]
stream0 = get_raw_stream(0)
triton_poi_fused__to_copy_0.run(primals_1, buf0, 256, grid=grid(256), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4, 4), (256, 64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [embedding, mul], Original ATen: [aten.embedding, aten.mul]
triton_poi_fused_embedding_mul_1.run(buf0, primals_2, buf1, 1024, grid=grid(1024), stream=stream0)
del primals_2
return (buf1, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
from torch import Tensor
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
class TokenEmbedding(nn.Module):
def __init__(self, vocab_size: 'int', emb_size):
super(TokenEmbedding, self).__init__()
self.embedding = nn.Embedding(vocab_size, emb_size)
self.emb_size = emb_size
def forward(self, tokens: 'Tensor'):
return self.embedding(tokens.long()) * math.sqrt(self.emb_size)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'vocab_size': 4, 'emb_size': 4}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__to_copy_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tmp0.to(tl.int64)
tl.store(out_ptr0 + x0, tmp1, xmask)
@triton.jit
def triton_poi_fused_embedding_mul_1(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 4, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tl.device_assert((0 <= tmp4) & (tmp4 < 4) | ~xmask,
'index out of bounds: 0 <= tmp4 < 4')
tmp6 = tl.load(in_ptr1 + (x0 + 4 * tmp4), xmask)
tmp7 = 2.0
tmp8 = tmp6 * tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.int64)
get_raw_stream(0)
triton_poi_fused__to_copy_0[grid(256)](primals_1, buf0, 256, XBLOCK
=128, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4, 4), (256, 64, 16, 4, 1),
torch.float32)
triton_poi_fused_embedding_mul_1[grid(1024)](buf0, primals_2, buf1,
1024, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
return buf1, buf0
class TokenEmbeddingNew(nn.Module):
def __init__(self, vocab_size: 'int', emb_size):
super(TokenEmbeddingNew, self).__init__()
self.embedding = nn.Embedding(vocab_size, emb_size)
self.emb_size = emb_size
def forward(self, input_0):
primals_2 = self.embedding.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| Lezcano/tutorials | TokenEmbedding | false | 5,522 | [
"BSD-3-Clause"
] | 1 | 24946b2e6d3d825afed6b35c1c4d618a70a88be8 | https://github.com/Lezcano/tutorials/tree/24946b2e6d3d825afed6b35c1c4d618a70a88be8 | import math
import torch
from torch import Tensor
import torch.nn as nn
import torch.quantization
import torch.onnx
import torch.nn.parallel
import torch.utils.data
import torch.fx
import torch.nn
import torch.optim
import torch.profiler
class Model(nn.Module):
def __init__(self, vocab_size: 'int', emb_size):
super().__init__()
self.embedding = nn.Embedding(vocab_size, emb_size)
self.emb_size = emb_size
def forward(self, tokens: 'Tensor'):
return self.embedding(tokens.long()) * math.sqrt(self.emb_size)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
LayerNorm | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/tc/ctcoccnojrifwsjhb4gqgfu5kxpt6dvdpv4qwca7cbgn27ktptbk.py
# Topologically Sorted Source Nodes: [mean, std, sub, add, ln_out, mul, ln_out_1], Original ATen: [aten.mean, aten.std, aten.sub, aten.add, aten.div, aten.mul]
# Source node to ATen node mapping:
# add => add
# ln_out => div
# ln_out_1 => add_1
# mean => mean
# mul => mul
# std => sqrt, var
# sub => sub
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%primals_1, [-1], True), kwargs = {})
# %var : [num_users=1] = call_function[target=torch.ops.aten.var.correction](args = (%primals_1, [-1]), kwargs = {correction: 1.0, keepdim: True})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%var,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %mean), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sqrt, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, %add), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_2, %div), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_3), kwargs = {})
triton_poi_fused_add_div_mean_mul_std_sub_0 = async_compile.triton('triton_poi_fused_add_div_mean_mul_std_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_mean_mul_std_sub_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 7, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tl.load(in_ptr1 + (4*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp8 = tmp6 + tmp7
tmp9 = 4.0
tmp10 = tmp8 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp2 - tmp10
tmp13 = tmp12 * tmp12
tmp14 = tmp3 - tmp10
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp10
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp7 - tmp10
tmp21 = tmp20 * tmp20
tmp22 = tmp19 + tmp21
tmp23 = 3.0
tmp24 = tmp22 / tmp23
tmp25 = libdevice.sqrt(tmp24)
tmp26 = 1e-06
tmp27 = tmp25 + tmp26
tmp28 = tmp11 / tmp27
tmp29 = tmp0 * tmp28
tmp31 = tmp29 + tmp30
tl.store(out_ptr0 + (x2), tmp31, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mean, std, sub, add, ln_out, mul, ln_out_1], Original ATen: [aten.mean, aten.std, aten.sub, aten.add, aten.div, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_mean_mul_std_sub_0.run(primals_2, primals_1, primals_3, buf0, 256, grid=grid(256), stream=stream0)
del primals_2
del primals_3
return (buf0, primals_1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class LayerNorm(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, x):
mean = x.mean(dim=-1, keepdim=True)
std = x.std(dim=-1, keepdim=True)
ln_out = (x - mean) / (std + self.eps)
ln_out = self.gamma * ln_out + self.beta
return ln_out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_hid': 4}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_0(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp8 = tmp6 + tmp7
tmp9 = 4.0
tmp10 = tmp8 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp2 - tmp10
tmp13 = tmp12 * tmp12
tmp14 = tmp3 - tmp10
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp10
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp7 - tmp10
tmp21 = tmp20 * tmp20
tmp22 = tmp19 + tmp21
tmp23 = 3.0
tmp24 = tmp22 / tmp23
tmp25 = libdevice.sqrt(tmp24)
tmp26 = 1e-06
tmp27 = tmp25 + tmp26
tmp28 = tmp11 / tmp27
tmp29 = tmp0 * tmp28
tmp31 = tmp29 + tmp30
tl.store(out_ptr0 + x2, tmp31, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_mean_mul_std_sub_0[grid(256)](primals_2,
primals_1, primals_3, buf0, 256, XBLOCK=256, num_warps=4,
num_stages=1)
del primals_2
del primals_3
return buf0, primals_1
class LayerNormNew(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super(LayerNormNew, self).__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, input_0):
primals_2 = self.gamma
primals_3 = self.beta
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| Lhx94As/PHO-LID | LayerNorm | false | 5,523 | [
"MIT"
] | 1 | 44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | https://github.com/Lhx94As/PHO-LID/tree/44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super().__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, x):
mean = x.mean(dim=-1, keepdim=True)
std = x.std(dim=-1, keepdim=True)
ln_out = (x - mean) / (std + self.eps)
ln_out = self.gamma * ln_out + self.beta
return ln_out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
ResAdaINConv2dLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/o5/co5utrqsbdulrbygkmgbtnft3s33dxmhlcrl7llsvtgrwilyozcn.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x => constant_pad_nd
# Graph fragment:
# %constant_pad_nd : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%primals_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 36864
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/pj/cpj6t7ch5tbjax2t7isz2rrooc27ef6tcwdhv2q3coqhpluuaotk.py
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add
# mv => mul, sum_1
# norm => pow_1, pow_2, sum_2
# truediv => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute, %primals_2), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, None), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 1e-12), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_1 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {5: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=(5,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), rmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (36 + r0), rmask, other=0.0)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (72 + r0), rmask, other=0.0)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (108 + r0), rmask, other=0.0)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = libdevice.sqrt(tmp23)
tmp25 = 1e-12
tmp26 = tmp24 + tmp25
tmp27 = tmp18 / tmp26
tl.store(out_ptr0 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp18, rmask)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp26, None)
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp27, rmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/og/cogzn2aqt2zqefp3tgxavposfk6wg5gaj3zipeiyyawkpex6e6wf.py
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
# Source node to ATen node mapping:
# mv_1 => mul_1, sum_3
# truediv => div
# Graph fragment:
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %div), kwargs = {})
# %sum_3 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [1]), kwargs = {})
triton_per_fused_div_mv_2 = async_compile.triton('triton_per_fused_div_mv_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_mv_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (36*x0)), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1), rmask, eviction_policy='evict_last', other=0.0)
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(rmask & xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qa/cqaed4ios3xqwlv4d3cciikkdz7d73vhwkegurd5cxca3y7htmvg.py
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_1 => add_1
# norm_1 => pow_3, pow_4, sum_4
# truediv_1 => div_1
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, None), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_4, 0.5), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_4, 1e-12), kwargs = {})
# %div_1 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, %add_1), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_3 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp8, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/c2/cc2arficwjs4sforhl25gdfmb3uzfg7hkw46gq3mxgv57jy52z32.py
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
# Source node to ATen node mapping:
# sigma => mul_3, sum_6
# Graph fragment:
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %sum_3), kwargs = {})
# %sum_6 : [num_users=2] = call_function[target=torch.ops.aten.sum.default](args = (%mul_3,), kwargs = {})
triton_per_fused_dot_4 = async_compile.triton('triton_per_fused_dot_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_dot_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/sb/csbqzjhtj4wya4qmyds7hrs7i55ypvfew7naorg3uvb7egvaiqs2.py
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_2 => div_2
# Graph fragment:
# %div_2 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_4, %expand), kwargs = {})
triton_poi_fused_div_5 = async_compile.triton('triton_poi_fused_div_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/uz/cuzeit4gmkb3dfi574sa6fq5xo72c2sqtw2e25dzqcvoo6tjrh6l.py
# Topologically Sorted Source Nodes: [x_1, mean, x_2, mean_2, add_2, var_2, x_3, mul_2, content, x_4, out], Original ATen: [aten.convolution, aten.mean, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu]
# Source node to ATen node mapping:
# add_2 => add_2
# content => add_3
# mean => mean
# mean_2 => mean_1
# mul_2 => mul_6
# out => add_4
# var_2 => rsqrt
# x_1 => convolution
# x_2 => sub
# x_3 => mul_5
# x_4 => gt, mul_7, where
# Graph fragment:
# %convolution : [num_users=3] = call_function[target=torch.ops.aten.convolution.default](args = (%constant_pad_nd, %div_2, %primals_5, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view_3, [2]), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%convolution, %view_4), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view_5, [2]), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_6, 1e-08), kwargs = {})
# %rsqrt : [num_users=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_5, %select), kwargs = {})
# %add_3 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_6, %select_1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%add_3, 0), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_3, 0.2), kwargs = {})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%gt, %add_3, %mul_7), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%where, %primals_1), kwargs = {})
triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6 = async_compile.triton('triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1024, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32', 8: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7, 8), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1', 'in_out_ptr2', 'in_out_ptr3'], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6(in_out_ptr0, in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1, in_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1024
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 4
x1 = (xindex // 4)
tmp0 = tl.load(in_out_ptr0 + (r2 + (16*x3)), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (x0 + (8*x1)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (4 + x0 + (8*x1)), xmask, eviction_policy='evict_last')
tmp29 = tl.load(in_ptr2 + (r2 + (16*x3)), xmask, other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.where(xmask, tmp3, 0)
tmp6 = tl.sum(tmp5, 1)[:, None]
tmp7 = 16.0
tmp8 = tmp6 / tmp7
tmp9 = tmp2 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.where(xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tmp14 / tmp7
tmp16 = 1e-08
tmp17 = tmp15 + tmp16
tmp18 = libdevice.rsqrt(tmp17)
tmp19 = tmp9 * tmp18
tmp21 = tmp19 * tmp20
tmp23 = tmp21 + tmp22
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp30 = tmp28 + tmp29
tl.store(in_out_ptr0 + (r2 + (16*x3)), tmp2, xmask)
tl.debug_barrier()
tl.store(in_out_ptr1 + (x3), tmp8, xmask)
tl.debug_barrier()
tl.store(in_out_ptr2 + (x3), tmp18, xmask)
tl.store(in_out_ptr3 + (r2 + (16*x3)), tmp30, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (36, ), (1, ))
assert_size_stride(primals_4, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((256, 4, 6, 6), (144, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(primals_1, buf0, 36864, grid=grid(36864), stream=stream0)
buf1 = empty_strided_cuda((36, ), (1, ), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2; del buf2 # reuse
buf22 = empty_strided_cuda((36, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_mv_1.run(buf3, primals_4, primals_2, buf1, buf22, 1, 36, grid=grid(1), stream=stream0)
buf4 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
triton_per_fused_div_mv_2.run(primals_4, buf1, buf3, buf4, 4, 36, grid=grid(4), stream=stream0)
buf6 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_3.run(buf4, buf6, 1, 4, grid=grid(1), stream=stream0)
buf7 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
triton_per_fused_dot_4.run(buf6, buf4, buf7, 1, 4, grid=grid(1), stream=stream0)
del buf4
buf8 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
triton_poi_fused_div_5.run(primals_4, buf7, buf8, 144, grid=grid(144), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (256, 4, 4, 4), (64, 16, 4, 1))
buf10 = buf9; del buf9 # reuse
buf11 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf12 = buf11; del buf11 # reuse
buf13 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf14 = reinterpret_tensor(buf13, (256, 4, 1, 1), (4, 1, 1, 1), 0); del buf13 # reuse
buf15 = empty_strided_cuda((256, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf16 = buf15; del buf15 # reuse
# Topologically Sorted Source Nodes: [x_1, mean, x_2, mean_2, add_2, var_2, x_3, mul_2, content, x_4, out], Original ATen: [aten.convolution, aten.mean, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu]
triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6.run(buf10, buf12, buf14, buf16, primals_5, primals_6, primals_1, 1024, 16, grid=grid(1024), stream=stream0)
del primals_1
del primals_5
# Topologically Sorted Source Nodes: [], Original ATen: []
buf17 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf17, (4, ), (1, ))
del buf1
# Topologically Sorted Source Nodes: [truediv], Original ATen: [aten.div]
buf23 = torch.ops.aten.set_.source_Tensor(primals_3, buf22)
assert_size_stride(buf23, (36, ), (1, ))
del primals_3
return (buf16, buf8, primals_2, primals_4, primals_6, buf0, buf3, buf6, buf7, buf8, buf10, reinterpret_tensor(buf12, (256, 4, 1, 1), (4, 1, 1, 1), 0), buf14, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((256, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((36, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((32, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, x, style):
x = self.pad(x)
x = self.conv2d(x)
x = self.norm(x, style)
if self.activation:
x = self.activation(x)
return x
class ResAdaINConv2dLayer(nn.Module):
def __init__(self, channels, kernel_size=3, stride=1, padding=1,
dilation=1, pad_type='zero', activation='lrelu', norm='none', sn=True):
super(ResAdaINConv2dLayer, self).__init__()
self.conv2d = AdaINConv2dLayer(channels, channels, kernel_size,
stride, padding, dilation, pad_type, activation, sn)
def forward(self, x, style):
residual = x
out = self.conv2d(x, style)
out = out + residual
return out
def get_inputs():
return [torch.rand([256, 4, 4, 4]), torch.rand([32, 4, 4, 4])]
def get_init_inputs():
return [[], {'channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10, other=0.0)
tl.store(out_ptr0 + x4, tmp11, None)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, rmask, other=0.0)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (36 + r0), rmask, other=0.0)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (72 + r0), rmask, other=0.0)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (108 + r0), rmask, other=0.0)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = libdevice.sqrt(tmp23)
tmp25 = 1e-12
tmp26 = tmp24 + tmp25
tmp27 = tmp18 / tmp26
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp18, rmask)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp26, None)
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp27, rmask)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 36 * x0), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + r1, rmask, eviction_policy='evict_last', other=0.0
)
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(rmask & xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp8, None)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6(
in_out_ptr0, in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1,
in_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 1024
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 4
x1 = xindex // 4
tmp0 = tl.load(in_out_ptr0 + (r2 + 16 * x3), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + (4 + x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp29 = tl.load(in_ptr2 + (r2 + 16 * x3), xmask, other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.where(xmask, tmp3, 0)
tmp6 = tl.sum(tmp5, 1)[:, None]
tmp7 = 16.0
tmp8 = tmp6 / tmp7
tmp9 = tmp2 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.where(xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tmp14 / tmp7
tmp16 = 1e-08
tmp17 = tmp15 + tmp16
tmp18 = libdevice.rsqrt(tmp17)
tmp19 = tmp9 * tmp18
tmp21 = tmp19 * tmp20
tmp23 = tmp21 + tmp22
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp30 = tmp28 + tmp29
tl.store(in_out_ptr0 + (r2 + 16 * x3), tmp2, xmask)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp8, xmask)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp18, xmask)
tl.store(in_out_ptr3 + (r2 + 16 * x3), tmp30, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (36,), (1,))
assert_size_stride(primals_4, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((256, 4, 6, 6), (144, 36, 6, 1), torch.
float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(36864)](primals_1, buf0,
36864, XBLOCK=256, num_warps=4, num_stages=1)
buf1 = empty_strided_cuda((36,), (1,), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
buf22 = empty_strided_cuda((36,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_mv_1[grid(1)](buf3,
primals_4, primals_2, buf1, buf22, 1, 36, XBLOCK=1, num_warps=2,
num_stages=1)
buf4 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_div_mv_2[grid(4)](primals_4, buf1, buf3, buf4, 4,
36, XBLOCK=1, num_warps=2, num_stages=1)
buf6 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_3[grid(1)](buf4, buf6,
1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf7 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_dot_4[grid(1)](buf6, buf4, buf7, 1, 4, XBLOCK=1,
num_warps=2, num_stages=1)
del buf4
buf8 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
triton_poi_fused_div_5[grid(144)](primals_4, buf7, buf8, 144,
XBLOCK=256, num_warps=4, num_stages=1)
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (256, 4, 4, 4), (64, 16, 4, 1))
buf10 = buf9
del buf9
buf11 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf12 = buf11
del buf11
buf13 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf14 = reinterpret_tensor(buf13, (256, 4, 1, 1), (4, 1, 1, 1), 0)
del buf13
buf15 = empty_strided_cuda((256, 4, 4, 4), (64, 16, 4, 1), torch.
float32)
buf16 = buf15
del buf15
triton_per_fused_add_convolution_leaky_relu_mean_mul_rsqrt_sub_6[grid
(1024)](buf10, buf12, buf14, buf16, primals_5, primals_6,
primals_1, 1024, 16, XBLOCK=32, num_warps=4, num_stages=1)
del primals_1
del primals_5
buf17 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf17, (4,), (1,))
del buf1
buf23 = torch.ops.aten.set_.source_Tensor(primals_3, buf22)
assert_size_stride(buf23, (36,), (1,))
del primals_3
return (buf16, buf8, primals_2, primals_4, primals_6, buf0, buf3, buf6,
buf7, buf8, buf10, reinterpret_tensor(buf12, (256, 4, 1, 1), (4, 1,
1, 1), 0), buf14)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, x, style):
x = self.pad(x)
x = self.conv2d(x)
x = self.norm(x, style)
if self.activation:
x = self.activation(x)
return x
class ResAdaINConv2dLayerNew(nn.Module):
def __init__(self, channels, kernel_size=3, stride=1, padding=1,
dilation=1, pad_type='zero', activation='lrelu', norm='none', sn=True):
super(ResAdaINConv2dLayerNew, self).__init__()
self.conv2d = AdaINConv2dLayer(channels, channels, kernel_size,
stride, padding, dilation, pad_type, activation, sn)
def forward(self, input_0, input_1):
primals_2 = self.conv2d.conv2d.module.bias
primals_5 = self.conv2d.conv2d.module.weight_u
primals_3 = self.conv2d.conv2d.module.weight_v
primals_4 = self.conv2d.conv2d.module.weight_bar
primals_1 = input_0
primals_6 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| LenKerr/Semantic-Colorization-GAN | ResAdaINConv2dLayer | false | 5,524 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super().__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activatio
# ... truncated (>4000 chars) for memory efficiency |
PositionwiseFeedForward | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/r3/cr3febcwm3t44fuoitsx3ou2p6xg4sk4f7unagmmrvffasxf47te.py
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# relu => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ji/cji7mw45fbdoanjc5e6qu3e2bf5d6jnnjabskl6onjlk7uv7oqud.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# x_2 => add
# x_3 => var_mean
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %primals_1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add, [3]), kwargs = {correction: 0, keepdim: True})
triton_poi_fused_add_native_layer_norm_1 = async_compile.triton('triton_poi_fused_add_native_layer_norm_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_native_layer_norm_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_native_layer_norm_1(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = tmp27 / tmp15
tl.store(out_ptr0 + (x0), tmp16, xmask)
tl.store(out_ptr1 + (x0), tmp28, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/xp/cxpe6vtgpqizlfgu2p5hfehvzox7zp73wjlhkybotxh42njvnais.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# x_2 => add
# x_3 => add_1, add_2, mul, mul_1, rsqrt, sub
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %primals_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-06), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_1,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %getitem_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %primals_6), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_7), kwargs = {})
triton_poi_fused_add_native_layer_norm_2 = async_compile.triton('triton_poi_fused_add_native_layer_norm_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_native_layer_norm_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_native_layer_norm_2(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp3 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x1), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp6 = 1e-06
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp4 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + (x2), tmp13, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_3, buf6, 256, grid=grid(256), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_1.run(buf2, primals_1, buf3, buf4, 64, grid=grid(64), stream=stream0)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_2.run(buf2, primals_1, buf3, buf4, primals_6, primals_7, buf5, 256, grid=grid(256), stream=stream0)
del buf3
del buf4
del primals_7
return (buf5, primals_1, primals_6, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), buf2, primals_4, buf6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.nn.init
class PositionwiseFeedForward(nn.Module):
""" A two-feed-forward-layer module """
def __init__(self, d_in, d_hid, dropout=0.1):
super().__init__()
self.w_1 = nn.Linear(d_in, d_hid)
self.w_2 = nn.Linear(d_hid, d_in)
self.layer_norm = nn.LayerNorm(d_in, eps=1e-06)
self.dropout = nn.Dropout(dropout)
def forward(self, x):
residual = x
x = self.w_2(F.relu(self.w_1(x)))
x = self.dropout(x)
x += residual
del residual
x = self.layer_norm(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_in': 4, 'd_hid': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_1(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = tmp27 / tmp15
tl.store(out_ptr0 + x0, tmp16, xmask)
tl.store(out_ptr1 + x0, tmp28, xmask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_2(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp6 = 1e-06
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp4 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + x2, tmp13, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_3, buf6, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
triton_poi_fused_add_native_layer_norm_1[grid(64)](buf2, primals_1,
buf3, buf4, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_native_layer_norm_2[grid(256)](buf2, primals_1,
buf3, buf4, primals_6, primals_7, buf5, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf3
del buf4
del primals_7
return buf5, primals_1, primals_6, reinterpret_tensor(buf1, (64, 4), (4,
1), 0), buf2, primals_4, buf6
class PositionwiseFeedForwardNew(nn.Module):
""" A two-feed-forward-layer module """
def __init__(self, d_in, d_hid, dropout=0.1):
super().__init__()
self.w_1 = nn.Linear(d_in, d_hid)
self.w_2 = nn.Linear(d_hid, d_in)
self.layer_norm = nn.LayerNorm(d_in, eps=1e-06)
self.dropout = nn.Dropout(dropout)
def forward(self, input_0):
primals_2 = self.w_1.weight
primals_3 = self.w_1.bias
primals_4 = self.w_2.weight
primals_5 = self.w_2.bias
primals_6 = self.layer_norm.weight
primals_7 = self.layer_norm.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| LiJiaBei-7/rivrl | PositionwiseFeedForward | false | 5,525 | [
"Apache-2.0"
] | 1 | f6078e4826c788704bb338d7a695ef879ff969f4 | https://github.com/LiJiaBei-7/rivrl/tree/f6078e4826c788704bb338d7a695ef879ff969f4 | import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.nn.init
class Model(nn.Module):
""" A two-feed-forward-layer module """
def __init__(self, d_in, d_hid, dropout=0.1):
super().__init__()
self.w_1 = nn.Linear(d_in, d_hid)
self.w_2 = nn.Linear(d_hid, d_in)
self.layer_norm = nn.LayerNorm(d_in, eps=1e-06)
self.dropout = nn.Dropout(dropout)
def forward(self, x):
residual = x
x = self.w_2(F.relu(self.w_1(x)))
x = self.dropout(x)
x += residual
del residual
x = self.layer_norm(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
DenseConv2dLayer_5C | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/td/ctdv3m5a33kovvtng5iilth4k6mtnyfcota6hhwoiqm34iumu7wi.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x => constant_pad_nd
# Graph fragment:
# %constant_pad_nd : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%primals_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/er/cerxfmiafmpnsn44ihnsss2pvd76g6x2i62qavnas5ygqo5mj43d.py
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat => cat
# x_3 => constant_pad_nd_1
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where], 1), kwargs = {})
# %constant_pad_nd_1 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_1 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 8
x3 = (xindex // 288)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + (x6), tmp30, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s3/cs34ah7ioinuak53u3oyhgsrsqeqfnovsb4djwnzfgkw44v6u547.py
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat_1 => cat_1
# x_6 => constant_pad_nd_2
# Graph fragment:
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1], 1), kwargs = {})
# %constant_pad_nd_2 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_2 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 12
x3 = (xindex // 432)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tmp30 = tl.full([1], 12, tl.int64)
tmp31 = tmp11 < tmp30
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + ((-5) + x0 + (4*x1) + (16*((-8) + x2)) + (64*x3)), tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + (x6), tmp42, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/2h/c2hdtntquyhqkvcjcbctuzny5hhijks5jcqplynkbwfrthtih25e.py
# Topologically Sorted Source Nodes: [cat_2], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat_2 => cat_2
# Graph fragment:
# %cat_2 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1, %where_2], 1), kwargs = {})
triton_poi_fused_cat_3 = async_compile.triton('triton_poi_fused_cat_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 16) % 16
x0 = xindex % 16
x2 = (xindex // 256)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + (16*((-4) + x1)) + (64*x2)), tmp9 & xmask, other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + (16*((-8) + x1)) + (64*x2)), tmp21 & xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tl.load(in_ptr3 + (x0 + (16*((-12) + x1)) + (64*x2)), tmp28 & xmask, other=0.0)
tmp32 = tmp31 > tmp11
tmp33 = tmp31 * tmp13
tmp34 = tl.where(tmp32, tmp31, tmp33)
tmp35 = tl.full(tmp34.shape, 0.0, tmp34.dtype)
tmp36 = tl.where(tmp28, tmp34, tmp35)
tmp37 = tl.where(tmp21, tmp27, tmp36)
tmp38 = tl.where(tmp9, tmp17, tmp37)
tmp39 = tl.where(tmp4, tmp5, tmp38)
tl.store(out_ptr0 + (x3), tmp39, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rc/crcdywpxyr2fxsveen5yolsghlkstdms4uvwp7zkfwhcmkjmpxpg.py
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x_9 => constant_pad_nd_3
# Graph fragment:
# %constant_pad_nd_3 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_2, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_4 = async_compile.triton('triton_poi_fused_constant_pad_nd_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_4(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4p/c4pwo5hnpxyletkwji2bfgdsgviax75psfwuih5qzyaah2skdcdn.py
# Topologically Sorted Source Nodes: [cat_3], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat_3 => cat_3
# Graph fragment:
# %cat_3 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1, %where_2, %where_3], 1), kwargs = {})
triton_poi_fused_cat_5 = async_compile.triton('triton_poi_fused_cat_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1280
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 16) % 20
x0 = xindex % 16
x2 = (xindex // 320)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + (16*((-4) + x1)) + (64*x2)), tmp9 & xmask, other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + (16*((-8) + x1)) + (64*x2)), tmp21 & xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tmp28 & tmp30
tmp32 = tl.load(in_ptr3 + (x0 + (16*((-12) + x1)) + (64*x2)), tmp31 & xmask, other=0.0)
tmp33 = tmp32 > tmp11
tmp34 = tmp32 * tmp13
tmp35 = tl.where(tmp33, tmp32, tmp34)
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp31, tmp35, tmp36)
tmp38 = tmp0 >= tmp29
tmp39 = tl.full([1], 20, tl.int64)
tmp40 = tmp0 < tmp39
tmp41 = tl.load(in_ptr4 + (x0 + (16*((-16) + x1)) + (64*x2)), tmp38 & xmask, other=0.0)
tmp42 = tmp41 > tmp11
tmp43 = tmp41 * tmp13
tmp44 = tl.where(tmp42, tmp41, tmp43)
tmp45 = tl.full(tmp44.shape, 0.0, tmp44.dtype)
tmp46 = tl.where(tmp38, tmp44, tmp45)
tmp47 = tl.where(tmp31, tmp37, tmp46)
tmp48 = tl.where(tmp21, tmp27, tmp47)
tmp49 = tl.where(tmp9, tmp17, tmp48)
tmp50 = tl.where(tmp4, tmp5, tmp49)
tl.store(out_ptr0 + (x3), tmp50, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s7/cs7remm7dxallug7mpmvc7novi2rckl5n32tcwbq5syx4cb4lqk6.py
# Topologically Sorted Source Nodes: [x_12], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x_12 => constant_pad_nd_4
# Graph fragment:
# %constant_pad_nd_4 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_3, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_6 = async_compile.triton('triton_poi_fused_constant_pad_nd_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_6(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2880
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/fz/cfzioooyib2f5ephr2m4wpireief66w5kf7iwt5ay6lq5g5n4k5g.py
# Topologically Sorted Source Nodes: [x_14], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x_14 => gt_4, mul_4, where_4
# Graph fragment:
# %gt_4 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_4, 0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_4, 0.2), kwargs = {})
# %where_4 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_4, %convolution_4, %mul_4), kwargs = {})
# %gt_5 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_4, 0), kwargs = {})
triton_poi_fused_leaky_relu_leaky_relu_backward_7 = async_compile.triton('triton_poi_fused_leaky_relu_leaky_relu_backward_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_leaky_relu_backward_7', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_7(in_out_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(in_out_ptr0 + (x0), tmp5, xmask)
tl.store(out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/is/cisodw5gxdmlgapnrqpu7bcsl5srnastcmg2c73tt3lqfqu2bmk6.py
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x_11 => gt_3, mul_3, where_3
# Graph fragment:
# %gt_3 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_3, 0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_3, 0.2), kwargs = {})
# %where_3 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_3, %convolution_3, %mul_3), kwargs = {})
# %gt_6 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_3, 0), kwargs = {})
triton_poi_fused_leaky_relu_leaky_relu_backward_8 = async_compile.triton('triton_poi_fused_leaky_relu_leaky_relu_backward_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_leaky_relu_backward_8', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_8(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
assert_size_stride(primals_5, (4, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (4, 20, 3, 3), (180, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(primals_1, buf0, 576, grid=grid(576), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_1.run(primals_1, buf1, buf2, 1152, grid=grid(1152), stream=stream0)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.convolution]
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_2.run(primals_1, buf1, buf3, buf4, 1728, grid=grid(1728), stream=stream0)
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.convolution]
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 16, 4, 4), (256, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_2], Original ATen: [aten.cat]
triton_poi_fused_cat_3.run(primals_1, buf1, buf3, buf5, buf6, 1024, grid=grid(1024), stream=stream0)
buf7 = empty_strided_cuda((4, 16, 6, 6), (576, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.constant_pad_nd]
triton_poi_fused_constant_pad_nd_4.run(buf6, buf7, 2304, grid=grid(2304), stream=stream0)
del buf6
# Topologically Sorted Source Nodes: [x_10], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_5, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4, 4), (64, 16, 4, 1))
buf9 = empty_strided_cuda((4, 20, 4, 4), (320, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_3], Original ATen: [aten.cat]
triton_poi_fused_cat_5.run(primals_1, buf1, buf3, buf5, buf8, buf9, 1280, grid=grid(1280), stream=stream0)
del primals_1
buf10 = empty_strided_cuda((4, 20, 6, 6), (720, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_12], Original ATen: [aten.constant_pad_nd]
triton_poi_fused_constant_pad_nd_6.run(buf9, buf10, 2880, grid=grid(2880), stream=stream0)
del buf9
# Topologically Sorted Source Nodes: [x_13], Original ATen: [aten.convolution]
buf11 = extern_kernels.convolution(buf10, primals_6, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 4, 4, 4), (64, 16, 4, 1))
buf12 = buf11; del buf11 # reuse
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_14], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_7.run(buf12, buf13, 256, grid=grid(256), stream=stream0)
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf8, buf14, 256, grid=grid(256), stream=stream0)
del buf8
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf5, buf15, 256, grid=grid(256), stream=stream0)
del buf5
buf16 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf3, buf16, 256, grid=grid(256), stream=stream0)
del buf3
buf17 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf1, buf17, 256, grid=grid(256), stream=stream0)
del buf1
return (buf12, primals_2, primals_3, primals_4, primals_5, primals_6, buf0, buf2, buf4, buf7, buf10, buf13, buf14, buf15, buf16, buf17, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 12, 3, 3), (108, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 20, 3, 3), (180, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class DenseConv2dLayer_5C(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(DenseConv2dLayer_5C, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv4 = Conv2dLayer(in_channels + latent_channels * 3,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv5 = Conv2dLayer(in_channels + latent_channels * 4,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, x):
x1 = self.conv1(x)
x2 = self.conv2(torch.cat((x, x1), 1))
x3 = self.conv3(torch.cat((x, x1, x2), 1))
x4 = self.conv4(torch.cat((x, x1, x2, x3), 1))
x5 = self.conv5(torch.cat((x, x1, x2, x3, x4), 1))
return x5
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'latent_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 8
x3 = xindex // 288
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tl.full([1], 8, tl.int64)
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + x6, tmp30, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 12
x3 = xindex // 432
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tl.full([1], 12, tl.int64)
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + (-5 + x0 + 4 * x1 + 16 * (-8 + x2) + 64 * x3),
tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + x6, tmp42, xmask)
@triton.jit
def triton_poi_fused_cat_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 16
x0 = xindex % 16
x2 = xindex // 256
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp9 & xmask,
other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + 16 * (-8 + x1) + 64 * x2), tmp21 &
xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tl.full([1], 16, tl.int64)
tmp31 = tl.load(in_ptr3 + (x0 + 16 * (-12 + x1) + 64 * x2), tmp28 &
xmask, other=0.0)
tmp32 = tmp31 > tmp11
tmp33 = tmp31 * tmp13
tmp34 = tl.where(tmp32, tmp31, tmp33)
tmp35 = tl.full(tmp34.shape, 0.0, tmp34.dtype)
tmp36 = tl.where(tmp28, tmp34, tmp35)
tmp37 = tl.where(tmp21, tmp27, tmp36)
tmp38 = tl.where(tmp9, tmp17, tmp37)
tmp39 = tl.where(tmp4, tmp5, tmp38)
tl.store(out_ptr0 + x3, tmp39, xmask)
@triton.jit
def triton_poi_fused_constant_pad_nd_4(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_cat_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1280
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 20
x0 = xindex % 16
x2 = xindex // 320
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp9 & xmask,
other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + 16 * (-8 + x1) + 64 * x2), tmp21 &
xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tmp28 & tmp30
tmp32 = tl.load(in_ptr3 + (x0 + 16 * (-12 + x1) + 64 * x2), tmp31 &
xmask, other=0.0)
tmp33 = tmp32 > tmp11
tmp34 = tmp32 * tmp13
tmp35 = tl.where(tmp33, tmp32, tmp34)
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp31, tmp35, tmp36)
tmp38 = tmp0 >= tmp29
tl.full([1], 20, tl.int64)
tmp41 = tl.load(in_ptr4 + (x0 + 16 * (-16 + x1) + 64 * x2), tmp38 &
xmask, other=0.0)
tmp42 = tmp41 > tmp11
tmp43 = tmp41 * tmp13
tmp44 = tl.where(tmp42, tmp41, tmp43)
tmp45 = tl.full(tmp44.shape, 0.0, tmp44.dtype)
tmp46 = tl.where(tmp38, tmp44, tmp45)
tmp47 = tl.where(tmp31, tmp37, tmp46)
tmp48 = tl.where(tmp21, tmp27, tmp47)
tmp49 = tl.where(tmp9, tmp17, tmp48)
tmp50 = tl.where(tmp4, tmp5, tmp49)
tl.store(out_ptr0 + x3, tmp50, xmask)
@triton.jit
def triton_poi_fused_constant_pad_nd_6(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2880
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_7(in_out_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(in_out_ptr0 + x0, tmp5, xmask)
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_8(in_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
assert_size_stride(primals_5, (4, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (4, 20, 3, 3), (180, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(576)](primals_1, buf0, 576,
XBLOCK=128, num_warps=4, num_stages=1)
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
triton_poi_fused_cat_constant_pad_nd_1[grid(1152)](primals_1, buf1,
buf2, 1152, XBLOCK=128, num_warps=4, num_stages=1)
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32
)
triton_poi_fused_cat_constant_pad_nd_2[grid(1728)](primals_1, buf1,
buf3, buf4, 1728, XBLOCK=128, num_warps=4, num_stages=1)
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 16, 4, 4), (256, 16, 4, 1), torch.float32
)
triton_poi_fused_cat_3[grid(1024)](primals_1, buf1, buf3, buf5,
buf6, 1024, XBLOCK=256, num_warps=4, num_stages=1)
buf7 = empty_strided_cuda((4, 16, 6, 6), (576, 36, 6, 1), torch.float32
)
triton_poi_fused_constant_pad_nd_4[grid(2304)](buf6, buf7, 2304,
XBLOCK=256, num_warps=4, num_stages=1)
del buf6
buf8 = extern_kernels.convolution(buf7, primals_5, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4, 4), (64, 16, 4, 1))
buf9 = empty_strided_cuda((4, 20, 4, 4), (320, 16, 4, 1), torch.float32
)
triton_poi_fused_cat_5[grid(1280)](primals_1, buf1, buf3, buf5,
buf8, buf9, 1280, XBLOCK=128, num_warps=4, num_stages=1)
del primals_1
buf10 = empty_strided_cuda((4, 20, 6, 6), (720, 36, 6, 1), torch.
float32)
triton_poi_fused_constant_pad_nd_6[grid(2880)](buf9, buf10, 2880,
XBLOCK=128, num_warps=4, num_stages=1)
del buf9
buf11 = extern_kernels.convolution(buf10, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 4, 4, 4), (64, 16, 4, 1))
buf12 = buf11
del buf11
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_7[grid(256)](buf12,
buf13, 256, XBLOCK=128, num_warps=4, num_stages=1)
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf8,
buf14, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf8
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf5,
buf15, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf5
buf16 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf3,
buf16, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf3
buf17 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf1,
buf17, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf1
return (buf12, primals_2, primals_3, primals_4, primals_5, primals_6,
buf0, buf2, buf4, buf7, buf10, buf13, buf14, buf15, buf16, buf17)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class DenseConv2dLayer_5CNew(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(DenseConv2dLayer_5CNew, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv4 = Conv2dLayer(in_channels + latent_channels * 3,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv5 = Conv2dLayer(in_channels + latent_channels * 4,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, input_0):
primals_2 = self.conv1.conv2d.weight
primals_3 = self.conv2.conv2d.weight
primals_4 = self.conv3.conv2d.weight
primals_5 = self.conv4.conv2d.weight
primals_6 = self.conv5.conv2d.weight
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| LenKerr/Semantic-Colorization-GAN | DenseConv2dLayer_5C | false | 5,526 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super().__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.acti
# ... truncated (>4000 chars) for memory efficiency |
DotProductAttention | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/fz/cfzmg4qtw6jgry4nhlwopodzjz62ll3n3ykfox77hwd2crdnlh2w.py
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weights => exp
# Graph fragment:
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%bmm, 1), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor, %amax_default), kwargs = {})
# %mul_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_tensor, 0.5), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%mul_tensor_1,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp3 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + (x2), tmp17, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kj/ckjtlefzavjukjsytvkak6ek26zmzexpcbnlwelx4k5kascjxlf3.py
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weights => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bmm], Original ATen: [aten.bmm]
extern_kernels.bmm(arg1_1, reinterpret_tensor(arg0_1, (4, 4, 4), (16, 1, 4), 0), out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf0, buf1, 64, grid=grid(64), stream=stream0)
buf2 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf1, buf2, 64, grid=grid(64), stream=stream0)
buf3 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [att_weights, att_out], Original ATen: [aten._softmax, aten.bmm]
extern_kernels.bmm(buf2, arg2_1, out=buf3)
del arg2_1
del buf2
return (buf3, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
arg2_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1, arg2_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
class DotProductAttention(nn.Module):
def __init__(self, k_dim, dropout=0.1):
super(DotProductAttention, self).__init__()
self.scale = 1.0 / math.sqrt(k_dim)
self.softmax = nn.Softmax(dim=-1)
self.dropout = nn.Dropout(dropout)
def forward(self, q, k, v, mask=None):
"""
:param q: (bz, q_len, q_dim)
:param k: (bz, k_len, k_dim)
:param v: (bz, v_len, v_dim)
k_len == v_len v_dim == q_dim
:param mask: (bz, k_len) 填充部分为0
:return: (bz, q_len, v_dim)
"""
att_score = torch.bmm(q, k.transpose(1, 2)).mul(self.scale)
if mask is not None:
att_score.masked_fill_(~mask[:, None, :], -1000000000.0)
att_weights = self.softmax(att_score)
if self.training:
att_weights = self.dropout(att_weights)
att_out = torch.bmm(att_weights, v)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [[], {'k_dim': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp3 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + x2, tmp17, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(arg1_1, reinterpret_tensor(arg0_1, (4, 4, 4), (
16, 1, 4), 0), out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(64)](buf0, buf1, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf2 = buf0
del buf0
triton_poi_fused__softmax_1[grid(64)](buf1, buf2, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf3 = buf1
del buf1
extern_kernels.bmm(buf2, arg2_1, out=buf3)
del arg2_1
del buf2
return buf3,
class DotProductAttentionNew(nn.Module):
def __init__(self, k_dim, dropout=0.1):
super(DotProductAttentionNew, self).__init__()
self.scale = 1.0 / math.sqrt(k_dim)
self.softmax = nn.Softmax(dim=-1)
self.dropout = nn.Dropout(dropout)
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | DotProductAttention | false | 5,527 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import math
import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, k_dim, dropout=0.1):
super().__init__()
self.scale = 1.0 / math.sqrt(k_dim)
self.softmax = nn.Softmax(dim=-1)
self.dropout = nn.Dropout(dropout)
def forward(self, q, k, v, mask=None):
"""
:param q: (bz, q_len, q_dim)
:param k: (bz, k_len, k_dim)
:param v: (bz, v_len, v_dim)
k_len == v_len v_dim == q_dim
:param mask: (bz, k_len) 填充部分为0
:return: (bz, q_len, v_dim)
"""
att_score = torch.bmm(q, k.transpose(1, 2)).mul(self.scale)
if mask is not None:
att_score.masked_fill_(~mask[:, None, :], -1000000000.0)
att_weights = self.softmax(att_score)
if self.training:
att_weights = self.dropout(att_weights)
att_out = torch.bmm(att_weights, v)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [4]
|
ResidualDenseBlock_5C | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/td/ctdv3m5a33kovvtng5iilth4k6mtnyfcota6hhwoiqm34iumu7wi.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x => constant_pad_nd
# Graph fragment:
# %constant_pad_nd : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%primals_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/er/cerxfmiafmpnsn44ihnsss2pvd76g6x2i62qavnas5ygqo5mj43d.py
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat => cat
# x_3 => constant_pad_nd_1
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where], 1), kwargs = {})
# %constant_pad_nd_1 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_1 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 8
x3 = (xindex // 288)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + (x6), tmp30, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s3/cs34ah7ioinuak53u3oyhgsrsqeqfnovsb4djwnzfgkw44v6u547.py
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat_1 => cat_1
# x_6 => constant_pad_nd_2
# Graph fragment:
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1], 1), kwargs = {})
# %constant_pad_nd_2 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_2 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 12
x3 = (xindex // 432)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tmp30 = tl.full([1], 12, tl.int64)
tmp31 = tmp11 < tmp30
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + ((-5) + x0 + (4*x1) + (16*((-8) + x2)) + (64*x3)), tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + (x6), tmp42, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/2h/c2hdtntquyhqkvcjcbctuzny5hhijks5jcqplynkbwfrthtih25e.py
# Topologically Sorted Source Nodes: [cat_2], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat_2 => cat_2
# Graph fragment:
# %cat_2 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1, %where_2], 1), kwargs = {})
triton_poi_fused_cat_3 = async_compile.triton('triton_poi_fused_cat_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 16) % 16
x0 = xindex % 16
x2 = (xindex // 256)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + (16*((-4) + x1)) + (64*x2)), tmp9 & xmask, other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + (16*((-8) + x1)) + (64*x2)), tmp21 & xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tl.load(in_ptr3 + (x0 + (16*((-12) + x1)) + (64*x2)), tmp28 & xmask, other=0.0)
tmp32 = tmp31 > tmp11
tmp33 = tmp31 * tmp13
tmp34 = tl.where(tmp32, tmp31, tmp33)
tmp35 = tl.full(tmp34.shape, 0.0, tmp34.dtype)
tmp36 = tl.where(tmp28, tmp34, tmp35)
tmp37 = tl.where(tmp21, tmp27, tmp36)
tmp38 = tl.where(tmp9, tmp17, tmp37)
tmp39 = tl.where(tmp4, tmp5, tmp38)
tl.store(out_ptr0 + (x3), tmp39, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rc/crcdywpxyr2fxsveen5yolsghlkstdms4uvwp7zkfwhcmkjmpxpg.py
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x_9 => constant_pad_nd_3
# Graph fragment:
# %constant_pad_nd_3 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_2, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_4 = async_compile.triton('triton_poi_fused_constant_pad_nd_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_4(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4p/c4pwo5hnpxyletkwji2bfgdsgviax75psfwuih5qzyaah2skdcdn.py
# Topologically Sorted Source Nodes: [cat_3], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat_3 => cat_3
# Graph fragment:
# %cat_3 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1, %where_2, %where_3], 1), kwargs = {})
triton_poi_fused_cat_5 = async_compile.triton('triton_poi_fused_cat_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1280
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 16) % 20
x0 = xindex % 16
x2 = (xindex // 320)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + (16*((-4) + x1)) + (64*x2)), tmp9 & xmask, other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + (16*((-8) + x1)) + (64*x2)), tmp21 & xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tmp28 & tmp30
tmp32 = tl.load(in_ptr3 + (x0 + (16*((-12) + x1)) + (64*x2)), tmp31 & xmask, other=0.0)
tmp33 = tmp32 > tmp11
tmp34 = tmp32 * tmp13
tmp35 = tl.where(tmp33, tmp32, tmp34)
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp31, tmp35, tmp36)
tmp38 = tmp0 >= tmp29
tmp39 = tl.full([1], 20, tl.int64)
tmp40 = tmp0 < tmp39
tmp41 = tl.load(in_ptr4 + (x0 + (16*((-16) + x1)) + (64*x2)), tmp38 & xmask, other=0.0)
tmp42 = tmp41 > tmp11
tmp43 = tmp41 * tmp13
tmp44 = tl.where(tmp42, tmp41, tmp43)
tmp45 = tl.full(tmp44.shape, 0.0, tmp44.dtype)
tmp46 = tl.where(tmp38, tmp44, tmp45)
tmp47 = tl.where(tmp31, tmp37, tmp46)
tmp48 = tl.where(tmp21, tmp27, tmp47)
tmp49 = tl.where(tmp9, tmp17, tmp48)
tmp50 = tl.where(tmp4, tmp5, tmp49)
tl.store(out_ptr0 + (x3), tmp50, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s7/cs7remm7dxallug7mpmvc7novi2rckl5n32tcwbq5syx4cb4lqk6.py
# Topologically Sorted Source Nodes: [x_12], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x_12 => constant_pad_nd_4
# Graph fragment:
# %constant_pad_nd_4 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_3, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_6 = async_compile.triton('triton_poi_fused_constant_pad_nd_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_6(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2880
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/yx/cyxxe53fwtxbbz4ado2vqkuowx34tuo7up564iwpnh2sggpenycv.py
# Topologically Sorted Source Nodes: [x_14, mul, x5], Original ATen: [aten.leaky_relu, aten.mul, aten.add, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# mul => mul_5
# x5 => add
# x_14 => gt_4, mul_4, where_4
# Graph fragment:
# %gt_4 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_4, 0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_4, 0.2), kwargs = {})
# %where_4 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_4, %convolution_4, %mul_4), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%where_4, 0.1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_5, %primals_1), kwargs = {})
# %gt_5 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_4, 0), kwargs = {})
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7 = async_compile.triton('triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i1', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp8 = tl.load(in_ptr1 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = 0.1
tmp7 = tmp5 * tmp6
tmp9 = tmp7 + tmp8
tmp10 = tmp5 > tmp1
tl.store(out_ptr0 + (x0), tmp9, xmask)
tl.store(out_ptr1 + (x0), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/is/cisodw5gxdmlgapnrqpu7bcsl5srnastcmg2c73tt3lqfqu2bmk6.py
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x_11 => gt_3, mul_3, where_3
# Graph fragment:
# %gt_3 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_3, 0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_3, 0.2), kwargs = {})
# %where_3 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_3, %convolution_3, %mul_3), kwargs = {})
# %gt_6 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_3, 0), kwargs = {})
triton_poi_fused_leaky_relu_leaky_relu_backward_8 = async_compile.triton('triton_poi_fused_leaky_relu_leaky_relu_backward_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_leaky_relu_backward_8', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_8(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
assert_size_stride(primals_5, (4, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (4, 20, 3, 3), (180, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(primals_1, buf0, 576, grid=grid(576), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_1.run(primals_1, buf1, buf2, 1152, grid=grid(1152), stream=stream0)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.convolution]
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_2.run(primals_1, buf1, buf3, buf4, 1728, grid=grid(1728), stream=stream0)
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.convolution]
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 16, 4, 4), (256, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_2], Original ATen: [aten.cat]
triton_poi_fused_cat_3.run(primals_1, buf1, buf3, buf5, buf6, 1024, grid=grid(1024), stream=stream0)
buf7 = empty_strided_cuda((4, 16, 6, 6), (576, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.constant_pad_nd]
triton_poi_fused_constant_pad_nd_4.run(buf6, buf7, 2304, grid=grid(2304), stream=stream0)
del buf6
# Topologically Sorted Source Nodes: [x_10], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_5, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4, 4), (64, 16, 4, 1))
buf9 = empty_strided_cuda((4, 20, 4, 4), (320, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_3], Original ATen: [aten.cat]
triton_poi_fused_cat_5.run(primals_1, buf1, buf3, buf5, buf8, buf9, 1280, grid=grid(1280), stream=stream0)
buf10 = empty_strided_cuda((4, 20, 6, 6), (720, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_12], Original ATen: [aten.constant_pad_nd]
triton_poi_fused_constant_pad_nd_6.run(buf9, buf10, 2880, grid=grid(2880), stream=stream0)
del buf9
# Topologically Sorted Source Nodes: [x_13], Original ATen: [aten.convolution]
buf11 = extern_kernels.convolution(buf10, primals_6, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 4, 4, 4), (64, 16, 4, 1))
buf12 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_14, mul, x5], Original ATen: [aten.leaky_relu, aten.mul, aten.add, aten.leaky_relu_backward]
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7.run(buf11, primals_1, buf12, buf13, 256, grid=grid(256), stream=stream0)
del buf11
del primals_1
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf8, buf14, 256, grid=grid(256), stream=stream0)
del buf8
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf5, buf15, 256, grid=grid(256), stream=stream0)
del buf5
buf16 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf3, buf16, 256, grid=grid(256), stream=stream0)
del buf3
buf17 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_8.run(buf1, buf17, 256, grid=grid(256), stream=stream0)
del buf1
return (buf12, primals_2, primals_3, primals_4, primals_5, primals_6, buf0, buf2, buf4, buf7, buf10, buf13, buf14, buf15, buf16, buf17, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 12, 3, 3), (108, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 20, 3, 3), (180, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class ResidualDenseBlock_5C(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(ResidualDenseBlock_5C, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv4 = Conv2dLayer(in_channels + latent_channels * 3,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv5 = Conv2dLayer(in_channels + latent_channels * 4,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, x):
residual = x
x1 = self.conv1(x)
x2 = self.conv2(torch.cat((x, x1), 1))
x3 = self.conv3(torch.cat((x, x1, x2), 1))
x4 = self.conv4(torch.cat((x, x1, x2, x3), 1))
x5 = self.conv5(torch.cat((x, x1, x2, x3, x4), 1))
x5 = 0.1 * x5 + residual
return x5
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'latent_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 8
x3 = xindex // 288
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tl.full([1], 8, tl.int64)
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + x6, tmp30, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 12
x3 = xindex // 432
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tl.full([1], 12, tl.int64)
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + (-5 + x0 + 4 * x1 + 16 * (-8 + x2) + 64 * x3),
tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + x6, tmp42, xmask)
@triton.jit
def triton_poi_fused_cat_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 16
x0 = xindex % 16
x2 = xindex // 256
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp9 & xmask,
other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + 16 * (-8 + x1) + 64 * x2), tmp21 &
xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tl.full([1], 16, tl.int64)
tmp31 = tl.load(in_ptr3 + (x0 + 16 * (-12 + x1) + 64 * x2), tmp28 &
xmask, other=0.0)
tmp32 = tmp31 > tmp11
tmp33 = tmp31 * tmp13
tmp34 = tl.where(tmp32, tmp31, tmp33)
tmp35 = tl.full(tmp34.shape, 0.0, tmp34.dtype)
tmp36 = tl.where(tmp28, tmp34, tmp35)
tmp37 = tl.where(tmp21, tmp27, tmp36)
tmp38 = tl.where(tmp9, tmp17, tmp37)
tmp39 = tl.where(tmp4, tmp5, tmp38)
tl.store(out_ptr0 + x3, tmp39, xmask)
@triton.jit
def triton_poi_fused_constant_pad_nd_4(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_cat_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1280
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 20
x0 = xindex % 16
x2 = xindex // 320
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp9 & xmask,
other=0.0)
tmp11 = 0.0
tmp12 = tmp10 > tmp11
tmp13 = 0.2
tmp14 = tmp10 * tmp13
tmp15 = tl.where(tmp12, tmp10, tmp14)
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp9, tmp15, tmp16)
tmp18 = tmp0 >= tmp7
tmp19 = tl.full([1], 12, tl.int64)
tmp20 = tmp0 < tmp19
tmp21 = tmp18 & tmp20
tmp22 = tl.load(in_ptr2 + (x0 + 16 * (-8 + x1) + 64 * x2), tmp21 &
xmask, other=0.0)
tmp23 = tmp22 > tmp11
tmp24 = tmp22 * tmp13
tmp25 = tl.where(tmp23, tmp22, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp21, tmp25, tmp26)
tmp28 = tmp0 >= tmp19
tmp29 = tl.full([1], 16, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tmp28 & tmp30
tmp32 = tl.load(in_ptr3 + (x0 + 16 * (-12 + x1) + 64 * x2), tmp31 &
xmask, other=0.0)
tmp33 = tmp32 > tmp11
tmp34 = tmp32 * tmp13
tmp35 = tl.where(tmp33, tmp32, tmp34)
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp31, tmp35, tmp36)
tmp38 = tmp0 >= tmp29
tl.full([1], 20, tl.int64)
tmp41 = tl.load(in_ptr4 + (x0 + 16 * (-16 + x1) + 64 * x2), tmp38 &
xmask, other=0.0)
tmp42 = tmp41 > tmp11
tmp43 = tmp41 * tmp13
tmp44 = tl.where(tmp42, tmp41, tmp43)
tmp45 = tl.full(tmp44.shape, 0.0, tmp44.dtype)
tmp46 = tl.where(tmp38, tmp44, tmp45)
tmp47 = tl.where(tmp31, tmp37, tmp46)
tmp48 = tl.where(tmp21, tmp27, tmp47)
tmp49 = tl.where(tmp9, tmp17, tmp48)
tmp50 = tl.where(tmp4, tmp5, tmp49)
tl.store(out_ptr0 + x3, tmp50, xmask)
@triton.jit
def triton_poi_fused_constant_pad_nd_6(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2880
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7(in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp8 = tl.load(in_ptr1 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = 0.1
tmp7 = tmp5 * tmp6
tmp9 = tmp7 + tmp8
tmp10 = tmp5 > tmp1
tl.store(out_ptr0 + x0, tmp9, xmask)
tl.store(out_ptr1 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_8(in_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
assert_size_stride(primals_5, (4, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (4, 20, 3, 3), (180, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(576)](primals_1, buf0, 576,
XBLOCK=128, num_warps=4, num_stages=1)
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
triton_poi_fused_cat_constant_pad_nd_1[grid(1152)](primals_1, buf1,
buf2, 1152, XBLOCK=128, num_warps=4, num_stages=1)
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32
)
triton_poi_fused_cat_constant_pad_nd_2[grid(1728)](primals_1, buf1,
buf3, buf4, 1728, XBLOCK=128, num_warps=4, num_stages=1)
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 16, 4, 4), (256, 16, 4, 1), torch.float32
)
triton_poi_fused_cat_3[grid(1024)](primals_1, buf1, buf3, buf5,
buf6, 1024, XBLOCK=256, num_warps=4, num_stages=1)
buf7 = empty_strided_cuda((4, 16, 6, 6), (576, 36, 6, 1), torch.float32
)
triton_poi_fused_constant_pad_nd_4[grid(2304)](buf6, buf7, 2304,
XBLOCK=256, num_warps=4, num_stages=1)
del buf6
buf8 = extern_kernels.convolution(buf7, primals_5, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4, 4), (64, 16, 4, 1))
buf9 = empty_strided_cuda((4, 20, 4, 4), (320, 16, 4, 1), torch.float32
)
triton_poi_fused_cat_5[grid(1280)](primals_1, buf1, buf3, buf5,
buf8, buf9, 1280, XBLOCK=128, num_warps=4, num_stages=1)
buf10 = empty_strided_cuda((4, 20, 6, 6), (720, 36, 6, 1), torch.
float32)
triton_poi_fused_constant_pad_nd_6[grid(2880)](buf9, buf10, 2880,
XBLOCK=128, num_warps=4, num_stages=1)
del buf9
buf11 = extern_kernels.convolution(buf10, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 4, 4, 4), (64, 16, 4, 1))
buf12 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_7[grid(256)](
buf11, primals_1, buf12, buf13, 256, XBLOCK=256, num_warps=4,
num_stages=1)
del buf11
del primals_1
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf8,
buf14, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf8
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf5,
buf15, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf5
buf16 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf3,
buf16, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf3
buf17 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_8[grid(256)](buf1,
buf17, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf1
return (buf12, primals_2, primals_3, primals_4, primals_5, primals_6,
buf0, buf2, buf4, buf7, buf10, buf13, buf14, buf15, buf16, buf17)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class ResidualDenseBlock_5CNew(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(ResidualDenseBlock_5CNew, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv4 = Conv2dLayer(in_channels + latent_channels * 3,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv5 = Conv2dLayer(in_channels + latent_channels * 4,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, input_0):
primals_2 = self.conv1.conv2d.weight
primals_3 = self.conv2.conv2d.weight
primals_4 = self.conv3.conv2d.weight
primals_5 = self.conv4.conv2d.weight
primals_6 = self.conv5.conv2d.weight
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| LenKerr/Semantic-Colorization-GAN | ResidualDenseBlock_5C | false | 5,528 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super().__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.acti
# ... truncated (>4000 chars) for memory efficiency |
TransposeAdaINConv2dLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ks/cksi7p5iq6amvatj4ylxlyaufmiucpgysxv6irpv4u55ro6d7sin.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten._unsafe_index]
# Source node to ATen node mapping:
# x => _unsafe_index
# Graph fragment:
# %_unsafe_index : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%primals_1, [None, None, %unsqueeze, %convert_element_type_1]), kwargs = {})
triton_poi_fused__unsafe_index_0 = async_compile.triton('triton_poi_fused__unsafe_index_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__unsafe_index_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__unsafe_index_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 65536
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x1 = (xindex // 8) % 8
x0 = xindex % 8
x2 = (xindex // 64)
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + (4*tmp4) + (16*x2)), None, eviction_policy='evict_last')
tl.store(out_ptr0 + (x4), tmp9, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ya/cya2grnbhraytq2wzrkx5sd2ottwnbrnd5ohd2xstcxyryneuc25.py
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add_4
# mv => mul_4, sum_1
# norm => pow_1, pow_2, sum_2
# truediv => div
# Graph fragment:
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute, %primals_2), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_4, [1]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, None), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %add_4 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 1e-12), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add_4), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_1 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {5: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 6), equal_to_1=(5,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp25, None)
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp26, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qi/cqiozgecuvqtnurxrggbllqpuci3n65ycew5qi5gdqg44ypxzegy.py
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
# Source node to ATen node mapping:
# mv_1 => mul_5, sum_3
# truediv => div
# Graph fragment:
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add_4), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %div), kwargs = {})
# %sum_3 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_5, [1]), kwargs = {})
triton_per_fused_div_mv_2 = async_compile.triton('triton_per_fused_div_mv_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_mv_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (64*x0)), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qa/cqaed4ios3xqwlv4d3cciikkdz7d73vhwkegurd5cxca3y7htmvg.py
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_1 => add_5
# norm_1 => pow_3, pow_4, sum_4
# truediv_1 => div_1
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, None), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_4, 0.5), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_4, 1e-12), kwargs = {})
# %div_1 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, %add_5), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_3 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp8, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/c2/cc2arficwjs4sforhl25gdfmb3uzfg7hkw46gq3mxgv57jy52z32.py
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
# Source node to ATen node mapping:
# sigma => mul_7, sum_6
# Graph fragment:
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %sum_3), kwargs = {})
# %sum_6 : [num_users=2] = call_function[target=torch.ops.aten.sum.default](args = (%mul_7,), kwargs = {})
triton_per_fused_dot_4 = async_compile.triton('triton_per_fused_dot_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_dot_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kw/ckwzptlssdpmtxi6pt23ik63xcuqar2giaakuqtgizxlg5weagc7.py
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_2 => div_2
# Graph fragment:
# %div_2 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_4, %expand), kwargs = {})
triton_poi_fused_div_5 = async_compile.triton('triton_poi_fused_div_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/bv/cbv6jakt3kcc7zdwd3bg7driglzwnrjamfcojzypctox67qmiqqu.py
# Topologically Sorted Source Nodes: [x_2, mean, x_3, mean_2, add_2, var_2, x_4, mul_2, content, x_5], Original ATen: [aten.convolution, aten.mean, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# add_2 => add_6
# content => add_7
# mean => mean
# mean_2 => mean_1
# mul_2 => mul_10
# var_2 => rsqrt
# x_2 => convolution
# x_3 => sub
# x_4 => mul_9
# x_5 => gt, mul_11, where
# Graph fragment:
# %convolution : [num_users=3] = call_function[target=torch.ops.aten.convolution.default](args = (%_unsafe_index, %div_2, %primals_5, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view_3, [2]), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%convolution, %view_4), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view_5, [2]), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_6, 1e-08), kwargs = {})
# %rsqrt : [num_users=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_6,), kwargs = {})
# %mul_9 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_10 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_9, %select), kwargs = {})
# %add_7 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_10, %select_1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%add_7, 0), kwargs = {})
# %mul_11 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_7, 0.2), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %add_7, %mul_11), kwargs = {})
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where, 0), kwargs = {})
triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6 = async_compile.triton('triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1024, 32],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*i1', 7: 'i32', 8: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1', 'in_out_ptr2', 'in_out_ptr3'], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6(in_out_ptr0, in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1024
rnumel = 25
RBLOCK: tl.constexpr = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r2 = rindex
x3 = xindex
x0 = xindex % 4
x1 = (xindex // 4)
tmp0 = tl.load(in_out_ptr0 + (r2 + (25*x3)), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (x0 + (8*x1)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (4 + x0 + (8*x1)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.where(rmask & xmask, tmp3, 0)
tmp6 = tl.sum(tmp5, 1)[:, None]
tmp7 = 25.0
tmp8 = tmp6 / tmp7
tmp9 = tmp2 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.where(rmask & xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tmp14 / tmp7
tmp16 = 1e-08
tmp17 = tmp15 + tmp16
tmp18 = libdevice.rsqrt(tmp17)
tmp19 = tmp9 * tmp18
tmp21 = tmp19 * tmp20
tmp23 = tmp21 + tmp22
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp29 = tmp28 > tmp24
tl.store(in_out_ptr0 + (r2 + (25*x3)), tmp2, rmask & xmask)
tl.debug_barrier()
tl.store(in_out_ptr1 + (x3), tmp8, xmask)
tl.debug_barrier()
tl.store(in_out_ptr2 + (x3), tmp18, xmask)
tl.store(in_out_ptr3 + (r2 + (25*x3)), tmp28, rmask & xmask)
tl.store(out_ptr0 + (r2 + (25*x3)), tmp29, rmask & xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (64, ), (1, ))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((256, 4, 8, 8), (256, 64, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten._unsafe_index]
stream0 = get_raw_stream(0)
triton_poi_fused__unsafe_index_0.run(primals_1, buf0, 65536, grid=grid(65536), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((64, ), (1, ), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2; del buf2 # reuse
buf23 = empty_strided_cuda((64, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_mv_1.run(buf3, primals_4, primals_2, buf1, buf23, 1, 64, grid=grid(1), stream=stream0)
buf4 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
triton_per_fused_div_mv_2.run(primals_4, buf1, buf3, buf4, 4, 64, grid=grid(4), stream=stream0)
buf6 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_3.run(buf4, buf6, 1, 4, grid=grid(1), stream=stream0)
buf7 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
triton_per_fused_dot_4.run(buf6, buf4, buf7, 1, 4, grid=grid(1), stream=stream0)
del buf4
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
triton_poi_fused_div_5.run(primals_4, buf7, buf8, 256, grid=grid(256), stream=stream0)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (256, 4, 5, 5), (100, 25, 5, 1))
buf10 = buf9; del buf9 # reuse
buf11 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf12 = buf11; del buf11 # reuse
buf13 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf14 = reinterpret_tensor(buf13, (256, 4, 1, 1), (4, 1, 1, 1), 0); del buf13 # reuse
buf15 = empty_strided_cuda((256, 4, 5, 5), (100, 25, 5, 1), torch.float32)
buf16 = buf15; del buf15 # reuse
buf17 = empty_strided_cuda((256, 4, 5, 5), (100, 25, 5, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2, mean, x_3, mean_2, add_2, var_2, x_4, mul_2, content, x_5], Original ATen: [aten.convolution, aten.mean, aten.sub, aten.add, aten.rsqrt, aten.mul, aten.leaky_relu, aten.leaky_relu_backward]
triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6.run(buf10, buf12, buf14, buf16, primals_5, primals_6, buf17, 1024, 25, grid=grid(1024), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [], Original ATen: []
buf18 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf18, (4, ), (1, ))
del buf1
# Topologically Sorted Source Nodes: [truediv], Original ATen: [aten.div]
buf24 = torch.ops.aten.set_.source_Tensor(primals_3, buf23)
assert_size_stride(buf24, (64, ), (1, ))
del primals_3
return (buf16, buf8, primals_2, primals_4, buf0, buf3, buf6, buf7, buf8, buf10, reinterpret_tensor(buf12, (256, 4, 1, 1), (4, 1, 1, 1), 0), buf14, reinterpret_tensor(primals_6, (256, 4, 1, 1), (8, 1, 1, 1), 0), buf17, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((256, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((32, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, x, style):
x = self.pad(x)
x = self.conv2d(x)
x = self.norm(x, style)
if self.activation:
x = self.activation(x)
return x
class TransposeAdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True,
scale_factor=2):
super(TransposeAdaINConv2dLayer, self).__init__()
self.scale_factor = scale_factor
self.conv2d = AdaINConv2dLayer(in_channels, out_channels,
kernel_size, stride, padding, dilation, pad_type, activation, sn)
def forward(self, x, style):
x = F.interpolate(x, scale_factor=self.scale_factor, mode='nearest')
x = self.conv2d(x, style)
return x
def get_inputs():
return [torch.rand([256, 4, 4, 4]), torch.rand([32, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__unsafe_index_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x1 = xindex // 8 % 8
x0 = xindex % 8
x2 = xindex // 64
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + 4 * tmp4 + 16 * x2), None,
eviction_policy='evict_last')
tl.store(out_ptr0 + x4, tmp9, None)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp25, None)
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp26, None)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 4
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 64 * x0), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp8, None)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6(
in_out_ptr0, in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1,
out_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 1024
rnumel = 25
RBLOCK: tl.constexpr = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r2 = rindex
x3 = xindex
x0 = xindex % 4
x1 = xindex // 4
tmp0 = tl.load(in_out_ptr0 + (r2 + 25 * x3), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + (4 + x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.where(rmask & xmask, tmp3, 0)
tmp6 = tl.sum(tmp5, 1)[:, None]
tmp7 = 25.0
tmp8 = tmp6 / tmp7
tmp9 = tmp2 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.where(rmask & xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tmp14 / tmp7
tmp16 = 1e-08
tmp17 = tmp15 + tmp16
tmp18 = libdevice.rsqrt(tmp17)
tmp19 = tmp9 * tmp18
tmp21 = tmp19 * tmp20
tmp23 = tmp21 + tmp22
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp29 = tmp28 > tmp24
tl.store(in_out_ptr0 + (r2 + 25 * x3), tmp2, rmask & xmask)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp8, xmask)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp18, xmask)
tl.store(in_out_ptr3 + (r2 + 25 * x3), tmp28, rmask & xmask)
tl.store(out_ptr0 + (r2 + 25 * x3), tmp29, rmask & xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (256, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (64,), (1,))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (32, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((256, 4, 8, 8), (256, 64, 8, 1), torch.
float32)
get_raw_stream(0)
triton_poi_fused__unsafe_index_0[grid(65536)](primals_1, buf0,
65536, XBLOCK=256, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((64,), (1,), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
buf23 = empty_strided_cuda((64,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_mv_1[grid(1)](buf3,
primals_4, primals_2, buf1, buf23, 1, 64, XBLOCK=1, num_warps=2,
num_stages=1)
buf4 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_div_mv_2[grid(4)](primals_4, buf1, buf3, buf4, 4,
64, XBLOCK=1, num_warps=2, num_stages=1)
buf6 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_3[grid(1)](buf4, buf6,
1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf7 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_dot_4[grid(1)](buf6, buf4, buf7, 1, 4, XBLOCK=1,
num_warps=2, num_stages=1)
del buf4
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_div_5[grid(256)](primals_4, buf7, buf8, 256,
XBLOCK=256, num_warps=4, num_stages=1)
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (256, 4, 5, 5), (100, 25, 5, 1))
buf10 = buf9
del buf9
buf11 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf12 = buf11
del buf11
buf13 = empty_strided_cuda((256, 4), (4, 1), torch.float32)
buf14 = reinterpret_tensor(buf13, (256, 4, 1, 1), (4, 1, 1, 1), 0)
del buf13
buf15 = empty_strided_cuda((256, 4, 5, 5), (100, 25, 5, 1), torch.
float32)
buf16 = buf15
del buf15
buf17 = empty_strided_cuda((256, 4, 5, 5), (100, 25, 5, 1), torch.bool)
triton_per_fused_add_convolution_leaky_relu_leaky_relu_backward_mean_mul_rsqrt_sub_6[
grid(1024)](buf10, buf12, buf14, buf16, primals_5, primals_6,
buf17, 1024, 25, XBLOCK=1, num_warps=2, num_stages=1)
del primals_5
buf18 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf18, (4,), (1,))
del buf1
buf24 = torch.ops.aten.set_.source_Tensor(primals_3, buf23)
assert_size_stride(buf24, (64,), (1,))
del primals_3
return (buf16, buf8, primals_2, primals_4, buf0, buf3, buf6, buf7, buf8,
buf10, reinterpret_tensor(buf12, (256, 4, 1, 1), (4, 1, 1, 1), 0),
buf14, reinterpret_tensor(primals_6, (256, 4, 1, 1), (8, 1, 1, 1),
0), buf17)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super(AdaptiveInstanceNorm2d, self).__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super(AdaINConv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation)
def forward(self, x, style):
x = self.pad(x)
x = self.conv2d(x)
x = self.norm(x, style)
if self.activation:
x = self.activation(x)
return x
class TransposeAdaINConv2dLayerNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True,
scale_factor=2):
super(TransposeAdaINConv2dLayerNew, self).__init__()
self.scale_factor = scale_factor
self.conv2d = AdaINConv2dLayer(in_channels, out_channels,
kernel_size, stride, padding, dilation, pad_type, activation, sn)
def forward(self, input_0, input_1):
primals_2 = self.conv2d.conv2d.module.bias
primals_5 = self.conv2d.conv2d.module.weight_u
primals_3 = self.conv2d.conv2d.module.weight_v
primals_4 = self.conv2d.conv2d.module.weight_bar
primals_1 = input_0
primals_6 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| LenKerr/Semantic-Colorization-GAN | TransposeAdaINConv2dLayer | false | 5,529 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class AdaptiveInstanceNorm2d(nn.Module):
def __init__(self, eps=1e-08):
super().__init__()
self.eps = eps
def IN_noWeight(self, x):
N, C = x.size(0), x.size(1)
mean = x.contiguous().view(N, C, -1).mean(2).contiguous().view(N, C,
1, 1)
x = x - mean
var = torch.mul(x, x)
var = var.contiguous().view(N, C, -1).mean(2).contiguous().view(N,
C, 1, 1)
var = torch.rsqrt(var + self.eps)
x = x * var
return x
def Apply_style(self, content, style):
style = style.contiguous().view([-1, 2, content.size(1), 1, 1])
content = content * style[:, 0] + style[:, 1]
return content
def forward(self, content, style):
normalized_content = self.IN_noWeight(content)
stylized_content = self.Apply_style(normalized_content, style)
return stylized_content
class AdaINConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
self.norm = AdaptiveInstanceNorm2d()
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigm
# ... truncated (>4000 chars) for memory efficiency |
TransposeConv2dLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/oj/cojl5mb3pzv5jbmfzjkbac5hekbmpvb72kof6ouyyasitrogdd6n.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten._unsafe_index]
# Source node to ATen node mapping:
# x => _unsafe_index
# Graph fragment:
# %_unsafe_index : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%primals_1, [None, None, %unsqueeze, %convert_element_type_1]), kwargs = {})
triton_poi_fused__unsafe_index_0 = async_compile.triton('triton_poi_fused__unsafe_index_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__unsafe_index_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__unsafe_index_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 8) % 8
x0 = xindex % 8
x2 = (xindex // 64)
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + (4*tmp4) + (16*x2)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x4), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ya/cya2grnbhraytq2wzrkx5sd2ottwnbrnd5ohd2xstcxyryneuc25.py
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add_4
# mv => mul_4, sum_1
# norm => pow_1, pow_2, sum_2
# truediv => div
# Graph fragment:
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute, %primals_2), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_4, [1]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, None), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %add_4 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 1e-12), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add_4), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_1 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {5: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 6), equal_to_1=(5,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp25, None)
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp26, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qi/cqiozgecuvqtnurxrggbllqpuci3n65ycew5qi5gdqg44ypxzegy.py
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
# Source node to ATen node mapping:
# mv_1 => mul_5, sum_3
# truediv => div
# Graph fragment:
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add_4), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %div), kwargs = {})
# %sum_3 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_5, [1]), kwargs = {})
triton_per_fused_div_mv_2 = async_compile.triton('triton_per_fused_div_mv_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_mv_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (64*x0)), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qa/cqaed4ios3xqwlv4d3cciikkdz7d73vhwkegurd5cxca3y7htmvg.py
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_1 => add_5
# norm_1 => pow_3, pow_4, sum_4
# truediv_1 => div_1
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, None), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_4, 0.5), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_4, 1e-12), kwargs = {})
# %div_1 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, %add_5), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_3 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp8, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/c2/cc2arficwjs4sforhl25gdfmb3uzfg7hkw46gq3mxgv57jy52z32.py
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
# Source node to ATen node mapping:
# sigma => mul_7, sum_6
# Graph fragment:
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %sum_3), kwargs = {})
# %sum_6 : [num_users=2] = call_function[target=torch.ops.aten.sum.default](args = (%mul_7,), kwargs = {})
triton_per_fused_dot_4 = async_compile.triton('triton_per_fused_dot_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_dot_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kw/ckwzptlssdpmtxi6pt23ik63xcuqar2giaakuqtgizxlg5weagc7.py
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_2 => div_2
# Graph fragment:
# %div_2 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_4, %expand), kwargs = {})
triton_poi_fused_div_5 = async_compile.triton('triton_poi_fused_div_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/n4/cn4gcl6zss6vqmwfppmtgzfmg5aubgn4jeyyouri7roaln7yxyx4.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x_3 => gt, mul_8, where
# Graph fragment:
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution, 0), kwargs = {})
# %mul_8 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution, 0.2), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %convolution, %mul_8), kwargs = {})
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where, 0), kwargs = {})
triton_poi_fused_leaky_relu_leaky_relu_backward_6 = async_compile.triton('triton_poi_fused_leaky_relu_leaky_relu_backward_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_leaky_relu_backward_6', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_6(in_out_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(in_out_ptr0 + (x0), tmp5, xmask)
tl.store(out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (64, ), (1, ))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 8, 8), (256, 64, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten._unsafe_index]
stream0 = get_raw_stream(0)
triton_poi_fused__unsafe_index_0.run(primals_1, buf0, 1024, grid=grid(1024), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((64, ), (1, ), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2; del buf2 # reuse
buf17 = empty_strided_cuda((64, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_mv_1.run(buf3, primals_4, primals_2, buf1, buf17, 1, 64, grid=grid(1), stream=stream0)
buf4 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [truediv, mv_1], Original ATen: [aten.div, aten.mv]
triton_per_fused_div_mv_2.run(primals_4, buf1, buf3, buf4, 4, 64, grid=grid(4), stream=stream0)
buf6 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_3.run(buf4, buf6, 1, 4, grid=grid(1), stream=stream0)
buf7 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sigma], Original ATen: [aten.dot]
triton_per_fused_dot_4.run(buf6, buf4, buf7, 1, 4, grid=grid(1), stream=stream0)
del buf4
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
triton_poi_fused_div_5.run(primals_4, buf7, buf8, 256, grid=grid(256), stream=stream0)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 4, 5, 5), (100, 25, 5, 1))
buf10 = buf9; del buf9 # reuse
buf11 = empty_strided_cuda((4, 4, 5, 5), (100, 25, 5, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_6.run(buf10, buf11, 400, grid=grid(400), stream=stream0)
# Topologically Sorted Source Nodes: [], Original ATen: []
buf12 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf12, (4, ), (1, ))
del buf1
# Topologically Sorted Source Nodes: [truediv], Original ATen: [aten.div]
buf18 = torch.ops.aten.set_.source_Tensor(primals_3, buf17)
assert_size_stride(buf18, (64, ), (1, ))
del primals_3
return (buf10, buf8, primals_2, primals_4, buf0, buf3, buf6, buf7, buf8, buf11, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class TransposeConv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True, scale_factor=2):
super(TransposeConv2dLayer, self).__init__()
self.scale_factor = scale_factor
self.conv2d = Conv2dLayer(in_channels, out_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
def forward(self, x):
x = F.interpolate(x, scale_factor=self.scale_factor, mode='nearest')
x = self.conv2d(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__unsafe_index_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 8 % 8
x0 = xindex % 8
x2 = xindex // 64
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + 4 * tmp4 + 16 * x2), xmask,
eviction_policy='evict_last')
tl.store(out_ptr0 + x4, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_1(in_out_ptr0, in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (64 + r0), None)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (128 + r0), None)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (192 + r0), None)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp18, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp25, None)
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp26, None)
@triton.jit
def triton_per_fused_div_mv_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 4
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 64 * x0), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_3(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp8, None)
@triton.jit
def triton_per_fused_dot_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
@triton.jit
def triton_poi_fused_div_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_6(in_out_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(in_out_ptr0 + x0, tmp5, xmask)
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (64,), (1,))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 8, 8), (256, 64, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__unsafe_index_0[grid(1024)](primals_1, buf0, 1024,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((64,), (1,), torch.float32)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
buf17 = empty_strided_cuda((64,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_mv_1[grid(1)](buf3,
primals_4, primals_2, buf1, buf17, 1, 64, XBLOCK=1, num_warps=2,
num_stages=1)
buf4 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_div_mv_2[grid(4)](primals_4, buf1, buf3, buf4, 4,
64, XBLOCK=1, num_warps=2, num_stages=1)
buf6 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_3[grid(1)](buf4, buf6,
1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf7 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_dot_4[grid(1)](buf6, buf4, buf7, 1, 4, XBLOCK=1,
num_warps=2, num_stages=1)
del buf4
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_div_5[grid(256)](primals_4, buf7, buf8, 256,
XBLOCK=256, num_warps=4, num_stages=1)
buf9 = extern_kernels.convolution(buf0, buf8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 4, 5, 5), (100, 25, 5, 1))
buf10 = buf9
del buf9
buf11 = empty_strided_cuda((4, 4, 5, 5), (100, 25, 5, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_6[grid(400)](buf10,
buf11, 400, XBLOCK=256, num_warps=4, num_stages=1)
buf12 = torch.ops.aten.set_.source_Tensor(primals_2, buf6)
assert_size_stride(buf12, (4,), (1,))
del buf1
buf18 = torch.ops.aten.set_.source_Tensor(primals_3, buf17)
assert_size_stride(buf18, (64,), (1,))
del primals_3
return (buf10, buf8, primals_2, primals_4, buf0, buf3, buf6, buf7, buf8,
buf11)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class TransposeConv2dLayerNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True, scale_factor=2):
super(TransposeConv2dLayerNew, self).__init__()
self.scale_factor = scale_factor
self.conv2d = Conv2dLayer(in_channels, out_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
def forward(self, input_0):
primals_2 = self.conv2d.conv2d.module.weight_u
primals_3 = self.conv2d.conv2d.module.weight_v
primals_1 = self.conv2d.conv2d.module.weight_bar
primals_4 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| LenKerr/Semantic-Colorization-GAN | TransposeConv2dLayer | false | 5,530 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super().__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation =
# ... truncated (>4000 chars) for memory efficiency |
BetaVAE_H | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/dq/cdq2ys2ixsgbmto6z3gv6fap6g6tzfirrsvo2mvozuimf2d5feuy.py
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# x => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [2, 2], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_convolution_relu_0 = async_compile.triton('triton_poi_fused_convolution_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[131072],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 123008
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 961) % 32
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/m2/cm2tzhnnikquck5qjqxil3z7xbvb4bmlv4dxtpfnobmoa2gn7kpo.py
# Topologically Sorted Source Nodes: [conv2d_1, x_1], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_1 => convolution_1
# x_1 => relu_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu, %primals_4, %primals_5, [2, 2], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_1,), kwargs = {})
triton_poi_fused_convolution_relu_1 = async_compile.triton('triton_poi_fused_convolution_relu_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32768],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 25088
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 196) % 32
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/yl/cylhzvm5azwoxpfzq6cmn4wrmtveszsjfaehpp6d3sldzewuqoy4.py
# Topologically Sorted Source Nodes: [conv2d_2, x_2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_2 => convolution_2
# x_2 => relu_2
# Graph fragment:
# %convolution_2 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_1, %primals_6, %primals_7, [2, 2], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_2 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_2,), kwargs = {})
triton_poi_fused_convolution_relu_2 = async_compile.triton('triton_poi_fused_convolution_relu_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16384],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 9216
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 36) % 64
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/7x/c7x644ci2kcridm32huqyaq4krk2jeobv5q45sjg2zadi4i5xifb.py
# Topologically Sorted Source Nodes: [conv2d_3, x_3], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_3 => convolution_3
# x_3 => relu_3
# Graph fragment:
# %convolution_3 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_2, %primals_8, %primals_9, [2, 2], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_3 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_3,), kwargs = {})
# %le_4 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_3, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_3 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_3', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_3(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 4) % 64
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x3), tmp4, xmask)
tl.store(out_ptr0 + (x3), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5q/c5qlaxbbxxxofs4euqefep4bsfkrxrk3yorlgw53njxbvityagxo.py
# Topologically Sorted Source Nodes: [div, std, mul, z], Original ATen: [aten.div, aten.exp, aten.mul, aten.add]
# Source node to ATen node mapping:
# div => div
# mul => mul
# std => exp
# z => add
# Graph fragment:
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%slice_4, 2), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%div,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp, %normal_functional), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%slice_2, %mul), kwargs = {})
triton_poi_fused_add_div_exp_mul_4 = async_compile.triton('triton_poi_fused_add_div_exp_mul_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_exp_mul_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_div_exp_mul_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 40
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 10
x1 = (xindex // 10)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (20*x1)), xmask)
tmp1 = tl.load(in_ptr0 + (10 + x0 + (20*x1)), xmask)
tmp5 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tl_math.exp(tmp3)
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tl.store(out_ptr0 + (x2), tmp7, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/du/cduf6krob444andnckdtxb622t46y6p2ense7shwixlcd3vtk2mf.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_5 => relu_4
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_13), kwargs = {})
# %relu_4 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor,), kwargs = {})
# %le_3 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_4, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_5 = async_compile.triton('triton_poi_fused_relu_threshold_backward_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_5(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 256
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dk/cdkqzxyla22osetdivjgvawrv3ugrhoc4pphb4useu7rir2oxw7c.py
# Topologically Sorted Source Nodes: [x_10], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_10 => convolution_7
# Graph fragment:
# %convolution_7 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_7, %primals_20, %primals_21, [2, 2], [0, 0], [1, 1], True, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_6 = async_compile.triton('triton_poi_fused_convolution_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_6', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_6(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 49152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 4096) % 3
tmp0 = tl.load(in_out_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x3), tmp2, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21 = args
args.clear()
assert_size_stride(primals_1, (32, 3, 4, 4), (48, 16, 4, 1))
assert_size_stride(primals_2, (32, ), (1, ))
assert_size_stride(primals_3, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_4, (32, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_5, (32, ), (1, ))
assert_size_stride(primals_6, (64, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_7, (64, ), (1, ))
assert_size_stride(primals_8, (64, 64, 4, 4), (1024, 16, 4, 1))
assert_size_stride(primals_9, (64, ), (1, ))
assert_size_stride(primals_10, (20, 256), (256, 1))
assert_size_stride(primals_11, (20, ), (1, ))
assert_size_stride(primals_12, (256, 10), (10, 1))
assert_size_stride(primals_13, (256, ), (1, ))
assert_size_stride(primals_14, (64, 64, 4, 4), (1024, 16, 4, 1))
assert_size_stride(primals_15, (64, ), (1, ))
assert_size_stride(primals_16, (64, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_17, (32, ), (1, ))
assert_size_stride(primals_18, (32, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_19, (32, ), (1, ))
assert_size_stride(primals_20, (32, 3, 4, 4), (48, 16, 4, 1))
assert_size_stride(primals_21, (3, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 32, 31, 31), (30752, 961, 31, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 123008, grid=grid(123008), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 32, 14, 14), (6272, 196, 14, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, x_1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_1.run(buf3, primals_5, 25088, grid=grid(25088), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 6, 6), (2304, 36, 6, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [conv2d_2, x_2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_2.run(buf5, primals_7, 9216, grid=grid(9216), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 2, 2), (256, 4, 2, 1))
buf7 = buf6; del buf6 # reuse
buf24 = empty_strided_cuda((4, 64, 2, 2), (256, 4, 2, 1), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_3, x_3], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_3.run(buf7, primals_9, buf24, 1024, grid=grid(1024), stream=stream0)
del primals_9
buf8 = empty_strided_cuda((4, 20), (20, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf7, (4, 256), (256, 1), 0), reinterpret_tensor(primals_10, (256, 20), (1, 256), 0), alpha=1, beta=1, out=buf8)
del primals_11
buf9 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
# Topologically Sorted Source Nodes: [eps], Original ATen: [aten.normal_functional]
buf10 = torch.ops.aten.normal_functional.default(buf9)
buf11 = buf10
del buf10
buf12 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [div, std, mul, z], Original ATen: [aten.div, aten.exp, aten.mul, aten.add]
triton_poi_fused_add_div_exp_mul_4.run(buf8, buf11, buf12, 40, grid=grid(40), stream=stream0)
buf13 = empty_strided_cuda((4, 256), (256, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf12, reinterpret_tensor(primals_12, (10, 256), (1, 10), 0), out=buf13)
buf14 = buf13; del buf13 # reuse
buf23 = empty_strided_cuda((4, 256), (256, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_5.run(buf14, primals_13, buf23, 1024, grid=grid(1024), stream=stream0)
del primals_13
# Topologically Sorted Source Nodes: [conv_transpose2d], Original ATen: [aten.convolution]
buf15 = extern_kernels.convolution(reinterpret_tensor(buf14, (4, 64, 2, 2), (256, 4, 2, 1), 0), primals_14, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf15, (4, 64, 6, 6), (2304, 36, 6, 1))
buf16 = buf15; del buf15 # reuse
# Topologically Sorted Source Nodes: [conv_transpose2d, x_7], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_2.run(buf16, primals_15, 9216, grid=grid(9216), stream=stream0)
del primals_15
# Topologically Sorted Source Nodes: [conv_transpose2d_1], Original ATen: [aten.convolution]
buf17 = extern_kernels.convolution(buf16, primals_16, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf17, (4, 32, 14, 14), (6272, 196, 14, 1))
buf18 = buf17; del buf17 # reuse
# Topologically Sorted Source Nodes: [conv_transpose2d_1, x_8], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_1.run(buf18, primals_17, 25088, grid=grid(25088), stream=stream0)
del primals_17
# Topologically Sorted Source Nodes: [conv_transpose2d_2], Original ATen: [aten.convolution]
buf19 = extern_kernels.convolution(buf18, primals_18, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(1, 1), groups=1, bias=None)
assert_size_stride(buf19, (4, 32, 31, 31), (30752, 961, 31, 1))
buf20 = buf19; del buf19 # reuse
# Topologically Sorted Source Nodes: [conv_transpose2d_2, x_9], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf20, primals_19, 123008, grid=grid(123008), stream=stream0)
del primals_19
# Topologically Sorted Source Nodes: [x_10], Original ATen: [aten.convolution]
buf21 = extern_kernels.convolution(buf20, primals_20, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf21, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf22 = buf21; del buf21 # reuse
# Topologically Sorted Source Nodes: [x_10], Original ATen: [aten.convolution]
triton_poi_fused_convolution_6.run(buf22, primals_21, 49152, grid=grid(49152), stream=stream0)
del primals_21
return (buf22, reinterpret_tensor(buf8, (4, 10), (20, 1), 0), reinterpret_tensor(buf8, (4, 10), (20, 1), 10), buf8, primals_1, primals_3, primals_4, primals_6, primals_8, primals_14, primals_16, primals_18, primals_20, buf1, buf3, buf5, reinterpret_tensor(buf7, (4, 256), (256, 1), 0), reinterpret_tensor(buf8, (4, 10), (20, 1), 10), buf11, buf12, reinterpret_tensor(buf14, (4, 64, 2, 2), (256, 4, 2, 1), 0), buf16, buf18, buf20, buf23, primals_12, primals_10, buf24, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((32, 3, 4, 4), (48, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 3, 64, 64), (12288, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((32, 32, 4, 4), (512, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 32, 4, 4), (512, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((64, 64, 4, 4), (1024, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((20, 256), (256, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((20, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((256, 10), (10, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((64, 64, 4, 4), (1024, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((64, 32, 4, 4), (512, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_18 = rand_strided((32, 32, 4, 4), (512, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_19 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_20 = rand_strided((32, 3, 4, 4), (48, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_21 = rand_strided((3, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def reparametrize(mu, logvar):
std = logvar.div(2).exp()
eps = std.data.new(std.size()).normal_()
return mu + std * eps
class Encoder_H(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=1):
super(Encoder_H, self).__init__()
self.conv2d_1 = nn.Conv2d(nc, 32, 4, 2, padding)
self.conv2d_2 = nn.Conv2d(32, 32, 4, 2, padding)
self.conv2d_3 = nn.Conv2d(32, 64, 4, 2, padding)
self.conv2d_4 = nn.Conv2d(64, 64, 4, 2, padding)
self.flatten_shape, self.dconv_size = self._get_conv_output(input_shape
, nc)
self.linear = nn.Linear(self.flatten_shape, z_dim * 2)
def _get_conv_output(self, shape, nc):
bs = 1
dummy_x = torch.empty(bs, nc, *shape)
x, dconv_size = self._forward_features(dummy_x)
flatten_shape = x.flatten(1).size(1)
return flatten_shape, dconv_size
def _forward_features(self, x):
size0 = x.shape[1:]
x = torch.relu(self.conv2d_1(x))
size1 = x.shape[1:]
x = torch.relu(self.conv2d_2(x))
size2 = x.shape[1:]
x = torch.relu(self.conv2d_3(x))
size3 = x.shape[1:]
x = torch.relu(self.conv2d_4(x))
size4 = x.shape[1:]
return x, [size0, size1, size2, size3, size4]
def forward(self, x):
x = torch.relu(self.conv2d_1(x))
x = torch.relu(self.conv2d_2(x))
x = torch.relu(self.conv2d_3(x))
x = torch.relu(self.conv2d_4(x))
x = self.linear(x.flatten(1))
return x
class Decoder_H(nn.Module):
def __init__(self, output_shape, z_dim=10, nc=3, padding=1):
super(Decoder_H, self).__init__()
self.output_shape = output_shape
flatten_shape = output_shape[-1][0] * output_shape[-1][1
] * output_shape[-1][2]
self.linear = nn.Linear(z_dim, flatten_shape)
self.conv2d_1 = nn.ConvTranspose2d(64, 64, 4, 2, padding)
self.conv2d_2 = nn.ConvTranspose2d(64, 32, 4, 2, padding)
self.conv2d_3 = nn.ConvTranspose2d(32, 32, 4, 2, padding)
self.conv2d_4 = nn.ConvTranspose2d(32, nc, 4, 2, padding)
def _forward_features(self, x):
x = torch.relu(self.conv2d_1(x, self.output_shape[3][1:]))
x = torch.relu(self.conv2d_2(x, self.output_shape[2][1:]))
x = torch.relu(self.conv2d_3(x, self.output_shape[1][1:]))
x = self.conv2d_4(x, self.output_shape[0][1:])
return x
def forward(self, x):
x = torch.relu(self.linear(x))
x = x.view(-1, *self.output_shape[4])
x = self._forward_features(x)
return x
class BetaVAE_H(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=0,
activation=nn.Identity()):
super(BetaVAE_H, self).__init__()
self.z_dim = z_dim
self.activation = activation
self.encoder = Encoder_H(input_shape=input_shape, nc=nc, z_dim=
z_dim, padding=padding)
self.decoder = Decoder_H(self.encoder.dconv_size, nc=nc, z_dim=
z_dim, padding=padding)
def forward(self, x):
distributions = self.encoder(x)
mu = distributions[:, :self.z_dim]
logvar = distributions[:, self.z_dim:]
z = reparametrize(mu, logvar)
x_recon = self.decoder(z)
return self.activation(x_recon), mu, logvar
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 123008
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 961 % 32
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_1(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 25088
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 196 % 32
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 9216
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 36 % 64
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_3(in_out_ptr0,
in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 64
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x3, tmp4, xmask)
tl.store(out_ptr0 + x3, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_div_exp_mul_4(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 40
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 10
x1 = xindex // 10
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 20 * x1), xmask)
tmp1 = tl.load(in_ptr0 + (10 + x0 + 20 * x1), xmask)
tmp5 = tl.load(in_ptr1 + x2, xmask)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tl_math.exp(tmp3)
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tl.store(out_ptr0 + x2, tmp7, xmask)
@triton.jit
def triton_poi_fused_relu_threshold_backward_5(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 256
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_convolution_6(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 3
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17,
primals_18, primals_19, primals_20, primals_21) = args
args.clear()
assert_size_stride(primals_1, (32, 3, 4, 4), (48, 16, 4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_4, (32, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (64, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (64, 64, 4, 4), (1024, 16, 4, 1))
assert_size_stride(primals_9, (64,), (1,))
assert_size_stride(primals_10, (20, 256), (256, 1))
assert_size_stride(primals_11, (20,), (1,))
assert_size_stride(primals_12, (256, 10), (10, 1))
assert_size_stride(primals_13, (256,), (1,))
assert_size_stride(primals_14, (64, 64, 4, 4), (1024, 16, 4, 1))
assert_size_stride(primals_15, (64,), (1,))
assert_size_stride(primals_16, (64, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_17, (32,), (1,))
assert_size_stride(primals_18, (32, 32, 4, 4), (512, 16, 4, 1))
assert_size_stride(primals_19, (32,), (1,))
assert_size_stride(primals_20, (32, 3, 4, 4), (48, 16, 4, 1))
assert_size_stride(primals_21, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(2,
2), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 32, 31, 31), (30752, 961, 31, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(123008)](buf1, primals_2,
123008, XBLOCK=512, num_warps=8, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 32, 14, 14), (6272, 196, 14, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_1[grid(25088)](buf3, primals_5,
25088, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 6, 6), (2304, 36, 6, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_2[grid(9216)](buf5, primals_7,
9216, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 2, 2), (256, 4, 2, 1))
buf7 = buf6
del buf6
buf24 = empty_strided_cuda((4, 64, 2, 2), (256, 4, 2, 1), torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_3[grid(1024)](buf7
, primals_9, buf24, 1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((4, 20), (20, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf7, (4, 256),
(256, 1), 0), reinterpret_tensor(primals_10, (256, 20), (1, 256
), 0), alpha=1, beta=1, out=buf8)
del primals_11
buf9 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
buf10 = torch.ops.aten.normal_functional.default(buf9)
buf11 = buf10
del buf10
buf12 = buf9
del buf9
triton_poi_fused_add_div_exp_mul_4[grid(40)](buf8, buf11, buf12, 40,
XBLOCK=64, num_warps=1, num_stages=1)
buf13 = empty_strided_cuda((4, 256), (256, 1), torch.float32)
extern_kernels.mm(buf12, reinterpret_tensor(primals_12, (10, 256),
(1, 10), 0), out=buf13)
buf14 = buf13
del buf13
buf23 = empty_strided_cuda((4, 256), (256, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_5[grid(1024)](buf14,
primals_13, buf23, 1024, XBLOCK=256, num_warps=4, num_stages=1)
del primals_13
buf15 = extern_kernels.convolution(reinterpret_tensor(buf14, (4, 64,
2, 2), (256, 4, 2, 1), 0), primals_14, stride=(2, 2), padding=(
0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf15, (4, 64, 6, 6), (2304, 36, 6, 1))
buf16 = buf15
del buf15
triton_poi_fused_convolution_relu_2[grid(9216)](buf16, primals_15,
9216, XBLOCK=256, num_warps=4, num_stages=1)
del primals_15
buf17 = extern_kernels.convolution(buf16, primals_16, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf17, (4, 32, 14, 14), (6272, 196, 14, 1))
buf18 = buf17
del buf17
triton_poi_fused_convolution_relu_1[grid(25088)](buf18, primals_17,
25088, XBLOCK=256, num_warps=4, num_stages=1)
del primals_17
buf19 = extern_kernels.convolution(buf18, primals_18, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(1, 1), groups=1, bias=None)
assert_size_stride(buf19, (4, 32, 31, 31), (30752, 961, 31, 1))
buf20 = buf19
del buf19
triton_poi_fused_convolution_relu_0[grid(123008)](buf20, primals_19,
123008, XBLOCK=512, num_warps=8, num_stages=1)
del primals_19
buf21 = extern_kernels.convolution(buf20, primals_20, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf21, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf22 = buf21
del buf21
triton_poi_fused_convolution_6[grid(49152)](buf22, primals_21,
49152, XBLOCK=256, num_warps=4, num_stages=1)
del primals_21
return (buf22, reinterpret_tensor(buf8, (4, 10), (20, 1), 0),
reinterpret_tensor(buf8, (4, 10), (20, 1), 10), buf8, primals_1,
primals_3, primals_4, primals_6, primals_8, primals_14, primals_16,
primals_18, primals_20, buf1, buf3, buf5, reinterpret_tensor(buf7,
(4, 256), (256, 1), 0), reinterpret_tensor(buf8, (4, 10), (20, 1),
10), buf11, buf12, reinterpret_tensor(buf14, (4, 64, 2, 2), (256, 4,
2, 1), 0), buf16, buf18, buf20, buf23, primals_12, primals_10, buf24)
def reparametrize(mu, logvar):
std = logvar.div(2).exp()
eps = std.data.new(std.size()).normal_()
return mu + std * eps
class Encoder_H(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=1):
super(Encoder_H, self).__init__()
self.conv2d_1 = nn.Conv2d(nc, 32, 4, 2, padding)
self.conv2d_2 = nn.Conv2d(32, 32, 4, 2, padding)
self.conv2d_3 = nn.Conv2d(32, 64, 4, 2, padding)
self.conv2d_4 = nn.Conv2d(64, 64, 4, 2, padding)
self.flatten_shape, self.dconv_size = self._get_conv_output(input_shape
, nc)
self.linear = nn.Linear(self.flatten_shape, z_dim * 2)
def _get_conv_output(self, shape, nc):
bs = 1
dummy_x = torch.empty(bs, nc, *shape)
x, dconv_size = self._forward_features(dummy_x)
flatten_shape = x.flatten(1).size(1)
return flatten_shape, dconv_size
def _forward_features(self, x):
size0 = x.shape[1:]
x = torch.relu(self.conv2d_1(x))
size1 = x.shape[1:]
x = torch.relu(self.conv2d_2(x))
size2 = x.shape[1:]
x = torch.relu(self.conv2d_3(x))
size3 = x.shape[1:]
x = torch.relu(self.conv2d_4(x))
size4 = x.shape[1:]
return x, [size0, size1, size2, size3, size4]
def forward(self, x):
x = torch.relu(self.conv2d_1(x))
x = torch.relu(self.conv2d_2(x))
x = torch.relu(self.conv2d_3(x))
x = torch.relu(self.conv2d_4(x))
x = self.linear(x.flatten(1))
return x
class Decoder_H(nn.Module):
def __init__(self, output_shape, z_dim=10, nc=3, padding=1):
super(Decoder_H, self).__init__()
self.output_shape = output_shape
flatten_shape = output_shape[-1][0] * output_shape[-1][1
] * output_shape[-1][2]
self.linear = nn.Linear(z_dim, flatten_shape)
self.conv2d_1 = nn.ConvTranspose2d(64, 64, 4, 2, padding)
self.conv2d_2 = nn.ConvTranspose2d(64, 32, 4, 2, padding)
self.conv2d_3 = nn.ConvTranspose2d(32, 32, 4, 2, padding)
self.conv2d_4 = nn.ConvTranspose2d(32, nc, 4, 2, padding)
def _forward_features(self, x):
x = torch.relu(self.conv2d_1(x, self.output_shape[3][1:]))
x = torch.relu(self.conv2d_2(x, self.output_shape[2][1:]))
x = torch.relu(self.conv2d_3(x, self.output_shape[1][1:]))
x = self.conv2d_4(x, self.output_shape[0][1:])
return x
def forward(self, x):
x = torch.relu(self.linear(x))
x = x.view(-1, *self.output_shape[4])
x = self._forward_features(x)
return x
class BetaVAE_HNew(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=0,
activation=nn.Identity()):
super(BetaVAE_HNew, self).__init__()
self.z_dim = z_dim
self.activation = activation
self.encoder = Encoder_H(input_shape=input_shape, nc=nc, z_dim=
z_dim, padding=padding)
self.decoder = Decoder_H(self.encoder.dconv_size, nc=nc, z_dim=
z_dim, padding=padding)
def forward(self, input_0):
primals_1 = self.encoder.conv2d_1.weight
primals_2 = self.encoder.conv2d_1.bias
primals_4 = self.encoder.conv2d_2.weight
primals_5 = self.encoder.conv2d_2.bias
primals_6 = self.encoder.conv2d_3.weight
primals_7 = self.encoder.conv2d_3.bias
primals_8 = self.encoder.conv2d_4.weight
primals_9 = self.encoder.conv2d_4.bias
primals_10 = self.encoder.linear.weight
primals_11 = self.encoder.linear.bias
primals_12 = self.decoder.linear.weight
primals_13 = self.decoder.linear.bias
primals_14 = self.decoder.conv2d_1.weight
primals_15 = self.decoder.conv2d_1.bias
primals_16 = self.decoder.conv2d_2.weight
primals_17 = self.decoder.conv2d_2.bias
primals_18 = self.decoder.conv2d_3.weight
primals_19 = self.decoder.conv2d_3.bias
primals_20 = self.decoder.conv2d_4.weight
primals_21 = self.decoder.conv2d_4.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21])
return output[0], output[1], output[2]
| KinWaiCheuk/Beta-VAE | BetaVAE_H | false | 5,531 | [
"MIT"
] | 1 | 57f538320fed76b54e8489656b11dc83c06d1584 | https://github.com/KinWaiCheuk/Beta-VAE/tree/57f538320fed76b54e8489656b11dc83c06d1584 | import torch
import torch.nn as nn
def reparametrize(mu, logvar):
std = logvar.div(2).exp()
eps = std.data.new(std.size()).normal_()
return mu + std * eps
class Encoder_H(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=1):
super().__init__()
self.conv2d_1 = nn.Conv2d(nc, 32, 4, 2, padding)
self.conv2d_2 = nn.Conv2d(32, 32, 4, 2, padding)
self.conv2d_3 = nn.Conv2d(32, 64, 4, 2, padding)
self.conv2d_4 = nn.Conv2d(64, 64, 4, 2, padding)
self.flatten_shape, self.dconv_size = self._get_conv_output(input_shape
, nc)
self.linear = nn.Linear(self.flatten_shape, z_dim * 2)
def _get_conv_output(self, shape, nc):
bs = 1
dummy_x = torch.empty(bs, nc, *shape)
x, dconv_size = self._forward_features(dummy_x)
flatten_shape = x.flatten(1).size(1)
return flatten_shape, dconv_size
def _forward_features(self, x):
size0 = x.shape[1:]
x = torch.relu(self.conv2d_1(x))
size1 = x.shape[1:]
x = torch.relu(self.conv2d_2(x))
size2 = x.shape[1:]
x = torch.relu(self.conv2d_3(x))
size3 = x.shape[1:]
x = torch.relu(self.conv2d_4(x))
size4 = x.shape[1:]
return x, [size0, size1, size2, size3, size4]
def forward(self, x):
x = torch.relu(self.conv2d_1(x))
x = torch.relu(self.conv2d_2(x))
x = torch.relu(self.conv2d_3(x))
x = torch.relu(self.conv2d_4(x))
x = self.linear(x.flatten(1))
return x
class Decoder_H(nn.Module):
def __init__(self, output_shape, z_dim=10, nc=3, padding=1):
super().__init__()
self.output_shape = output_shape
flatten_shape = output_shape[-1][0] * output_shape[-1][1
] * output_shape[-1][2]
self.linear = nn.Linear(z_dim, flatten_shape)
self.conv2d_1 = nn.ConvTranspose2d(64, 64, 4, 2, padding)
self.conv2d_2 = nn.ConvTranspose2d(64, 32, 4, 2, padding)
self.conv2d_3 = nn.ConvTranspose2d(32, 32, 4, 2, padding)
self.conv2d_4 = nn.ConvTranspose2d(32, nc, 4, 2, padding)
def _forward_features(self, x):
x = torch.relu(self.conv2d_1(x, self.output_shape[3][1:]))
x = torch.relu(self.conv2d_2(x, self.output_shape[2][1:]))
x = torch.relu(self.conv2d_3(x, self.output_shape[1][1:]))
x = self.conv2d_4(x, self.output_shape[0][1:])
return x
def forward(self, x):
x = torch.relu(self.linear(x))
x = x.view(-1, *self.output_shape[4])
x = self._forward_features(x)
return x
class Model(nn.Module):
def __init__(self, input_shape=(64, 64), z_dim=10, nc=3, padding=0,
activation=nn.Identity()):
super().__init__()
self.z_dim = z_dim
self.activation = activation
self.encoder = Encoder_H(input_shape=input_shape, nc=nc, z_dim=
z_dim, padding=padding)
self.decoder = Decoder_H(self.encoder.dconv_size, nc=nc, z_dim=
z_dim, padding=padding)
def forward(self, x):
distributions = self.encoder(x)
mu = distributions[:, :self.z_dim]
logvar = distributions[:, self.z_dim:]
z = reparametrize(mu, logvar)
x_recon = self.decoder(z)
return self.activation(x_recon), mu, logvar
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return []
|
MultiHeadAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/u5/cu56dhpcth43gy4shrd7mcexf4nfa6qetnnhwe4mno4v6ug76h6j.py
# Topologically Sorted Source Nodes: [att_out_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# att_out_1 => clone_3
# Graph fragment:
# %clone_3 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_7,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_0 = async_compile.triton('triton_poi_fused_clone_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tl.store(out_ptr0 + (x0), tmp0, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (16, 4), (4, 1))
assert_size_stride(primals_5, (16, ), (1, ))
assert_size_stride(primals_6, (16, 4), (4, 1))
assert_size_stride(primals_7, (16, ), (1, ))
assert_size_stride(primals_8, (16, 4), (4, 1))
assert_size_stride(primals_9, (16, ), (1, ))
assert_size_stride(primals_10, (4, 16), (16, 1))
assert_size_stride(primals_11, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 16), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_4
del primals_5
buf1 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(primals_2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 16), (1, 4), 0), alpha=1, beta=1, out=buf1)
del primals_6
del primals_7
buf2 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_9, reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_8, (4, 16), (1, 4), 0), alpha=1, beta=1, out=buf2)
del primals_8
del primals_9
# Topologically Sorted Source Nodes: [], Original ATen: []
buf3 = torch.ops.aten._scaled_dot_product_efficient_attention.default(reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 4, 16, 1), 0), reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 4, 16, 1), 0), reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 4, 16, 1), 0), None, True, scale=0.5)
buf4 = buf3[0]
buf5 = buf3[1]
buf6 = buf3[2]
buf7 = buf3[3]
del buf3
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [att_out_1], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(buf4, buf8, 256, grid=grid(256), stream=stream0)
buf9 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf8, (16, 16), (16, 1), 0), reinterpret_tensor(primals_10, (16, 4), (1, 16), 0), alpha=1, beta=1, out=buf9)
del primals_11
return (reinterpret_tensor(buf9, (4, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 4, 16, 1), 0), reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 4, 16, 1), 0), reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 4, 16, 1), 0), buf4, buf5, buf6, buf7, reinterpret_tensor(buf8, (16, 16), (16, 1), 0), primals_10, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((16, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((16, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((16, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 16), (16, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
class SelfAttention(nn.Module):
def __init__(self, dropout=0.1):
super(SelfAttention, self).__init__()
self.softmax = nn.Softmax(dim=-1)
self._dropout = nn.Dropout(dropout)
def forward(self, q, k, v, pad_mask=None):
"""
:param q: [bz, len_q, Q]
:param k: [bz, len_k, K]
:param v: [bz, len_v, V]
:param pad_mask: [bz, len_q, len_k] 填充部分的mask
more: Q==K, len_k==len_v
:return: [bz, len_q, V]
"""
att_weights = torch.matmul(q, k.transpose(-1, -2)).div(math.sqrt(k.
size(-1)))
if pad_mask is not None:
att_weights.masked_fill_(pad_mask, -1000000000.0)
soft_att_weights = self.softmax(att_weights)
if self.training:
soft_att_weights = self._dropout(soft_att_weights)
att_out = torch.matmul(soft_att_weights, v)
return att_out
class MultiHeadAttention(nn.Module):
def __init__(self, d_model, d_k, d_v, nb_heads, dropout=0.1):
super(MultiHeadAttention, self).__init__()
self._d_model = d_model
self._d_k = d_k
self._d_v = d_v
self._nb_heads = nb_heads
self._linear_qs = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_ks = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_vs = nn.Linear(in_features=d_model, out_features=d_v *
nb_heads)
self._linear_out = nn.Linear(in_features=d_v * nb_heads,
out_features=d_model)
self._self_attention = SelfAttention(dropout)
self._dropout = nn.Dropout(dropout)
self.reset_parameters()
def reset_parameters(self):
nn.init.normal_(self._linear_qs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_ks.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_vs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_out.weight, mean=0, std=math.sqrt(1 /
self._d_model))
def forward(self, q, k, v, att_mask=None):
"""
:param q: [bz, len_q, d_model]
:param k: [bz, len_k, d_model]
:param v: [bz, len_v, d_model]
:param att_mask: [bz, len_k]
more: Q == K, len_k==len_v
:return: [bz, len_q, d_model]
"""
bz, len_q, _ = q.size()
bz, len_k, _ = k.size()
bz, len_v, _ = v.size()
q_fc = self._linear_qs(q).reshape(bz, len_q, self._nb_heads, -1
).transpose(1, 2)
k_fc = self._linear_ks(k).reshape(bz, len_k, self._nb_heads, -1
).transpose(1, 2)
v_fc = self._linear_vs(v).reshape(bz, len_v, self._nb_heads, -1
).transpose(1, 2)
if att_mask is not None:
att_mask = att_mask[:, None, None, :]
att_out = self._self_attention(q_fc, k_fc, v_fc, att_mask)
att_out = att_out.transpose(1, 2).reshape(bz, len_q, -1)
multi_head = self._linear_out(att_out)
if self.training:
multi_head = self._dropout(multi_head)
return multi_head
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [[], {'d_model': 4, 'd_k': 4, 'd_v': 4, 'nb_heads': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tl.store(out_ptr0 + x0, tmp0, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (16, 4), (4, 1))
assert_size_stride(primals_5, (16,), (1,))
assert_size_stride(primals_6, (16, 4), (4, 1))
assert_size_stride(primals_7, (16,), (1,))
assert_size_stride(primals_8, (16, 4), (4, 1))
assert_size_stride(primals_9, (16,), (1,))
assert_size_stride(primals_10, (4, 16), (16, 1))
assert_size_stride(primals_11, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_1, (16,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 16), (1, 4),
0), alpha=1, beta=1, out=buf0)
del primals_4
del primals_5
buf1 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(primals_2, (16,
4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 16), (1, 4),
0), alpha=1, beta=1, out=buf1)
del primals_6
del primals_7
buf2 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.addmm(primals_9, reinterpret_tensor(primals_3, (16,
4), (4, 1), 0), reinterpret_tensor(primals_8, (4, 16), (1, 4),
0), alpha=1, beta=1, out=buf2)
del primals_8
del primals_9
buf3 = torch.ops.aten._scaled_dot_product_efficient_attention.default(
reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 4, 16, 1), 0),
reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 4, 16, 1), 0),
reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 4, 16, 1), 0), None,
True, scale=0.5)
buf4 = buf3[0]
buf5 = buf3[1]
buf6 = buf3[2]
buf7 = buf3[3]
del buf3
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(256)](buf4, buf8, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf9 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf8, (16, 16),
(16, 1), 0), reinterpret_tensor(primals_10, (16, 4), (1, 16), 0
), alpha=1, beta=1, out=buf9)
del primals_11
return reinterpret_tensor(buf9, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_2, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0
), reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 4, 16, 1), 0
), reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 4, 16, 1), 0
), reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 4, 16, 1), 0
), buf4, buf5, buf6, buf7, reinterpret_tensor(buf8, (16, 16), (16,
1), 0), primals_10
class SelfAttention(nn.Module):
def __init__(self, dropout=0.1):
super(SelfAttention, self).__init__()
self.softmax = nn.Softmax(dim=-1)
self._dropout = nn.Dropout(dropout)
def forward(self, q, k, v, pad_mask=None):
"""
:param q: [bz, len_q, Q]
:param k: [bz, len_k, K]
:param v: [bz, len_v, V]
:param pad_mask: [bz, len_q, len_k] 填充部分的mask
more: Q==K, len_k==len_v
:return: [bz, len_q, V]
"""
att_weights = torch.matmul(q, k.transpose(-1, -2)).div(math.sqrt(k.
size(-1)))
if pad_mask is not None:
att_weights.masked_fill_(pad_mask, -1000000000.0)
soft_att_weights = self.softmax(att_weights)
if self.training:
soft_att_weights = self._dropout(soft_att_weights)
att_out = torch.matmul(soft_att_weights, v)
return att_out
class MultiHeadAttentionNew(nn.Module):
def __init__(self, d_model, d_k, d_v, nb_heads, dropout=0.1):
super(MultiHeadAttentionNew, self).__init__()
self._d_model = d_model
self._d_k = d_k
self._d_v = d_v
self._nb_heads = nb_heads
self._linear_qs = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_ks = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_vs = nn.Linear(in_features=d_model, out_features=d_v *
nb_heads)
self._linear_out = nn.Linear(in_features=d_v * nb_heads,
out_features=d_model)
self._self_attention = SelfAttention(dropout)
self._dropout = nn.Dropout(dropout)
self.reset_parameters()
def reset_parameters(self):
nn.init.normal_(self._linear_qs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_ks.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_vs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_out.weight, mean=0, std=math.sqrt(1 /
self._d_model))
def forward(self, input_0, input_1, input_2):
primals_4 = self._linear_qs.weight
primals_5 = self._linear_qs.bias
primals_6 = self._linear_ks.weight
primals_7 = self._linear_ks.bias
primals_8 = self._linear_vs.weight
primals_9 = self._linear_vs.bias
primals_10 = self._linear_out.weight
primals_11 = self._linear_out.bias
primals_1 = input_0
primals_2 = input_1
primals_3 = input_2
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | MultiHeadAttention | false | 5,532 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import math
import torch
import torch.nn as nn
class SelfAttention(nn.Module):
def __init__(self, dropout=0.1):
super().__init__()
self.softmax = nn.Softmax(dim=-1)
self._dropout = nn.Dropout(dropout)
def forward(self, q, k, v, pad_mask=None):
"""
:param q: [bz, len_q, Q]
:param k: [bz, len_k, K]
:param v: [bz, len_v, V]
:param pad_mask: [bz, len_q, len_k] 填充部分的mask
more: Q==K, len_k==len_v
:return: [bz, len_q, V]
"""
att_weights = torch.matmul(q, k.transpose(-1, -2)).div(math.sqrt(k.
size(-1)))
if pad_mask is not None:
att_weights.masked_fill_(pad_mask, -1000000000.0)
soft_att_weights = self.softmax(att_weights)
if self.training:
soft_att_weights = self._dropout(soft_att_weights)
att_out = torch.matmul(soft_att_weights, v)
return att_out
class Model(nn.Module):
def __init__(self, d_model, d_k, d_v, nb_heads, dropout=0.1):
super().__init__()
self._d_model = d_model
self._d_k = d_k
self._d_v = d_v
self._nb_heads = nb_heads
self._linear_qs = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_ks = nn.Linear(in_features=d_model, out_features=d_k *
nb_heads)
self._linear_vs = nn.Linear(in_features=d_model, out_features=d_v *
nb_heads)
self._linear_out = nn.Linear(in_features=d_v * nb_heads,
out_features=d_model)
self._self_attention = SelfAttention(dropout)
self._dropout = nn.Dropout(dropout)
self.reset_parameters()
def reset_parameters(self):
nn.init.normal_(self._linear_qs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_ks.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_vs.weight, mean=0, std=math.sqrt(1 /
self._d_model))
nn.init.normal_(self._linear_out.weight, mean=0, std=math.sqrt(1 /
self._d_model))
def forward(self, q, k, v, att_mask=None):
"""
:param q: [bz, len_q, d_model]
:param k: [bz, len_k, d_model]
:param v: [bz, len_v, d_model]
:param att_mask: [bz, len_k]
more: Q == K, len_k==len_v
:return: [bz, len_q, d_model]
"""
bz, len_q, _ = q.size()
bz, len_k, _ = k.size()
bz, len_v, _ = v.size()
q_fc = self._linear_qs(q).reshape(bz, len_q, self._nb_heads, -1
).transpose(1, 2)
k_fc = self._linear_ks(k).reshape(bz, len_k, self._nb_heads, -1
).transpose(1, 2)
v_fc = self._linear_vs(v).reshape(bz, len_v, self._nb_heads, -1
).transpose(1, 2)
if att_mask is not None:
att_mask = att_mask[:, None, None, :]
att_out = self._self_attention(q_fc, k_fc, v_fc, att_mask)
att_out = att_out.transpose(1, 2).reshape(bz, len_q, -1)
multi_head = self._linear_out(att_out)
if self.training:
multi_head = self._dropout(multi_head)
return multi_head
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [4, 4, 4, 4]
|
AttnMerge | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/qz/cqza6p5fjiie2hfiu5dfjqqugrnzziwuwxzlhzy2aa7khopxjbym.py
# Topologically Sorted Source Nodes: [alpha], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# alpha => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_1, [1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_1, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x3), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ry/cry3o7ciiisvaqxldoaatx6nejwjg2z36x5a6gra4n32ey4vmhsr.py
# Topologically Sorted Source Nodes: [alpha, mul, out], Original ATen: [aten._softmax, aten.mul, aten.sum]
# Source node to ATen node mapping:
# alpha => div, sum_1
# mul => mul
# out => sum_2
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %primals_2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
triton_poi_fused__softmax_mul_sum_1 = async_compile.triton('triton_poi_fused__softmax_mul_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_mul_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_mul_sum_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = (xindex // 16)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (64*x1)), xmask)
tmp1 = tl.load(in_ptr0 + (16 + x0 + (64*x1)), xmask)
tmp3 = tl.load(in_ptr0 + (32 + x0 + (64*x1)), xmask)
tmp5 = tl.load(in_ptr0 + (48 + x0 + (64*x1)), xmask)
tmp8 = tl.load(in_ptr1 + (x0 + (64*x1)), xmask)
tmp11 = tl.load(in_ptr1 + (16 + x0 + (64*x1)), xmask)
tmp15 = tl.load(in_ptr1 + (32 + x0 + (64*x1)), xmask)
tmp19 = tl.load(in_ptr1 + (48 + x0 + (64*x1)), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = tmp0 / tmp6
tmp9 = tmp7 * tmp8
tmp10 = tmp1 / tmp6
tmp12 = tmp10 * tmp11
tmp13 = tmp9 + tmp12
tmp14 = tmp3 / tmp6
tmp16 = tmp14 * tmp15
tmp17 = tmp13 + tmp16
tmp18 = tmp5 / tmp6
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tl.store(out_ptr0 + (x2), tmp21, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [hx], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_2, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [alpha], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [alpha, mul, out], Original ATen: [aten._softmax, aten.mul, aten.sum]
triton_poi_fused__softmax_mul_sum_1.run(buf1, primals_2, buf2, 64, grid=grid(64), stream=stream0)
del buf1
return (buf2, primals_2, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class AttnMerge(nn.Module):
def __init__(self, hn_size):
super(AttnMerge, self).__init__()
self.fc = nn.Linear(hn_size, hn_size, bias=False)
def forward(self, x):
hx = self.fc(x)
alpha = F.softmax(hx, dim=1)
out = torch.sum(alpha * x, dim=1)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'hn_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x3, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_mul_sum_1(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = xindex // 16
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 64 * x1), xmask)
tmp1 = tl.load(in_ptr0 + (16 + x0 + 64 * x1), xmask)
tmp3 = tl.load(in_ptr0 + (32 + x0 + 64 * x1), xmask)
tmp5 = tl.load(in_ptr0 + (48 + x0 + 64 * x1), xmask)
tmp8 = tl.load(in_ptr1 + (x0 + 64 * x1), xmask)
tmp11 = tl.load(in_ptr1 + (16 + x0 + 64 * x1), xmask)
tmp15 = tl.load(in_ptr1 + (32 + x0 + 64 * x1), xmask)
tmp19 = tl.load(in_ptr1 + (48 + x0 + 64 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = tmp0 / tmp6
tmp9 = tmp7 * tmp8
tmp10 = tmp1 / tmp6
tmp12 = tmp10 * tmp11
tmp13 = tmp9 + tmp12
tmp14 = tmp3 / tmp6
tmp16 = tmp14 * tmp15
tmp17 = tmp13 + tmp16
tmp18 = tmp5 / tmp6
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tl.store(out_ptr0 + x2, tmp21, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_2, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_mul_sum_1[grid(64)](buf1, primals_2, buf2,
64, XBLOCK=64, num_warps=1, num_stages=1)
del buf1
return buf2, primals_2, buf0
class AttnMergeNew(nn.Module):
def __init__(self, hn_size):
super(AttnMergeNew, self).__init__()
self.fc = nn.Linear(hn_size, hn_size, bias=False)
def forward(self, input_0):
primals_1 = self.fc.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | AttnMerge | false | 5,533 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, hn_size):
super().__init__()
self.fc = nn.Linear(hn_size, hn_size, bias=False)
def forward(self, x):
hx = self.fc(x)
alpha = F.softmax(hx, dim=1)
out = torch.sum(alpha * x, dim=1)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
Highway | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/oy/coygqjpoh5r5lolh6b5eagjttwpgk4w2ou6uwxyoprhw5g2ddsgw.py
# Topologically Sorted Source Nodes: [gate, mul, sub, mul_1, x], Original ATen: [aten.sigmoid, aten.mul, aten.rsub, aten.add]
# Source node to ATen node mapping:
# gate => sigmoid
# mul => mul
# mul_1 => mul_1
# sub => sub
# x => add
# Graph fragment:
# %sigmoid : [num_users=2] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_1,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid, %primals_4), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sigmoid), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %primals_3), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %mul_1), kwargs = {})
triton_poi_fused_add_mul_rsub_sigmoid_0 = async_compile.triton('triton_poi_fused_add_mul_rsub_sigmoid_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_rsub_sigmoid_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_rsub_sigmoid_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp2 = tl.load(in_ptr1 + (x0), xmask)
tmp6 = tl.load(in_ptr2 + (x0), xmask)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = 1.0
tmp5 = tmp4 - tmp1
tmp7 = tmp5 * tmp6
tmp8 = tmp3 + tmp7
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [gate, mul, sub, mul_1, x], Original ATen: [aten.sigmoid, aten.mul, aten.rsub, aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mul_rsub_sigmoid_0.run(buf0, primals_4, primals_3, buf1, 256, grid=grid(256), stream=stream0)
return (buf1, primals_3, primals_4, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Highway(nn.Module):
def __init__(self, x_hidden):
super(Highway, self).__init__()
self.lin = nn.Linear(x_hidden, x_hidden)
def forward(self, x1, x2):
gate = torch.sigmoid(self.lin(x1))
x = torch.mul(gate, x2) + torch.mul(1 - gate, x1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'x_hidden': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_rsub_sigmoid_0(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp2 = tl.load(in_ptr1 + x0, xmask)
tmp6 = tl.load(in_ptr2 + x0, xmask)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = 1.0
tmp5 = tmp4 - tmp1
tmp7 = tmp5 * tmp6
tmp8 = tmp3 + tmp7
tl.store(out_ptr0 + x0, tmp8, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_rsub_sigmoid_0[grid(256)](buf0, primals_4,
primals_3, buf1, 256, XBLOCK=256, num_warps=4, num_stages=1)
return buf1, primals_3, primals_4, buf0
class HighwayNew(nn.Module):
def __init__(self, x_hidden):
super(HighwayNew, self).__init__()
self.lin = nn.Linear(x_hidden, x_hidden)
def forward(self, input_0, input_1):
primals_1 = self.lin.weight
primals_2 = self.lin.bias
primals_3 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| LinXueyuanStdio/EchoEA | Highway | false | 5,534 | [
"Apache-2.0"
] | 1 | d9b8564023cca71678dec44cf8cab3f91736448a | https://github.com/LinXueyuanStdio/EchoEA/tree/d9b8564023cca71678dec44cf8cab3f91736448a | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, x_hidden):
super().__init__()
self.lin = nn.Linear(x_hidden, x_hidden)
def forward(self, x1, x2):
gate = torch.sigmoid(self.lin(x1))
x = torch.mul(gate, x2) + torch.mul(1 - gate, x1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
ChannelPool | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/uc/cucdaa5tqnxykdmw5yqh7ir5ac35phopjcobljrg4rrtlnfjtuwd.py
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%unsqueeze, %unsqueeze_1], 1), kwargs = {})
triton_poi_fused_cat_0 = async_compile.triton('triton_poi_fused_cat_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 16) % 2
x0 = xindex % 16
x2 = (xindex // 32)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp9 = triton_helpers.maximum(tmp7, tmp8)
tmp10 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp11 = triton_helpers.maximum(tmp9, tmp10)
tmp12 = tl.full(tmp11.shape, 0.0, tmp11.dtype)
tmp13 = tl.where(tmp4, tmp11, tmp12)
tmp14 = tmp0 >= tmp3
tmp15 = tl.full([1], 2, tl.int64)
tmp16 = tmp0 < tmp15
tmp17 = tl.load(in_ptr0 + (x0 + (64*x2)), tmp14 & xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), tmp14 & xmask, eviction_policy='evict_last', other=0.0)
tmp19 = tmp17 + tmp18
tmp20 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), tmp14 & xmask, eviction_policy='evict_last', other=0.0)
tmp21 = tmp19 + tmp20
tmp22 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), tmp14 & xmask, eviction_policy='evict_last', other=0.0)
tmp23 = tmp21 + tmp22
tmp24 = 4.0
tmp25 = tmp23 / tmp24
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp14, tmp25, tmp26)
tmp28 = tl.where(tmp4, tmp13, tmp27)
tl.store(out_ptr0 + (x3), tmp28, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(arg0_1, buf0, 128, grid=grid(128), stream=stream0)
del arg0_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class ChannelPool(nn.Module):
def forward(self, x):
return torch.cat((torch.max(x, 1)[0].unsqueeze(1), torch.mean(x, 1)
.unsqueeze(1)), dim=1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 2
x0 = xindex % 16
x2 = xindex // 32
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 64 * x2), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp9 = triton_helpers.maximum(tmp7, tmp8)
tmp10 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = triton_helpers.maximum(tmp9, tmp10)
tmp12 = tl.full(tmp11.shape, 0.0, tmp11.dtype)
tmp13 = tl.where(tmp4, tmp11, tmp12)
tmp14 = tmp0 >= tmp3
tl.full([1], 2, tl.int64)
tmp17 = tl.load(in_ptr0 + (x0 + 64 * x2), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp18 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp19 = tmp17 + tmp18
tmp20 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp21 = tmp19 + tmp20
tmp22 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp23 = tmp21 + tmp22
tmp24 = 4.0
tmp25 = tmp23 / tmp24
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp14, tmp25, tmp26)
tmp28 = tl.where(tmp4, tmp13, tmp27)
tl.store(out_ptr0 + x3, tmp28, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(128)](arg0_1, buf0, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class ChannelPoolNew(nn.Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| Linus4world/mrs-gan | ChannelPool | false | 5,535 | [
"BSD-2-Clause"
] | 1 | 64669251584a7421cce3a5173983a2275dcb438a | https://github.com/Linus4world/mrs-gan/tree/64669251584a7421cce3a5173983a2275dcb438a | import torch
import torch.nn as nn
class Model(nn.Module):
def forward(self, x):
return torch.cat((torch.max(x, 1)[0].unsqueeze(1), torch.mean(x, 1)
.unsqueeze(1)), dim=1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
ResidualDenseBlock_3C | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/td/ctdv3m5a33kovvtng5iilth4k6mtnyfcota6hhwoiqm34iumu7wi.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x => constant_pad_nd
# Graph fragment:
# %constant_pad_nd : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%primals_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/er/cerxfmiafmpnsn44ihnsss2pvd76g6x2i62qavnas5ygqo5mj43d.py
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat => cat
# x_3 => constant_pad_nd_1
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where], 1), kwargs = {})
# %constant_pad_nd_1 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_1 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 8
x3 = (xindex // 288)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + (x6), tmp30, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s3/cs34ah7ioinuak53u3oyhgsrsqeqfnovsb4djwnzfgkw44v6u547.py
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
# Source node to ATen node mapping:
# cat_1 => cat_1
# x_6 => constant_pad_nd_2
# Graph fragment:
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %where, %where_1], 1), kwargs = {})
# %constant_pad_nd_2 : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%cat_1, [1, 1, 1, 1], 0.0), kwargs = {})
triton_poi_fused_cat_constant_pad_nd_2 = async_compile.triton('triton_poi_fused_cat_constant_pad_nd_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_constant_pad_nd_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 6
x0 = xindex % 6
x2 = (xindex // 36) % 12
x3 = (xindex // 432)
x6 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = (-1) + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp12 = tmp11 >= tmp1
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2) + (64*x3)), tmp14 & xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + ((-5) + x0 + (4*x1) + (16*((-4) + x2)) + (64*x3)), tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tmp30 = tl.full([1], 12, tl.int64)
tmp31 = tmp11 < tmp30
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + ((-5) + x0 + (4*x1) + (16*((-8) + x2)) + (64*x3)), tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + (x6), tmp42, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ya/cyaqnnpliwjolqezg24obzz6ccba4jeqnoqwry7pv7itm4cqy6ao.py
# Topologically Sorted Source Nodes: [x_8, mul, x3], Original ATen: [aten.leaky_relu, aten.mul, aten.add, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# mul => mul_3
# x3 => add
# x_8 => gt_2, mul_2, where_2
# Graph fragment:
# %gt_2 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_2, 0), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_2, 0.2), kwargs = {})
# %where_2 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_2, %convolution_2, %mul_2), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%where_2, 0.1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_3, %primals_1), kwargs = {})
# %gt_3 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_2, 0), kwargs = {})
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3 = async_compile.triton('triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i1', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp8 = tl.load(in_ptr1 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = 0.1
tmp7 = tmp5 * tmp6
tmp9 = tmp7 + tmp8
tmp10 = tmp5 > tmp1
tl.store(out_ptr0 + (x0), tmp9, xmask)
tl.store(out_ptr1 + (x0), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/my/cmymu55pgdw2f6aywkz53vy2ux4ld5ava4e5srxurzv3xcxqknfg.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x_5 => gt_1, mul_1, where_1
# Graph fragment:
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_1, 0), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_1, 0.2), kwargs = {})
# %where_1 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt_1, %convolution_1, %mul_1), kwargs = {})
# %gt_4 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where_1, 0), kwargs = {})
triton_poi_fused_leaky_relu_leaky_relu_backward_4 = async_compile.triton('triton_poi_fused_leaky_relu_leaky_relu_backward_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_leaky_relu_backward_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_4(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(primals_1, buf0, 576, grid=grid(576), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat, x_3], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_1.run(primals_1, buf1, buf2, 1152, grid=grid(1152), stream=stream0)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.convolution]
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat_1, x_6], Original ATen: [aten.cat, aten.constant_pad_nd]
triton_poi_fused_cat_constant_pad_nd_2.run(primals_1, buf1, buf3, buf4, 1728, grid=grid(1728), stream=stream0)
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.convolution]
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_8, mul, x3], Original ATen: [aten.leaky_relu, aten.mul, aten.add, aten.leaky_relu_backward]
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3.run(buf5, primals_1, buf6, buf7, 256, grid=grid(256), stream=stream0)
del buf5
del primals_1
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_4.run(buf3, buf8, 256, grid=grid(256), stream=stream0)
del buf3
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.leaky_relu, aten.leaky_relu_backward]
triton_poi_fused_leaky_relu_leaky_relu_backward_4.run(buf1, buf9, 256, grid=grid(256), stream=stream0)
del buf1
return (buf6, primals_2, primals_3, primals_4, buf0, buf2, buf4, buf7, buf8, buf9, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 12, 3, 3), (108, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class ResidualDenseBlock_3C(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(ResidualDenseBlock_3C, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, x):
residual = x
x1 = self.conv1(x)
x2 = self.conv2(torch.cat((x, x1), 1))
x3 = self.conv3(torch.cat((x, x1, x2), 1))
x3 = 0.1 * x3 + residual
return x3
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'latent_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_1(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 8
x3 = xindex // 288
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tl.full([1], 8, tl.int64)
tmp19 = tmp16 & tmp10
tmp20 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp19 & xmask, other=0.0)
tmp21 = 0.0
tmp22 = tmp20 > tmp21
tmp23 = 0.2
tmp24 = tmp20 * tmp23
tmp25 = tl.where(tmp22, tmp20, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp19, tmp25, tmp26)
tmp28 = tl.where(tmp13, tmp15, tmp27)
tmp29 = tl.full(tmp28.shape, 0.0, tmp28.dtype)
tmp30 = tl.where(tmp10, tmp28, tmp29)
tl.store(out_ptr0 + x6, tmp30, xmask)
@triton.jit
def triton_poi_fused_cat_constant_pad_nd_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1728
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 6
x0 = xindex % 6
x2 = xindex // 36 % 12
x3 = xindex // 432
x6 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = -1 + x0
tmp6 = tmp5 >= tmp1
tmp7 = tmp5 < tmp3
tmp8 = tmp2 & tmp4
tmp9 = tmp8 & tmp6
tmp10 = tmp9 & tmp7
tmp11 = x2
tmp13 = tmp11 < tmp3
tmp14 = tmp13 & tmp10
tmp15 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2 + 64 * x3), tmp14 &
xmask, other=0.0)
tmp16 = tmp11 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp11 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tmp19 & tmp10
tmp21 = tl.load(in_ptr1 + (-5 + x0 + 4 * x1 + 16 * (-4 + x2) + 64 * x3),
tmp20 & xmask, other=0.0)
tmp22 = 0.0
tmp23 = tmp21 > tmp22
tmp24 = 0.2
tmp25 = tmp21 * tmp24
tmp26 = tl.where(tmp23, tmp21, tmp25)
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp20, tmp26, tmp27)
tmp29 = tmp11 >= tmp17
tl.full([1], 12, tl.int64)
tmp32 = tmp29 & tmp10
tmp33 = tl.load(in_ptr2 + (-5 + x0 + 4 * x1 + 16 * (-8 + x2) + 64 * x3),
tmp32 & xmask, other=0.0)
tmp34 = tmp33 > tmp22
tmp35 = tmp33 * tmp24
tmp36 = tl.where(tmp34, tmp33, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp32, tmp36, tmp37)
tmp39 = tl.where(tmp19, tmp28, tmp38)
tmp40 = tl.where(tmp13, tmp15, tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp10, tmp40, tmp41)
tl.store(out_ptr0 + x6, tmp42, xmask)
@triton.jit
def triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3(in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp8 = tl.load(in_ptr1 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = 0.1
tmp7 = tmp5 * tmp6
tmp9 = tmp7 + tmp8
tmp10 = tmp5 > tmp1
tl.store(out_ptr0 + x0, tmp9, xmask)
tl.store(out_ptr1 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_leaky_relu_backward_4(in_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.0
tmp2 = tmp0 > tmp1
tmp3 = 0.2
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = tmp5 > tmp1
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (4, 12, 3, 3), (108, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(576)](primals_1, buf0, 576,
XBLOCK=128, num_warps=4, num_stages=1)
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 4, 4, 4), (64, 16, 4, 1))
buf2 = empty_strided_cuda((4, 8, 6, 6), (288, 36, 6, 1), torch.float32)
triton_poi_fused_cat_constant_pad_nd_1[grid(1152)](primals_1, buf1,
buf2, 1152, XBLOCK=128, num_warps=4, num_stages=1)
buf3 = extern_kernels.convolution(buf2, primals_3, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 4, 4), (64, 16, 4, 1))
buf4 = empty_strided_cuda((4, 12, 6, 6), (432, 36, 6, 1), torch.float32
)
triton_poi_fused_cat_constant_pad_nd_2[grid(1728)](primals_1, buf1,
buf3, buf4, 1728, XBLOCK=128, num_warps=4, num_stages=1)
buf5 = extern_kernels.convolution(buf4, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_leaky_relu_leaky_relu_backward_mul_3[grid(256)](
buf5, primals_1, buf6, buf7, 256, XBLOCK=128, num_warps=4,
num_stages=1)
del buf5
del primals_1
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_4[grid(256)](buf3,
buf8, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf3
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_leaky_relu_leaky_relu_backward_4[grid(256)](buf1,
buf9, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf1
return (buf6, primals_2, primals_3, primals_4, buf0, buf2, buf4, buf7,
buf8, buf9)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class ResidualDenseBlock_3CNew(nn.Module):
def __init__(self, in_channels, latent_channels, kernel_size=3, stride=
1, padding=1, dilation=1, pad_type='zero', activation='lrelu', norm
='none', sn=False):
super(ResidualDenseBlock_3CNew, self).__init__()
self.conv1 = Conv2dLayer(in_channels, latent_channels, kernel_size,
stride, padding, dilation, pad_type, activation, norm, sn)
self.conv2 = Conv2dLayer(in_channels + latent_channels,
latent_channels, kernel_size, stride, padding, dilation,
pad_type, activation, norm, sn)
self.conv3 = Conv2dLayer(in_channels + latent_channels * 2,
in_channels, kernel_size, stride, padding, dilation, pad_type,
activation, norm, sn)
def forward(self, input_0):
primals_2 = self.conv1.conv2d.weight
primals_3 = self.conv2.conv2d.weight
primals_4 = self.conv3.conv2d.weight
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| LenKerr/Semantic-Colorization-GAN | ResidualDenseBlock_3C | false | 5,536 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super().__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.acti
# ... truncated (>4000 chars) for memory efficiency |
KMaxPool1d | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/qf/cqfhskey3zun6jxe3jyn6ntmnpvxnpmqob4lhxlqwvyiiesmz5gl.py
# Topologically Sorted Source Nodes: [sort, gather], Original ATen: [aten.sort, aten.gather]
# Source node to ATen node mapping:
# gather => gather
# sort => sort
# Graph fragment:
# %sort : [num_users=1] = call_function[target=torch.ops.aten.sort.default](args = (%getitem_1, 2), kwargs = {})
# %gather : [num_users=1] = call_function[target=torch.ops.aten.gather.default](args = (%arg0_1, 2, %getitem_2), kwargs = {})
triton_per_fused_gather_sort_0 = async_compile.triton('triton_per_fused_gather_sort_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[16, 4],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*i64', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_gather_sort_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_gather_sort_0(in_ptr0, in_ptr1, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 16
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (4*x0)), xmask, other=0.0)
tmp1 = r1
tmp2 = tmp1.to(tl.int16)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5, tmp6, = triton_helpers.sort_with_index(tmp3, tmp4, None, 1, stable=False, descending=False)
tmp7 = tl.full([XBLOCK, RBLOCK], 4, tl.int32)
tmp8 = tmp5 + tmp7
tmp9 = tmp5 < 0
tmp10 = tl.where(tmp9, tmp8, tmp5)
tl.device_assert(((0 <= tmp10) & (tmp10 < 4)) | ~(xmask), "index out of bounds: 0 <= tmp10 < 4")
tmp12 = tl.load(in_ptr1 + (tmp10 + (4*x0)), xmask, eviction_policy='evict_last')
tl.store(out_ptr1 + (r1 + (4*x0)), tmp12, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [topk], Original ATen: [aten.topk]
buf0 = torch.ops.aten.topk.default(arg0_1, 4, 2)
buf2 = buf0[1]
del buf0
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sort, gather], Original ATen: [aten.sort, aten.gather]
stream0 = get_raw_stream(0)
triton_per_fused_gather_sort_0.run(buf2, arg0_1, buf5, 16, 4, grid=grid(16), stream=stream0)
del arg0_1
del buf2
return (buf5, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class KMaxPool1d(nn.Module):
def __init__(self, top_k: 'int'):
super(KMaxPool1d, self).__init__()
self.top_k = top_k
def forward(self, inputs):
assert inputs.dim() == 3
top_idxs = torch.topk(inputs, k=self.top_k, dim=2)[1]
sorted_top_idxs = top_idxs.sort(dim=2)[0]
return inputs.gather(dim=2, index=sorted_top_idxs)
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'top_k': 4}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_gather_sort_0(in_ptr0, in_ptr1, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 16
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 4 * x0), xmask, other=0.0)
tmp1 = r1
tmp2 = tmp1.to(tl.int16)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5, _tmp6 = triton_helpers.sort_with_index(tmp3, tmp4, None, 1,
stable=False, descending=False)
tmp7 = tl.full([XBLOCK, RBLOCK], 4, tl.int32)
tmp8 = tmp5 + tmp7
tmp9 = tmp5 < 0
tmp10 = tl.where(tmp9, tmp8, tmp5)
tl.device_assert((0 <= tmp10) & (tmp10 < 4) | ~xmask,
'index out of bounds: 0 <= tmp10 < 4')
tmp12 = tl.load(in_ptr1 + (tmp10 + 4 * x0), xmask, eviction_policy=
'evict_last')
tl.store(out_ptr1 + (r1 + 4 * x0), tmp12, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten.topk.default(arg0_1, 4, 2)
buf2 = buf0[1]
del buf0
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_gather_sort_0[grid(16)](buf2, arg0_1, buf5, 16, 4,
XBLOCK=8, num_warps=2, num_stages=1)
del arg0_1
del buf2
return buf5,
class KMaxPool1dNew(nn.Module):
def __init__(self, top_k: 'int'):
super(KMaxPool1dNew, self).__init__()
self.top_k = top_k
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | KMaxPool1d | false | 5,537 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, top_k: 'int'):
super().__init__()
self.top_k = top_k
def forward(self, inputs):
assert inputs.dim() == 3
top_idxs = torch.topk(inputs, k=self.top_k, dim=2)[1]
sorted_top_idxs = top_idxs.sort(dim=2)[0]
return inputs.gather(dim=2, index=sorted_top_idxs)
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4]
|
Biaffine | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/bc/cbcett6ey62xkijoadrmiqwnmvdqa242vrdqwxiw4pvecwqjoged.py
# Topologically Sorted Source Nodes: [input1], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# input1 => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %full_default], -1), kwargs = {})
triton_poi_fused_cat_0 = async_compile.triton('triton_poi_fused_cat_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 80
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 5
x1 = (xindex // 5)
x2 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((4*x1) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 5, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = 1.0
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp6, tmp9, tmp10)
tmp12 = tl.where(tmp4, tmp5, tmp11)
tl.store(out_ptr0 + (x2), tmp12, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (5, 5), (5, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 5), (20, 5, 1), torch.float32)
# Topologically Sorted Source Nodes: [input1], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(primals_1, buf0, 80, grid=grid(80), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((16, 5), (5, 1), torch.float32)
# Topologically Sorted Source Nodes: [affine], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf0, (16, 5), (5, 1), 0), reinterpret_tensor(primals_3, (5, 5), (1, 5), 0), out=buf1)
del primals_3
buf2 = empty_strided_cuda((4, 4, 5), (20, 5, 1), torch.float32)
# Topologically Sorted Source Nodes: [input2], Original ATen: [aten.cat]
triton_poi_fused_cat_0.run(primals_2, buf2, 80, grid=grid(80), stream=stream0)
del primals_2
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bmm], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf1, (4, 4, 5), (20, 5, 1), 0), reinterpret_tensor(buf2, (4, 5, 4), (20, 1, 5), 0), out=buf3)
del buf1
return (reinterpret_tensor(buf3, (4, 4, 4), (16, 1, 4), 0), reinterpret_tensor(buf0, (16, 5), (5, 1), 0), buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((5, 5), (5, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Biaffine(nn.Module):
def __init__(self, in_features, out_features=1, bias=(True, True)):
super(Biaffine, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.bias = bias
self.linear_input_size = in_features + bias[0]
self.linear_output_size = out_features * (in_features + bias[1])
self.linear = nn.Linear(in_features=self.linear_input_size,
out_features=self.linear_output_size, bias=False)
self.reset_parameters()
@property
def weight(self):
return self.linear.weight
def reset_parameters(self):
nn.init.xavier_uniform_(self.linear.weight)
def forward(self, input1, input2):
batch_size, len1, _dim1 = input1.size()
batch_size, len2, _dim2 = input2.size()
if self.bias[0]:
ones = input1.data.new_ones(batch_size, len1, 1)
input1 = torch.cat((input1, ones), dim=-1)
if self.bias[1]:
ones = input2.data.new_ones(batch_size, len2, 1)
input2 = torch.cat((input2, ones), dim=-1)
affine = self.linear(input1)
affine = affine.reshape(batch_size, len1 * self.out_features, -1)
biaffine = torch.bmm(affine, input2.transpose(1, 2)).transpose(1, 2)
biaffine = biaffine.reshape((batch_size, len2, len1, -1)).squeeze(-1)
return biaffine
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 80
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 5
x1 = xindex // 5
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 5, tl.int64)
tmp9 = 1.0
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp6, tmp9, tmp10)
tmp12 = tl.where(tmp4, tmp5, tmp11)
tl.store(out_ptr0 + x2, tmp12, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (5, 5), (5, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 5), (20, 5, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(80)](primals_1, buf0, 80, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((16, 5), (5, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (16, 5), (5, 1), 0),
reinterpret_tensor(primals_3, (5, 5), (1, 5), 0), out=buf1)
del primals_3
buf2 = empty_strided_cuda((4, 4, 5), (20, 5, 1), torch.float32)
triton_poi_fused_cat_0[grid(80)](primals_2, buf2, 80, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_2
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf1, (4, 4, 5), (20, 5, 1),
0), reinterpret_tensor(buf2, (4, 5, 4), (20, 1, 5), 0), out=buf3)
del buf1
return reinterpret_tensor(buf3, (4, 4, 4), (16, 1, 4), 0
), reinterpret_tensor(buf0, (16, 5), (5, 1), 0), buf2
class BiaffineNew(nn.Module):
def __init__(self, in_features, out_features=1, bias=(True, True)):
super(BiaffineNew, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.bias = bias
self.linear_input_size = in_features + bias[0]
self.linear_output_size = out_features * (in_features + bias[1])
self.linear = nn.Linear(in_features=self.linear_input_size,
out_features=self.linear_output_size, bias=False)
self.reset_parameters()
@property
def weight(self):
return self.linear.weight
def reset_parameters(self):
nn.init.xavier_uniform_(self.linear.weight)
def forward(self, input_0, input_1):
primals_3 = self.linear.weight
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | Biaffine | false | 5,538 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_features, out_features=1, bias=(True, True)):
super().__init__()
self.in_features = in_features
self.out_features = out_features
self.bias = bias
self.linear_input_size = in_features + bias[0]
self.linear_output_size = out_features * (in_features + bias[1])
self.linear = nn.Linear(in_features=self.linear_input_size,
out_features=self.linear_output_size, bias=False)
self.reset_parameters()
@property
def weight(self):
return self.linear.weight
def reset_parameters(self):
nn.init.xavier_uniform_(self.linear.weight)
def forward(self, input1, input2):
batch_size, len1, _dim1 = input1.size()
batch_size, len2, _dim2 = input2.size()
if self.bias[0]:
ones = input1.data.new_ones(batch_size, len1, 1)
input1 = torch.cat((input1, ones), dim=-1)
if self.bias[1]:
ones = input2.data.new_ones(batch_size, len2, 1)
input2 = torch.cat((input2, ones), dim=-1)
affine = self.linear(input1)
affine = affine.reshape(batch_size, len1 * self.out_features, -1)
biaffine = torch.bmm(affine, input2.transpose(1, 2)).transpose(1, 2)
biaffine = biaffine.reshape((batch_size, len2, len1, -1)).squeeze(-1)
return biaffine
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4]
|
L2Norm | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/3i/c3iedxosa453ppy7izskpdsxaah4sgqlvqpbehs3lyv5xhslwtpv.py
# Topologically Sorted Source Nodes: [pow_1, sum_1, norm, x, out], Original ATen: [aten.pow, aten.sum, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# norm => sqrt
# out => mul
# pow_1 => pow_1
# sum_1 => sum_1
# x => div
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%primals_1, 2), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [1], True), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%sum_1,), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_1, %sqrt), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %primals_2), kwargs = {})
triton_poi_fused_div_mul_pow_sqrt_sum_0 = async_compile.triton('triton_poi_fused_div_mul_pow_sqrt_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_mul_pow_sqrt_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_mul_pow_sqrt_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
x1 = (xindex // 16) % 4
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = tmp0 / tmp12
tmp15 = tmp13 * tmp14
tl.store(out_ptr0 + (x3), tmp15, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1, 4, 1, 1), (4, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_1, sum_1, norm, x, out], Original ATen: [aten.pow, aten.sum, aten.sqrt, aten.div, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_div_mul_pow_sqrt_sum_0.run(primals_1, primals_2, buf0, 256, grid=grid(256), stream=stream0)
del primals_2
return (buf0, primals_1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((1, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.init as init
class L2Norm(nn.Module):
def __init__(self, channels, scale):
super(L2Norm, self).__init__()
self.channels = channels
self.scale = scale
self.rescale_factors = nn.Parameter(torch.FloatTensor(1, channels,
1, 1))
self.reset_params()
def reset_params(self):
init.constant_(self.rescale_factors, self.scale)
def forward(self, x):
norm = x.pow(2).sum(dim=1, keepdim=True).sqrt()
x = x / norm
out = x * self.rescale_factors
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'channels': 4, 'scale': 1.0}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.nn.init as init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_div_mul_pow_sqrt_sum_0(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp14 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = tmp0 / tmp12
tmp15 = tmp13 * tmp14
tl.store(out_ptr0 + x3, tmp15, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1, 4, 1, 1), (4, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_mul_pow_sqrt_sum_0[grid(256)](primals_1,
primals_2, buf0, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
return buf0, primals_1
class L2NormNew(nn.Module):
def __init__(self, channels, scale):
super(L2NormNew, self).__init__()
self.channels = channels
self.scale = scale
self.rescale_factors = nn.Parameter(torch.FloatTensor(1, channels,
1, 1))
self.reset_params()
def reset_params(self):
init.constant_(self.rescale_factors, self.scale)
def forward(self, input_0):
primals_2 = self.rescale_factors
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| Liuhongzhi2018/Car_detection | L2Norm | false | 5,539 | [
"MIT"
] | 1 | f32fea9c348c691ccc30b9804a4f3fa32732bbae | https://github.com/Liuhongzhi2018/Car_detection/tree/f32fea9c348c691ccc30b9804a4f3fa32732bbae | import torch
import torch.nn as nn
import torch.nn.init as init
class Model(nn.Module):
def __init__(self, channels, scale):
super().__init__()
self.channels = channels
self.scale = scale
self.rescale_factors = nn.Parameter(torch.FloatTensor(1, channels,
1, 1))
self.reset_params()
def reset_params(self):
init.constant_(self.rescale_factors, self.scale)
def forward(self, x):
norm = x.pow(2).sum(dim=1, keepdim=True).sqrt()
x = x / norm
out = x * self.rescale_factors
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 1.0]
|
PositionWiseFeedForward | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/zy/czylxf6rfbnbz2ddgd3xovxwjqnfen7sgqej5mnv46j2fekwnniz.py
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# output => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[131072],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 131072
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 2048
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, None)
tl.store(out_ptr0 + (x2), tmp6, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/mx/cmxyblvny6ryx4zft5bojylqtaa5rz2epkem2quffgbnvnc2z4fy.py
# Topologically Sorted Source Nodes: [add, mean, std], Original ATen: [aten.add, aten.mean, aten.std]
# Source node to ATen node mapping:
# add => add
# mean => mean
# std => var
# Graph fragment:
# %add : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %primals_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add, [-1], True), kwargs = {})
# %var : [num_users=1] = call_function[target=torch.ops.aten.var.correction](args = (%add, [-1]), kwargs = {correction: 1.0, keepdim: True})
triton_poi_fused_add_mean_std_1 = async_compile.triton('triton_poi_fused_add_mean_std_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mean_std_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mean_std_1(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = 3.0
tmp29 = tmp27 / tmp28
tl.store(in_out_ptr0 + (x0), tmp29, xmask)
tl.store(out_ptr0 + (x0), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/da/cdabe3pqqspscvgqgnm3vaiwdzwvoew5ch5fapm37mrjq7zo4e55.py
# Topologically Sorted Source Nodes: [add, mean, std, sub, add_1, ln_out, mul, ln_out_1], Original ATen: [aten.add, aten.mean, aten.std, aten.sub, aten.div, aten.mul]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# ln_out => div
# ln_out_1 => add_2
# mean => mean
# mul => mul
# std => sqrt
# sub => sub
# Graph fragment:
# %add : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %primals_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add, [-1], True), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%var,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %mean), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sqrt, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, %add_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_6, %div), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_7), kwargs = {})
triton_poi_fused_add_div_mean_mul_std_sub_2 = async_compile.triton('triton_poi_fused_add_div_mean_mul_std_sub_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_mean_mul_std_sub_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_2(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tl.load(in_ptr2 + (x2), xmask)
tmp4 = tl.load(in_ptr3 + (x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + (x1), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 - tmp4
tmp7 = libdevice.sqrt(tmp6)
tmp8 = 1e-06
tmp9 = tmp7 + tmp8
tmp10 = tmp5 / tmp9
tmp11 = tmp0 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + (x2), tmp13, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (2048, 4), (4, 1))
assert_size_stride(primals_3, (2048, ), (1, ))
assert_size_stride(primals_4, (4, 2048), (2048, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 2048), (2048, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 2048), (1, 4), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 2048), (32768, 8192, 2048, 1), 0); del buf0 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 2048), (32768, 8192, 2048, 1), torch.bool)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_3, buf7, 131072, grid=grid(131072), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 2048), (2048, 1), 0), reinterpret_tensor(primals_4, (2048, 4), (1, 2048), 0), alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = buf3; del buf3 # reuse
buf5 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [add, mean, std], Original ATen: [aten.add, aten.mean, aten.std]
triton_poi_fused_add_mean_std_1.run(buf4, buf2, primals_1, buf5, 64, grid=grid(64), stream=stream0)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [add, mean, std, sub, add_1, ln_out, mul, ln_out_1], Original ATen: [aten.add, aten.mean, aten.std, aten.sub, aten.div, aten.mul]
triton_poi_fused_add_div_mean_mul_std_sub_2.run(primals_6, buf2, primals_1, buf5, buf4, primals_7, buf6, 256, grid=grid(256), stream=stream0)
del buf4
del buf5
del primals_7
return (buf6, primals_1, primals_6, reinterpret_tensor(buf1, (64, 2048), (2048, 1), 0), buf2, primals_4, buf7, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((2048, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2048, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 2048), (2048, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.init as init
class LayerNorm(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, x):
mean = x.mean(dim=-1, keepdim=True)
std = x.std(dim=-1, keepdim=True)
ln_out = (x - mean) / (std + self.eps)
ln_out = self.gamma * ln_out + self.beta
return ln_out
class Linear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super(Linear, self).__init__()
self.linear = nn.Linear(in_features, out_features, bias=bias)
init.xavier_normal_(self.linear.weight)
init.zeros_(self.linear.bias)
def forward(self, inputs):
return self.linear(inputs)
class PositionWiseFeedForward(nn.Module):
def __init__(self, d_model, d_ff=2048, dropout=0.1):
super(PositionWiseFeedForward, self).__init__()
self.relu = nn.ReLU()
self.fc1 = Linear(d_model, d_ff)
self.fc2 = Linear(d_ff, d_model)
self.dropout = nn.Dropout(dropout)
self.layernorm = LayerNorm(d_model)
def forward(self, x):
residual = x
output = self.relu(self.fc1(x))
output = self.dropout(self.fc2(output))
output = self.layernorm(output + residual)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.nn.init as init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 2048
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_add_mean_std_1(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = 3.0
tmp29 = tmp27 / tmp28
tl.store(in_out_ptr0 + x0, tmp29, xmask)
tl.store(out_ptr0 + x0, tmp16, xmask)
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_2(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tl.load(in_ptr2 + x2, xmask)
tmp4 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + x1, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 - tmp4
tmp7 = libdevice.sqrt(tmp6)
tmp8 = 1e-06
tmp9 = tmp7 + tmp8
tmp10 = tmp5 / tmp9
tmp11 = tmp0 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + x2, tmp13, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (2048, 4), (4, 1))
assert_size_stride(primals_3, (2048,), (1,))
assert_size_stride(primals_4, (4, 2048), (2048, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 2048), (2048, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 2048), (1, 4), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 2048), (32768, 8192, 2048,
1), 0)
del buf0
buf7 = empty_strided_cuda((4, 4, 4, 2048), (32768, 8192, 2048, 1),
torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(131072)](buf1,
primals_3, buf7, 131072, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 2048),
(2048, 1), 0), reinterpret_tensor(primals_4, (2048, 4), (1,
2048), 0), alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = buf3
del buf3
buf5 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
triton_poi_fused_add_mean_std_1[grid(64)](buf4, buf2, primals_1,
buf5, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_std_sub_2[grid(256)](primals_6,
buf2, primals_1, buf5, buf4, primals_7, buf6, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf4
del buf5
del primals_7
return buf6, primals_1, primals_6, reinterpret_tensor(buf1, (64, 2048),
(2048, 1), 0), buf2, primals_4, buf7
class LayerNorm(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, x):
mean = x.mean(dim=-1, keepdim=True)
std = x.std(dim=-1, keepdim=True)
ln_out = (x - mean) / (std + self.eps)
ln_out = self.gamma * ln_out + self.beta
return ln_out
class Linear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super(Linear, self).__init__()
self.linear = nn.Linear(in_features, out_features, bias=bias)
init.xavier_normal_(self.linear.weight)
init.zeros_(self.linear.bias)
def forward(self, inputs):
return self.linear(inputs)
class PositionWiseFeedForwardNew(nn.Module):
def __init__(self, d_model, d_ff=2048, dropout=0.1):
super(PositionWiseFeedForwardNew, self).__init__()
self.relu = nn.ReLU()
self.fc1 = Linear(d_model, d_ff)
self.fc2 = Linear(d_ff, d_model)
self.dropout = nn.Dropout(dropout)
self.layernorm = LayerNorm(d_model)
def forward(self, input_0):
primals_2 = self.fc1.linear.weight
primals_3 = self.fc1.linear.bias
primals_4 = self.fc2.linear.weight
primals_5 = self.fc2.linear.bias
primals_6 = self.layernorm.gamma
primals_7 = self.layernorm.beta
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| Lhx94As/PHO-LID | PositionWiseFeedForward | false | 5,540 | [
"MIT"
] | 1 | 44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | https://github.com/Lhx94As/PHO-LID/tree/44843b25b977dd6e0b77b520dbe3f2ff1ea633cd | import torch
import torch.nn as nn
import torch.nn.init as init
class LayerNorm(nn.Module):
def __init__(self, d_hid, eps=1e-06):
super().__init__()
self.gamma = nn.Parameter(torch.ones(d_hid))
self.beta = nn.Parameter(torch.zeros(d_hid))
self.eps = eps
def forward(self, x):
mean = x.mean(dim=-1, keepdim=True)
std = x.std(dim=-1, keepdim=True)
ln_out = (x - mean) / (std + self.eps)
ln_out = self.gamma * ln_out + self.beta
return ln_out
class Linear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear = nn.Linear(in_features, out_features, bias=bias)
init.xavier_normal_(self.linear.weight)
init.zeros_(self.linear.bias)
def forward(self, inputs):
return self.linear(inputs)
class Model(nn.Module):
def __init__(self, d_model, d_ff=2048, dropout=0.1):
super().__init__()
self.relu = nn.ReLU()
self.fc1 = Linear(d_model, d_ff)
self.fc2 = Linear(d_ff, d_model)
self.dropout = nn.Dropout(dropout)
self.layernorm = LayerNorm(d_model)
def forward(self, x):
residual = x
output = self.relu(self.fc1(x))
output = self.dropout(self.fc2(output))
output = self.layernorm(output + residual)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
VAE | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/iv/civ5tj3syvyx4mu3v7dhhyywq75qhwo5ijklk46fa3gmvy43dnjf.py
# Topologically Sorted Source Nodes: [mul, std, mul_1, z_1, add_1, pow_1, sub, exp_1, sub_1, sum_1, mul_2, kl_loss], Original ATen: [aten.mul, aten.exp, aten.add, aten.pow, aten.sub, aten.sum, aten.div]
# Source node to ATen node mapping:
# add_1 => add_1
# exp_1 => exp_1
# kl_loss => div
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# pow_1 => pow_1
# std => exp
# sub => sub
# sub_1 => sub_1
# sum_1 => sum_1
# z_1 => add
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_3, 0.5), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%mul,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%randn, %exp), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %view_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, 1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_1, 2), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_1, %pow_1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%view_3,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%sub, %exp_1), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%sub_1,), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_1, -0.5), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_2, 4), kwargs = {})
triton_per_fused_add_div_exp_mul_pow_sub_sum_0 = async_compile.triton('triton_per_fused_add_div_exp_mul_pow_sub_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {6: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 7), equal_to_1=(6,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_exp_mul_pow_sub_sum_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_exp_mul_pow_sub_sum_0(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex % 16
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (r2), None)
tmp6 = tl.load(in_ptr2 + (r2), None)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp0 * tmp4
tmp7 = tmp5 + tmp6
tmp8 = tl_math.exp(tmp1)
tmp9 = 1.0
tmp10 = tmp1 + tmp9
tmp11 = tmp6 * tmp6
tmp12 = tmp10 - tmp11
tmp13 = tmp12 - tmp8
tmp14 = tl.broadcast_to(tmp13, [RBLOCK])
tmp16 = triton_helpers.promote_to_tensor(tl.sum(tmp14, 0))
tmp17 = -0.5
tmp18 = tmp16 * tmp17
tmp19 = 0.25
tmp20 = tmp18 * tmp19
tl.store(out_ptr0 + (tl.broadcast_to(r2, [RBLOCK])), tmp7, None)
tl.store(out_ptr1 + (tl.broadcast_to(r2, [RBLOCK])), tmp8, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp20, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qx/cqxtiuijob6zwtx2fbrdeze2bxnzigjok4rc5o4e7udmnr4hlmi7.py
# Topologically Sorted Source Nodes: [dist_loss], Original ATen: [aten.mse_loss]
# Source node to ATen node mapping:
# dist_loss => mean, pow_2, sub_2
# Graph fragment:
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_5, %primals_3), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub_2, 2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%pow_2,), kwargs = {})
triton_per_fused_mse_loss_1 = async_compile.triton('triton_per_fused_mse_loss_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mse_loss_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mse_loss_1(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [RBLOCK])
tmp6 = triton_helpers.promote_to_tensor(tl.sum(tmp4, 0))
tmp7 = 256.0
tmp8 = tmp6 / tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp8, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mean], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [logv], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf1)
del primals_4
del primals_5
# Topologically Sorted Source Nodes: [z], Original ATen: [aten.randn]
buf2 = torch.ops.aten.randn.default([4, 4], device=device(type='cuda', index=0), pin_memory=False)
buf3 = buf2
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf7 = empty_strided_cuda((), (), torch.float32)
buf10 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [mul, std, mul_1, z_1, add_1, pow_1, sub, exp_1, sub_1, sum_1, mul_2, kl_loss], Original ATen: [aten.mul, aten.exp, aten.add, aten.pow, aten.sub, aten.sum, aten.div]
stream0 = get_raw_stream(0)
triton_per_fused_add_div_exp_mul_pow_sub_sum_0.run(buf10, buf3, buf1, buf0, buf4, buf6, 1, 256, grid=grid(1), stream=stream0)
buf5 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [restruct_hidden], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf4, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf5)
del primals_7
buf8 = empty_strided_cuda((), (), torch.float32)
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [dist_loss], Original ATen: [aten.mse_loss]
triton_per_fused_mse_loss_1.run(buf9, buf5, primals_3, 1, 256, grid=grid(1), stream=stream0)
return (buf9, buf10, primals_3, buf0, buf1, buf3, reinterpret_tensor(buf4, (64, 4), (4, 1), 0), buf5, buf6, primals_6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class VAE(nn.Module):
def __init__(self, input_dim, latent_dim):
super(VAE, self).__init__()
self.latent_dim = latent_dim
self.hidden2mean = nn.Linear(input_dim, latent_dim)
self.hidden2logv = nn.Linear(input_dim, latent_dim)
self.latent2hidden = nn.Linear(latent_dim, input_dim)
self.reset_params()
def reset_params(self):
nn.init.xavier_uniform_(self.hidden2mean.weight)
nn.init.xavier_uniform_(self.hidden2logv.weight)
nn.init.xavier_uniform_(self.latent2hidden.weight)
def forward(self, h):
mean = self.hidden2mean(h)
logv = self.hidden2logv(h)
std = torch.exp(0.5 * logv)
z = torch.randn((h.size(0), self.latent_dim), device=h.device)
z = z * std + mean
restruct_hidden = self.latent2hidden(z)
kl_loss = -0.5 * torch.sum(1 + logv - mean.pow(2) - logv.exp()
) / logv.size(0)
dist_loss = F.mse_loss(restruct_hidden, h)
return dist_loss, kl_loss
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_dim': 4, 'latent_dim': 4}]
| import torch
from torch import device
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_add_div_exp_mul_pow_sub_sum_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex % 16
r2 = rindex
tmp0 = tl.load(in_ptr0 + r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + r2, None)
tmp6 = tl.load(in_ptr2 + r2, None)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp0 * tmp4
tmp7 = tmp5 + tmp6
tmp8 = tl_math.exp(tmp1)
tmp9 = 1.0
tmp10 = tmp1 + tmp9
tmp11 = tmp6 * tmp6
tmp12 = tmp10 - tmp11
tmp13 = tmp12 - tmp8
tmp14 = tl.broadcast_to(tmp13, [RBLOCK])
tmp16 = triton_helpers.promote_to_tensor(tl.sum(tmp14, 0))
tmp17 = -0.5
tmp18 = tmp16 * tmp17
tmp19 = 0.25
tmp20 = tmp18 * tmp19
tl.store(out_ptr0 + tl.broadcast_to(r2, [RBLOCK]), tmp7, None)
tl.store(out_ptr1 + tl.broadcast_to(r2, [RBLOCK]), tmp8, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp20, None)
@triton.jit
def triton_per_fused_mse_loss_1(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [RBLOCK])
tmp6 = triton_helpers.promote_to_tensor(tl.sum(tmp4, 0))
tmp7 = 256.0
tmp8 = tmp6 / tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp8, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf1)
del primals_4
del primals_5
buf2 = torch.ops.aten.randn.default([4, 4], device=device(type=
'cuda', index=0), pin_memory=False)
buf3 = buf2
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf7 = empty_strided_cuda((), (), torch.float32)
buf10 = buf7
del buf7
get_raw_stream(0)
triton_per_fused_add_div_exp_mul_pow_sub_sum_0[grid(1)](buf10, buf3,
buf1, buf0, buf4, buf6, 1, 256, num_warps=2, num_stages=1)
buf5 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(buf4, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf5)
del primals_7
buf8 = empty_strided_cuda((), (), torch.float32)
buf9 = buf8
del buf8
triton_per_fused_mse_loss_1[grid(1)](buf9, buf5, primals_3, 1, 256,
num_warps=2, num_stages=1)
return buf9, buf10, primals_3, buf0, buf1, buf3, reinterpret_tensor(buf4,
(64, 4), (4, 1), 0), buf5, buf6, primals_6
class VAENew(nn.Module):
def __init__(self, input_dim, latent_dim):
super(VAENew, self).__init__()
self.latent_dim = latent_dim
self.hidden2mean = nn.Linear(input_dim, latent_dim)
self.hidden2logv = nn.Linear(input_dim, latent_dim)
self.latent2hidden = nn.Linear(latent_dim, input_dim)
self.reset_params()
def reset_params(self):
nn.init.xavier_uniform_(self.hidden2mean.weight)
nn.init.xavier_uniform_(self.hidden2logv.weight)
nn.init.xavier_uniform_(self.latent2hidden.weight)
def forward(self, input_0):
primals_1 = self.hidden2mean.weight
primals_2 = self.hidden2mean.bias
primals_4 = self.hidden2logv.weight
primals_5 = self.hidden2logv.bias
primals_6 = self.latent2hidden.weight
primals_7 = self.latent2hidden.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0], output[1]
| LindgeW/DomainAdaption4DependencyParsing | VAE | false | 5,541 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, input_dim, latent_dim):
super().__init__()
self.latent_dim = latent_dim
self.hidden2mean = nn.Linear(input_dim, latent_dim)
self.hidden2logv = nn.Linear(input_dim, latent_dim)
self.latent2hidden = nn.Linear(latent_dim, input_dim)
self.reset_params()
def reset_params(self):
nn.init.xavier_uniform_(self.hidden2mean.weight)
nn.init.xavier_uniform_(self.hidden2logv.weight)
nn.init.xavier_uniform_(self.latent2hidden.weight)
def forward(self, h):
mean = self.hidden2mean(h)
logv = self.hidden2logv(h)
std = torch.exp(0.5 * logv)
z = torch.randn((h.size(0), self.latent_dim), device=h.device)
z = z * std + mean
restruct_hidden = self.latent2hidden(z)
kl_loss = -0.5 * torch.sum(1 + logv - mean.pow(2) - logv.exp()
) / logv.size(0)
dist_loss = F.mse_loss(restruct_hidden, h)
return dist_loss, kl_loss
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
BiaffineAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/3l/c3lo77c7wjxasxrhtr6wesb72ods2d2rxnxhbfieun7j2wukm3wn.py
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_4, %primals_3], -1), kwargs = {})
triton_poi_fused_cat_0 = async_compile.triton('triton_poi_fused_cat_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = (xindex // 8)
x2 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((4*x1) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((4*x1) + ((-4) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x2), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/bd/cbd5vrzay2jbzmubtgdwxlp37vuwvz6hkpnirlmnp5x7bwlay5gq.py
# Topologically Sorted Source Nodes: [att_weigths], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weigths => amax, clone, exp, sub, sum_1
# Graph fragment:
# %clone : [num_users=2] = call_function[target=torch.ops.aten.clone.default](args = (%permute_1,), kwargs = {memory_format: torch.contiguous_format})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%clone, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clone, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 9, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (16*x1)), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x0 + (16*x1)), xmask)
tmp5 = tl.load(in_ptr0 + (4 + x0 + (16*x1)), xmask)
tmp7 = tl.load(in_ptr2 + (4 + x0 + (16*x1)), xmask)
tmp10 = tl.load(in_ptr0 + (8 + x0 + (16*x1)), xmask)
tmp12 = tl.load(in_ptr2 + (8 + x0 + (16*x1)), xmask)
tmp15 = tl.load(in_ptr0 + (12 + x0 + (16*x1)), xmask)
tmp17 = tl.load(in_ptr2 + (12 + x0 + (16*x1)), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp5 + tmp1
tmp8 = tmp6 + tmp7
tmp9 = triton_helpers.maximum(tmp4, tmp8)
tmp11 = tmp10 + tmp1
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp9, tmp13)
tmp16 = tmp15 + tmp1
tmp18 = tmp16 + tmp17
tmp19 = triton_helpers.maximum(tmp14, tmp18)
tmp20 = tmp4 - tmp19
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp8 - tmp19
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp19
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp28 = tmp18 - tmp19
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp27 + tmp29
tl.store(out_ptr0 + (x2), tmp19, xmask)
tl.store(out_ptr1 + (x2), tmp30, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dr/cdrn6bnobxdzhbn4umg6cz5cycfofs62kppoaiycjgik6nkueiql.py
# Topologically Sorted Source Nodes: [att_weigths], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weigths => clone, div, exp, sub
# Graph fragment:
# %clone : [num_users=2] = call_function[target=torch.ops.aten.clone.default](args = (%permute_1,), kwargs = {memory_format: torch.contiguous_format})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clone, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_2 = async_compile.triton('triton_poi_fused__softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y1 = (yindex // 4)
y0 = yindex % 4
tmp0 = tl.load(in_ptr0 + (x2 + (4*y3)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x2 + (4*y3)), xmask & ymask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x2 + (4*y1)), xmask & ymask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr4 + (x2 + (4*y1)), xmask & ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tl.store(out_ptr0 + (y0 + (4*x2) + (16*y1)), tmp9, xmask & ymask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_5, (4, 8), (8, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten._trilinear]
buf0 = torch.ops.aten._trilinear.default(reinterpret_tensor(primals_4, (16, 4), (4, 1), 0), primals_1, reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), [1, 3], [0], [1, 2], [2, 3])
del primals_1
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(primals_4, primals_3, buf2, 128, grid=grid(128), stream=stream0)
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf2, (16, 8), (8, 1), 0), reinterpret_tensor(primals_5, (8, 4), (1, 8), 0), out=buf3)
del primals_5
buf4 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf5 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [att_weigths], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf1, primals_2, buf3, buf4, buf5, 16, grid=grid(16), stream=stream0)
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [att_weigths], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf1, primals_2, buf3, buf4, buf5, buf6, 16, 4, grid=grid(16, 4), stream=stream0)
del buf1
del buf4
del buf5
del primals_2
buf7 = reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [att_out], Original ATen: [aten.bmm]
extern_kernels.bmm(buf6, primals_3, out=buf7)
return (buf7, primals_3, reinterpret_tensor(primals_4, (16, 4), (4, 1), 0), reinterpret_tensor(buf2, (16, 8), (8, 1), 0), buf6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 8), (8, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class BiaffineAttention(nn.Module):
def __init__(self, in1_features, in2_features, num_label, bias=True):
super(BiaffineAttention, self).__init__()
self.bilinear = nn.Bilinear(in1_features, in2_features, num_label,
bias=bias)
self.linear = nn.Linear(in1_features + in2_features, num_label,
bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, head, dep):
"""
:param head: [batch, seq_len, hidden] 输入特征1, 即label-head
:param dep: [batch, seq_len, hidden] 输入特征2, 即label-dep
:return output: [batch, seq_len, num_cls] 每个元素对应类别的概率图
"""
output = self.bilinear(head, dep)
biaff_score = output + self.linear(torch.cat((head, dep), dim=-1))
biaff_score = biaff_score.transpose(1, 2)
att_weigths = self.softmax(biaff_score)
att_out = torch.bmm(att_weigths, dep)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'in1_features': 4, 'in2_features': 4, 'num_label': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 16 * x1), xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x0 + 16 * x1), xmask)
tmp5 = tl.load(in_ptr0 + (4 + x0 + 16 * x1), xmask)
tmp7 = tl.load(in_ptr2 + (4 + x0 + 16 * x1), xmask)
tmp10 = tl.load(in_ptr0 + (8 + x0 + 16 * x1), xmask)
tmp12 = tl.load(in_ptr2 + (8 + x0 + 16 * x1), xmask)
tmp15 = tl.load(in_ptr0 + (12 + x0 + 16 * x1), xmask)
tmp17 = tl.load(in_ptr2 + (12 + x0 + 16 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp5 + tmp1
tmp8 = tmp6 + tmp7
tmp9 = triton_helpers.maximum(tmp4, tmp8)
tmp11 = tmp10 + tmp1
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp9, tmp13)
tmp16 = tmp15 + tmp1
tmp18 = tmp16 + tmp17
tmp19 = triton_helpers.maximum(tmp14, tmp18)
tmp20 = tmp4 - tmp19
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp8 - tmp19
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp19
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp28 = tmp18 - tmp19
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp27 + tmp29
tl.store(out_ptr0 + x2, tmp19, xmask)
tl.store(out_ptr1 + x2, tmp30, xmask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
out_ptr0, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y1 = yindex // 4
y0 = yindex % 4
tmp0 = tl.load(in_ptr0 + (x2 + 4 * y3), xmask & ymask, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x2 + 4 * y3), xmask & ymask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr3 + (x2 + 4 * y1), xmask & ymask, eviction_policy=
'evict_last')
tmp8 = tl.load(in_ptr4 + (x2 + 4 * y1), xmask & ymask, eviction_policy=
'evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tl.store(out_ptr0 + (y0 + 4 * x2 + 16 * y1), tmp9, xmask & ymask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_5, (4, 8), (8, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten._trilinear.default(reinterpret_tensor(
primals_4, (16, 4), (4, 1), 0), primals_1, reinterpret_tensor(
primals_3, (16, 4), (4, 1), 0), [1, 3], [0], [1, 2], [2, 3])
del primals_1
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(128)](primals_4, primals_3, buf2, 128,
XBLOCK=128, num_warps=4, num_stages=1)
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (16, 8), (8, 1), 0),
reinterpret_tensor(primals_5, (8, 4), (1, 8), 0), out=buf3)
del primals_5
buf4 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf5 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused__softmax_1[grid(16)](buf1, primals_2, buf3, buf4,
buf5, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_2[grid(16, 4)](buf1, primals_2, buf3,
buf4, buf5, buf6, 16, 4, XBLOCK=4, YBLOCK=8, num_warps=1,
num_stages=1)
del buf1
del buf4
del buf5
del primals_2
buf7 = reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0)
del buf3
extern_kernels.bmm(buf6, primals_3, out=buf7)
return buf7, primals_3, reinterpret_tensor(primals_4, (16, 4), (4, 1), 0
), reinterpret_tensor(buf2, (16, 8), (8, 1), 0), buf6
class BiaffineAttentionNew(nn.Module):
def __init__(self, in1_features, in2_features, num_label, bias=True):
super(BiaffineAttentionNew, self).__init__()
self.bilinear = nn.Bilinear(in1_features, in2_features, num_label,
bias=bias)
self.linear = nn.Linear(in1_features + in2_features, num_label,
bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, input_0, input_1):
primals_1 = self.bilinear.weight
primals_2 = self.bilinear.bias
primals_5 = self.linear.weight
primals_3 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | BiaffineAttention | false | 5,542 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in1_features, in2_features, num_label, bias=True):
super().__init__()
self.bilinear = nn.Bilinear(in1_features, in2_features, num_label,
bias=bias)
self.linear = nn.Linear(in1_features + in2_features, num_label,
bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, head, dep):
"""
:param head: [batch, seq_len, hidden] 输入特征1, 即label-head
:param dep: [batch, seq_len, hidden] 输入特征2, 即label-dep
:return output: [batch, seq_len, num_cls] 每个元素对应类别的概率图
"""
output = self.bilinear(head, dep)
biaff_score = output + self.linear(torch.cat((head, dep), dim=-1))
biaff_score = biaff_score.transpose(1, 2)
att_weigths = self.softmax(biaff_score)
att_out = torch.bmm(att_weigths, dep)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
FocalLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/tj/ctjenrmpmnfu2d3pk5t7qu4ih2v4kysuba4mugrps3thgibhpgig.py
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits, mul_2, sub_2, mul_3, add_1, pred_sigmoid, sub, mul, sub_1, mul_1, pt, pow_1, focal_weight, loss, loss_1, loss_cls], Original ATen: [aten.binary_cross_entropy_with_logits, aten.mul, aten.rsub, aten.add, aten.sigmoid, aten.pow, aten.mean]
# Source node to ATen node mapping:
# add_1 => add_1
# binary_cross_entropy_with_logits => abs_1, exp, full_default, log1p, minimum, mul_5, neg, sub_3, sub_4, sub_5
# focal_weight => mul_4
# loss => mul_6
# loss_1 => mean
# loss_cls => mul_7
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# pow_1 => pow_1
# pred_sigmoid => sigmoid
# pt => add
# sub => sub
# sub_1 => sub_1
# sub_2 => sub_2
# Graph fragment:
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_3, %arg0_1), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %minimum : [num_users=1] = call_function[target=torch.ops.aten.minimum.default](args = (%full_default, %arg0_1), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%arg0_1,), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%abs_1,), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%neg,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %sub_4 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%minimum, %log1p), kwargs = {})
# %sub_5 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_5, %sub_4), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, 0.25), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_2, 0.75), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %mul_3), kwargs = {})
# %sigmoid : [num_users=2] = call_function[target=torch.ops.aten.sigmoid.default](args = (%arg0_1,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sigmoid), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %arg1_1), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid, %sub_1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %mul_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%add, 2.0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_1, %pow_1), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_5, %mul_4), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%mul_6,), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mean, 1.0), kwargs = {})
triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0 = async_compile.triton('triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp3 = tl.load(in_ptr1 + (r0), None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = 0.25
tmp14 = tmp0 * tmp13
tmp15 = 0.75
tmp16 = tmp2 * tmp15
tmp17 = tmp14 + tmp16
tmp18 = tl.sigmoid(tmp3)
tmp19 = tmp1 - tmp18
tmp20 = tmp19 * tmp0
tmp21 = tmp18 * tmp2
tmp22 = tmp20 + tmp21
tmp23 = tmp22 * tmp22
tmp24 = tmp17 * tmp23
tmp25 = tmp12 * tmp24
tmp26 = tl.broadcast_to(tmp25, [RBLOCK])
tmp28 = triton_helpers.promote_to_tensor(tl.sum(tmp26, 0))
tmp29 = 256.0
tmp30 = tmp28 / tmp29
tmp31 = tmp30 * tmp1
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp31, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits, mul_2, sub_2, mul_3, add_1, pred_sigmoid, sub, mul, sub_1, mul_1, pt, pow_1, focal_weight, loss, loss_1, loss_cls], Original ATen: [aten.binary_cross_entropy_with_logits, aten.mul, aten.rsub, aten.add, aten.sigmoid, aten.pow, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0.run(buf1, arg1_1, arg0_1, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def sigmoid_focal_loss(pred, target, weight=None, gamma=2.0, alpha=0.25,
reduction='mean', avg_factor=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Dafaults to None.
gamma (float): The gamma for calculating the modulating factor.
Defaults to 2.0.
alpha (float): A balanced form for Focal Loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' ,
loss is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
focal_weight = (alpha * target + (1 - alpha) * (1 - target)) * pt.pow(gamma
)
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class FocalLoss(nn.Module):
"""Focal loss.
Args:
gamma (float): Focusing parameter in focal loss.
Defaults to 2.0.
alpha (float): The parameter in balanced form of focal
loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss into
a scalar. Options are "none" and "mean". Defaults to 'mean'.
loss_weight (float): Weight of loss. Defaults to 1.0.
"""
def __init__(self, gamma=2.0, alpha=0.25, reduction='mean', loss_weight=1.0
):
super(FocalLoss, self).__init__()
self.gamma = gamma
self.alpha = alpha
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, pred, target, weight=None, avg_factor=None,
reduction_override=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction
with shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, \\*). Dafaults to None.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
reduction_override (str, optional): The method used to reduce the
loss into a scalar. Options are "none", "mean" and "sum".
Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
loss_cls = self.loss_weight * sigmoid_focal_loss(pred, target,
weight, gamma=self.gamma, alpha=self.alpha, reduction=reduction,
avg_factor=avg_factor)
return loss_cls
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
import torch.nn.functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0(
in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp3 = tl.load(in_ptr1 + r0, None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = 0.25
tmp14 = tmp0 * tmp13
tmp15 = 0.75
tmp16 = tmp2 * tmp15
tmp17 = tmp14 + tmp16
tmp18 = tl.sigmoid(tmp3)
tmp19 = tmp1 - tmp18
tmp20 = tmp19 * tmp0
tmp21 = tmp18 * tmp2
tmp22 = tmp20 + tmp21
tmp23 = tmp22 * tmp22
tmp24 = tmp17 * tmp23
tmp25 = tmp12 * tmp24
tmp26 = tl.broadcast_to(tmp25, [RBLOCK])
tmp28 = triton_helpers.promote_to_tensor(tl.sum(tmp26, 0))
tmp29 = 256.0
tmp30 = tmp28 / tmp29
tmp31 = tmp30 * tmp1
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp31, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0[
grid(1)](buf1, arg1_1, arg0_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def sigmoid_focal_loss(pred, target, weight=None, gamma=2.0, alpha=0.25,
reduction='mean', avg_factor=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Dafaults to None.
gamma (float): The gamma for calculating the modulating factor.
Defaults to 2.0.
alpha (float): A balanced form for Focal Loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' ,
loss is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
focal_weight = (alpha * target + (1 - alpha) * (1 - target)) * pt.pow(gamma
)
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class FocalLossNew(nn.Module):
"""Focal loss.
Args:
gamma (float): Focusing parameter in focal loss.
Defaults to 2.0.
alpha (float): The parameter in balanced form of focal
loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss into
a scalar. Options are "none" and "mean". Defaults to 'mean'.
loss_weight (float): Weight of loss. Defaults to 1.0.
"""
def __init__(self, gamma=2.0, alpha=0.25, reduction='mean', loss_weight=1.0
):
super(FocalLossNew, self).__init__()
self.gamma = gamma
self.alpha = alpha
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| LiuXiaoxuanPKU/actnn-mmcls | FocalLoss | false | 5,543 | [
"Apache-2.0"
] | 1 | c97d1116d54ddb3f9b1e51baebe25ffb2b3f7b75 | https://github.com/LiuXiaoxuanPKU/actnn-mmcls/tree/c97d1116d54ddb3f9b1e51baebe25ffb2b3f7b75 | import torch
import torch.nn as nn
import torch.nn.functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def sigmoid_focal_loss(pred, target, weight=None, gamma=2.0, alpha=0.25,
reduction='mean', avg_factor=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Dafaults to None.
gamma (float): The gamma for calculating the modulating factor.
Defaults to 2.0.
alpha (float): A balanced form for Focal Loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' ,
loss is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
focal_weight = (alpha * target + (1 - alpha) * (1 - target)) * pt.pow(gamma
)
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class Model(nn.Module):
"""Focal loss.
Args:
gamma (float): Focusing parameter in focal loss.
Defaults to 2.0.
alpha (float): The parameter in balanced form of focal
loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss into
a scalar. Options are "none" and "mean". Defaults to 'mean'.
loss_weight (float): Weight of loss. Defaults to 1.0.
"""
def __init__(self, gamma=2.0, alpha=0.25, reduction='mean', loss_weight=1.0
):
super().__init__()
self.gamma = gamma
self.alpha = alpha
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, pred, target, weight=None, avg_factor=None,
reduction_override=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction
with shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise
# ... truncated (>4000 chars) for memory efficiency |
BasicBlock_IN | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/xk/cxka55ttd2soaycbsbmcsfd6abljife3im4atrw25snc4lgpizpp.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.reflection_pad2d]
# Source node to ATen node mapping:
# x => iota
# Graph fragment:
# %iota : [num_users=2] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
triton_poi_fused_reflection_pad2d_0 = async_compile.triton('triton_poi_fused_reflection_pad2d_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*i64', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0,), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_reflection_pad2d_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_reflection_pad2d_0(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tl.store(out_ptr0 + (x0), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/u3/cu3esk352lfpwhdx5q3cisxtqfmayk6lbllkai24wfn3z3eotxka.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.reflection_pad2d]
# Source node to ATen node mapping:
# x => _unsafe_index, _unsafe_index_1
# Graph fragment:
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%primals_1, [None, None, %sub_1, None]), kwargs = {})
# %_unsafe_index_1 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index, [None, None, None, %sub_1]), kwargs = {})
triton_poi_fused_reflection_pad2d_1 = async_compile.triton('triton_poi_fused_reflection_pad2d_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_reflection_pad2d_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_reflection_pad2d_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4) % 4
x2 = (xindex // 16)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + ((-1)*(tl_math.abs((-3) + x0))) + ((-4)*(tl_math.abs((-3) + x1))) + (16*x2)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/u2/cu2h7kepp2vussiw5mawdeynnancc7laf3dn7xmj6m66txxk463j.py
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add
# mv => mul, sum_1
# norm => pow_1, pow_2, sum_2
# truediv => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute, %primals_2), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, None), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 1e-12), kwargs = {})
# %div : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_2 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_2(in_ptr0, in_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (4 + r0), None)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (8 + r0), None)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (12 + r0), None)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr2 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp26, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dd/cdd6hg7bodqqsl6rt7hkho7fppp37iuwsdghsonkdm35yx4beecq.py
# Topologically Sorted Source Nodes: [mv_1, norm_1, add_1, truediv_1, sigma], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div, aten.dot]
# Source node to ATen node mapping:
# add_1 => add_1
# mv_1 => mul_1, sum_3
# norm_1 => pow_3, pow_4, sum_4
# sigma => mul_3, sum_6
# truediv_1 => div_1
# Graph fragment:
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %div), kwargs = {})
# %sum_3 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [1]), kwargs = {})
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, None), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_4, 0.5), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_4, 1e-12), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, %add_1), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, %sum_3), kwargs = {})
# %sum_6 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%mul_3,), kwargs = {})
triton_per_fused_add_div_dot_linalg_vector_norm_mv_3 = async_compile.triton('triton_per_fused_add_div_dot_linalg_vector_norm_mv_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {4: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_dot_linalg_vector_norm_mv_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_dot_linalg_vector_norm_mv_3(in_ptr0, in_ptr1, out_ptr2, out_ptr3, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (4*r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (1 + (4*r0)), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (2 + (4*r0)), None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (3 + (4*r0)), None, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tmp27 = tmp26 * tmp18
tmp28 = tl.broadcast_to(tmp27, [XBLOCK, RBLOCK])
tmp30 = tl.sum(tmp28, 1)[:, None]
tl.store(out_ptr3 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp26, None)
tl.store(out_ptr2 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp30, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/iy/ciyfrn3h65grkri5tse22oq6oz24zqdxpthkaxnywq5nlwwcu426.py
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_2 => div_2
# Graph fragment:
# %div_2 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_4, %expand), kwargs = {})
triton_poi_fused_div_4 = async_compile.triton('triton_poi_fused_div_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/j2/cj25y5kypks7hldzyxgj75p5rupfp4lc5cvxsqotcl5sq3q2acok.py
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten._native_batch_norm_legit]
# Source node to ATen node mapping:
# x_2 => add_2, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view_3, [0, 2, 3]), kwargs = {correction: 0, keepdim: True})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
triton_per_fused__native_batch_norm_legit_5 = async_compile.triton('triton_per_fused__native_batch_norm_legit_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[16, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__native_batch_norm_legit_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 4, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__native_batch_norm_legit_5(in_out_ptr0, in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 16
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (16*x0)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.full([XBLOCK, 1], 16, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = tl.where(xmask, tmp13, 0)
tmp16 = tl.sum(tmp15, 1)[:, None]
tmp17 = 16.0
tmp18 = tmp16 / tmp17
tmp19 = 1e-05
tmp20 = tmp18 + tmp19
tmp21 = libdevice.rsqrt(tmp20)
tl.debug_barrier()
tl.store(in_out_ptr0 + (x0), tmp21, xmask)
tl.store(out_ptr0 + (x0), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/z5/cz5isnfhua7uwsnx7hxjka3axj4urd3jqvilqt34xz47cbtq74ao.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.reflection_pad2d]
# Source node to ATen node mapping:
# x_4 => _unsafe_index_2, _unsafe_index_3
# Graph fragment:
# %_unsafe_index_2 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%view_6, [None, None, %sub_6, None]), kwargs = {})
# %_unsafe_index_3 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index_2, [None, None, None, %sub_6]), kwargs = {})
triton_poi_fused_reflection_pad2d_6 = async_compile.triton('triton_poi_fused_reflection_pad2d_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_reflection_pad2d_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_reflection_pad2d_6(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = (xindex // 6) % 6
x2 = (xindex // 36)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + ((-1)*(tl_math.abs((-3) + (tl_math.abs((-1) + x0))))) + ((-4)*(tl_math.abs((-3) + (tl_math.abs((-1) + x1))))) + (16*x2)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x2), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = tmp4 > tmp5
tmp7 = 0.2
tmp8 = tmp4 * tmp7
tmp9 = tl.where(tmp6, tmp4, tmp8)
tl.store(out_ptr0 + (x3), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/7q/c7qakexxwln2kazpcla524ogq3z3pzbf5zjqmylyo6xhvnxv26in.py
# Topologically Sorted Source Nodes: [mv_3, norm_2, add_2, truediv_3], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_2 => add_3
# mv_3 => mul_6, sum_7
# norm_2 => pow_5, pow_6, sum_8
# truediv_3 => div_3
# Graph fragment:
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_1, %primals_5), kwargs = {})
# %sum_7 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_6, [1]), kwargs = {})
# %pow_5 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_7, 2), kwargs = {})
# %sum_8 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_5, None), kwargs = {})
# %pow_6 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_8, 0.5), kwargs = {})
# %add_3 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_6, 1e-12), kwargs = {})
# %div_3 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_7, %add_3), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_mv_7 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_mv_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {5: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=(5,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_mv_7', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_7(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), rmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (36 + r0), rmask, other=0.0)
tmp5 = tl.load(in_ptr1 + (1))
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (72 + r0), rmask, other=0.0)
tmp10 = tl.load(in_ptr1 + (2))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (108 + r0), rmask, other=0.0)
tmp15 = tl.load(in_ptr1 + (3))
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = libdevice.sqrt(tmp23)
tmp25 = 1e-12
tmp26 = tmp24 + tmp25
tmp27 = tmp18 / tmp26
tl.store(out_ptr0 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp18, rmask)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp26, None)
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp27, rmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/cg/ccgxk6hsvkfuxxj43enqcdea57y4x5rki37xnq6pkpids7y5n7rw.py
# Topologically Sorted Source Nodes: [truediv_3, mv_4], Original ATen: [aten.div, aten.mv]
# Source node to ATen node mapping:
# mv_4 => mul_7, sum_9
# truediv_3 => div_3
# Graph fragment:
# %div_3 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_7, %add_3), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_8, %div_3), kwargs = {})
# %sum_9 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_7, [1]), kwargs = {})
triton_per_fused_div_mv_8 = async_compile.triton('triton_per_fused_div_mv_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_mv_8', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_mv_8(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (36*x0)), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1), rmask, eviction_policy='evict_last', other=0.0)
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(rmask & xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/xy/cxyvlheoomg3phbyiazvdpun3iy3baokavqqz6edie2evohlyp4h.py
# Topologically Sorted Source Nodes: [norm_3, add_3, truediv_4], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
# Source node to ATen node mapping:
# add_3 => add_4
# norm_3 => pow_7, pow_8, sum_10
# truediv_4 => div_4
# Graph fragment:
# %pow_7 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_9, 2), kwargs = {})
# %sum_10 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_7, None), kwargs = {})
# %pow_8 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_10, 0.5), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_8, 1e-12), kwargs = {})
# %div_4 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_9, %add_4), kwargs = {})
triton_per_fused_add_div_linalg_vector_norm_9 = async_compile.triton('triton_per_fused_add_div_linalg_vector_norm_9', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_linalg_vector_norm_9', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_9(in_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + (tl.broadcast_to(r0, [XBLOCK, RBLOCK])), tmp8, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rr/crr6tv3ypuiw25pt55zuvsw3hjc35kiyz2jbsaoih3lleqjyzlaw.py
# Topologically Sorted Source Nodes: [sigma_1], Original ATen: [aten.dot]
# Source node to ATen node mapping:
# sigma_1 => mul_9, sum_12
# Graph fragment:
# %mul_9 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_4, %sum_9), kwargs = {})
# %sum_12 : [num_users=2] = call_function[target=torch.ops.aten.sum.default](args = (%mul_9,), kwargs = {})
triton_per_fused_dot_10 = async_compile.triton('triton_per_fused_dot_10', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_dot_10', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_dot_10(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/6k/c6k5avjbseb37trg5jk5ln36xmokh6sfxm5ccsm3zpxtlrhgun64.py
# Topologically Sorted Source Nodes: [truediv_5], Original ATen: [aten.div]
# Source node to ATen node mapping:
# truediv_5 => div_5
# Graph fragment:
# %div_5 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_7, %expand_1), kwargs = {})
triton_poi_fused_div_11 = async_compile.triton('triton_poi_fused_div_11', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_11', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_11(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5d/c5ddtjdheky6atm26rexxirggj3mn2xfjk6rvqazcv7mhetauk5r.py
# Topologically Sorted Source Nodes: [x_6, out, out_1], Original ATen: [aten._native_batch_norm_legit, aten.add, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# out => add_6
# out_1 => relu
# x_6 => add_5, rsqrt_1, var_mean_1
# Graph fragment:
# %var_mean_1 : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view_11, [0, 2, 3]), kwargs = {correction: 0, keepdim: True})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_2, 1e-05), kwargs = {})
# %rsqrt_1 : [num_users=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_5,), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_14, %primals_1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_6,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12 = async_compile.triton('triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[16, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*i1', 6: 'i32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 4, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 16
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (16*x0)), xmask, other=0.0)
tmp29 = tl.load(in_ptr1 + (r1 + (16*x0)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.full([XBLOCK, 1], 16, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = tl.where(xmask, tmp13, 0)
tmp16 = tl.sum(tmp15, 1)[:, None]
tmp17 = 16.0
tmp18 = tmp16 / tmp17
tmp19 = 1e-05
tmp20 = tmp18 + tmp19
tmp21 = libdevice.rsqrt(tmp20)
tmp22 = tmp0 - tmp10
tmp23 = tmp22 * tmp21
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp30 = tmp28 + tmp29
tmp31 = tl.full([1, 1], 0, tl.int32)
tmp32 = triton_helpers.maximum(tmp31, tmp30)
tmp33 = tmp32 <= tmp24
tl.debug_barrier()
tl.store(in_out_ptr0 + (x0), tmp21, xmask)
tl.store(out_ptr1 + (r1 + (16*x0)), tmp32, xmask)
tl.store(out_ptr2 + (r1 + (16*x0)), tmp33, xmask)
tl.store(out_ptr0 + (x0), tmp10, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (36, ), (1, ))
assert_size_stride(primals_7, (4, 4, 3, 3), (36, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.reflection_pad2d]
stream0 = get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0.run(buf0, 4, grid=grid(4), stream=stream0)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.reflection_pad2d]
triton_poi_fused_reflection_pad2d_1.run(primals_1, buf1, 256, grid=grid(256), stream=stream0)
buf4 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv, norm, add, truediv], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_mv_2.run(primals_4, primals_2, buf4, 1, 4, grid=grid(1), stream=stream0)
buf7 = empty_strided_cuda((), (), torch.float32)
buf30 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv_1, norm_1, add_1, truediv_1, sigma], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div, aten.dot]
triton_per_fused_add_div_dot_linalg_vector_norm_mv_3.run(primals_4, buf4, buf7, buf30, 1, 4, grid=grid(1), stream=stream0)
buf8 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_2], Original ATen: [aten.div]
triton_poi_fused_div_4.run(primals_4, buf7, buf8, 16, grid=grid(16), stream=stream0)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.convolution]
buf9 = extern_kernels.convolution(buf1, buf8, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 4, 4, 4), (64, 16, 4, 1))
buf10 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 16, 16), torch.float32)
buf13 = reinterpret_tensor(buf11, (1, 16, 1, 1), (16, 1, 1, 1), 0); del buf11 # reuse
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten._native_batch_norm_legit]
triton_per_fused__native_batch_norm_legit_5.run(buf13, buf9, buf10, 16, 16, grid=grid(16), stream=stream0)
buf14 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.reflection_pad2d]
triton_poi_fused_reflection_pad2d_6.run(buf9, buf10, buf13, buf14, 576, grid=grid(576), stream=stream0)
buf15 = empty_strided_cuda((36, ), (1, ), torch.float32)
buf16 = buf7; del buf7 # reuse
buf17 = buf16; del buf16 # reuse
buf44 = empty_strided_cuda((36, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [mv_3, norm_2, add_2, truediv_3], Original ATen: [aten.mv, aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_mv_7.run(buf17, primals_7, primals_5, buf15, buf44, 1, 36, grid=grid(1), stream=stream0)
buf18 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [truediv_3, mv_4], Original ATen: [aten.div, aten.mv]
triton_per_fused_div_mv_8.run(primals_7, buf15, buf17, buf18, 4, 36, grid=grid(4), stream=stream0)
buf20 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [norm_3, add_3, truediv_4], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
triton_per_fused_add_div_linalg_vector_norm_9.run(buf18, buf20, 1, 4, grid=grid(1), stream=stream0)
buf21 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sigma_1], Original ATen: [aten.dot]
triton_per_fused_dot_10.run(buf20, buf18, buf21, 1, 4, grid=grid(1), stream=stream0)
del buf18
buf22 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [truediv_5], Original ATen: [aten.div]
triton_poi_fused_div_11.run(primals_7, buf21, buf22, 144, grid=grid(144), stream=stream0)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
buf23 = extern_kernels.convolution(buf14, buf22, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf23, (4, 4, 4, 4), (64, 16, 4, 1))
buf24 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 1, 1), torch.float32)
buf25 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 16, 16), torch.float32)
buf27 = reinterpret_tensor(buf25, (1, 16, 1, 1), (16, 1, 1, 1), 0); del buf25 # reuse
buf28 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf29 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_6, out, out_1], Original ATen: [aten._native_batch_norm_legit, aten.add, aten.relu, aten.threshold_backward]
triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12.run(buf27, buf23, primals_1, buf24, buf28, buf29, 16, 16, grid=grid(16), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [norm_1, add_1, truediv_1], Original ATen: [aten.linalg_vector_norm, aten.add, aten.div]
buf31 = torch.ops.aten.set_.source_Tensor(primals_2, buf30)
assert_size_stride(buf31, (4, ), (1, ))
del primals_2
# Topologically Sorted Source Nodes: [], Original ATen: []
buf35 = torch.ops.aten.set_.source_Tensor(primals_3, buf4)
assert_size_stride(buf35, (4, ), (1, ))
del primals_3
# Topologically Sorted Source Nodes: [], Original ATen: []
buf39 = torch.ops.aten.set_.source_Tensor(primals_5, buf20)
assert_size_stride(buf39, (4, ), (1, ))
del buf15
# Topologically Sorted Source Nodes: [truediv_3], Original ATen: [aten.div]
buf45 = torch.ops.aten.set_.source_Tensor(primals_6, buf44)
assert_size_stride(buf45, (36, ), (1, ))
del primals_6
return (buf28, buf8, buf22, primals_4, primals_5, primals_7, buf0, buf1, buf4, buf8, buf9, buf10, buf13, buf14, buf17, buf20, buf21, buf22, buf23, buf24, buf27, buf29, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((36, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class BasicBlock_IN(nn.Module):
def __init__(self, inplanes, planes, stride=1, first=False):
super(BasicBlock_IN, self).__init__()
self.expansion = 4
self.stride = stride
self.first = first
self.conv1 = Conv2dLayer(inplanes, planes, 1, stride, 0, pad_type=
'reflect', activation='lrelu', norm='in')
self.conv2 = Conv2dLayer(planes, planes, 3, 1, 1, pad_type=
'reflect', activation='lrelu', norm='in')
if stride != 1 or first is True:
self.downsample = Conv2dLayer(inplanes, planes * self.expansion,
1, stride, 0, pad_type='reflect', activation='lrelu', norm='bn'
)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
identity = x
out = self.conv1(x)
out = self.conv2(out)
if self.stride != 1 or self.first is True:
identity = self.downsample(x)
out = out + identity
out = self.relu(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'inplanes': 4, 'planes': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_reflection_pad2d_0(out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tl.store(out_ptr0 + x0, tmp0, xmask)
@triton.jit
def triton_poi_fused_reflection_pad2d_1(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4 % 4
x2 = xindex // 16
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + -1 * tl_math.abs(-3 + x0) + -4 * tl_math
.abs(-3 + x1) + 16 * x2), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_2(in_ptr0, in_ptr1,
out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (4 + r0), None)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (8 + r0), None)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (12 + r0), None)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tl.store(out_ptr2 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp26, None)
@triton.jit
def triton_per_fused_add_div_dot_linalg_vector_norm_mv_3(in_ptr0, in_ptr1,
out_ptr2, out_ptr3, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + 4 * r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (1 + 4 * r0), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (2 + 4 * r0), None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (3 + 4 * r0), None, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.sum(tmp20, 1)[:, None]
tmp23 = libdevice.sqrt(tmp22)
tmp24 = 1e-12
tmp25 = tmp23 + tmp24
tmp26 = tmp18 / tmp25
tmp27 = tmp26 * tmp18
tmp28 = tl.broadcast_to(tmp27, [XBLOCK, RBLOCK])
tmp30 = tl.sum(tmp28, 1)[:, None]
tl.store(out_ptr3 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp26, None)
tl.store(out_ptr2 + tl.full([XBLOCK, 1], 0, tl.int32), tmp30, None)
@triton.jit
def triton_poi_fused_div_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_per_fused__native_batch_norm_legit_5(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 16 * x0), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tl.where(xmask, tmp1, 0)
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.full([XBLOCK, 1], 16, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = tl.where(xmask, tmp13, 0)
tmp16 = tl.sum(tmp15, 1)[:, None]
tmp17 = 16.0
tmp18 = tmp16 / tmp17
tmp19 = 1e-05
tmp20 = tmp18 + tmp19
tmp21 = libdevice.rsqrt(tmp20)
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp21, xmask)
tl.store(out_ptr0 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused_reflection_pad2d_6(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = xindex // 6 % 6
x2 = xindex // 36
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + -1 * tl_math.abs(-3 + tl_math.abs(-1 +
x0)) + -4 * tl_math.abs(-3 + tl_math.abs(-1 + x1)) + 16 * x2),
xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x2, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = tmp4 > tmp5
tmp7 = 0.2
tmp8 = tmp4 * tmp7
tmp9 = tl.where(tmp6, tmp4, tmp8)
tl.store(out_ptr0 + x3, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_mv_7(in_out_ptr0, in_ptr0,
in_ptr1, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, rmask, other=0.0)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.load(in_ptr0 + (36 + r0), rmask, other=0.0)
tmp5 = tl.load(in_ptr1 + 1)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp9 = tl.load(in_ptr0 + (72 + r0), rmask, other=0.0)
tmp10 = tl.load(in_ptr1 + 2)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp14 = tl.load(in_ptr0 + (108 + r0), rmask, other=0.0)
tmp15 = tl.load(in_ptr1 + 3)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp3 = tmp0 * tmp2
tmp7 = tmp4 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp9 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp14 * tmp16
tmp18 = tmp13 + tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = libdevice.sqrt(tmp23)
tmp25 = 1e-12
tmp26 = tmp24 + tmp25
tmp27 = tmp18 / tmp26
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp18, rmask)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp26, None)
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp27, rmask)
@triton.jit
def triton_per_fused_div_mv_8(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 36 * x0), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + r1, rmask, eviction_policy='evict_last', other=0.0
)
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4 = tmp1 / tmp3
tmp5 = tmp0 * tmp4
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(rmask & xmask, tmp6, 0)
tmp9 = tl.sum(tmp8, 1)[:, None]
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_per_fused_add_div_linalg_vector_norm_9(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp6 = 1e-12
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp8, None)
@triton.jit
def triton_per_fused_dot_10(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
@triton.jit
def triton_poi_fused_div_11(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 / tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12(
in_out_ptr0, in_ptr0, in_ptr1, out_ptr0, out_ptr1, out_ptr2, xnumel,
rnumel, XBLOCK: tl.constexpr):
xnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 16 * x0), xmask, other=0.0)
tmp29 = tl.load(in_ptr1 + (r1 + 16 * x0), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tl.where(xmask, tmp1, 0)
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.full([XBLOCK, 1], 16, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = tl.where(xmask, tmp13, 0)
tmp16 = tl.sum(tmp15, 1)[:, None]
tmp17 = 16.0
tmp18 = tmp16 / tmp17
tmp19 = 1e-05
tmp20 = tmp18 + tmp19
tmp21 = libdevice.rsqrt(tmp20)
tmp22 = tmp0 - tmp10
tmp23 = tmp22 * tmp21
tmp24 = 0.0
tmp25 = tmp23 > tmp24
tmp26 = 0.2
tmp27 = tmp23 * tmp26
tmp28 = tl.where(tmp25, tmp23, tmp27)
tmp30 = tmp28 + tmp29
tmp31 = tl.full([1, 1], 0, tl.int32)
tmp32 = triton_helpers.maximum(tmp31, tmp30)
tmp33 = tmp32 <= tmp24
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp21, xmask)
tl.store(out_ptr1 + (r1 + 16 * x0), tmp32, xmask)
tl.store(out_ptr2 + (r1 + 16 * x0), tmp33, xmask)
tl.store(out_ptr0 + x0, tmp10, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (36,), (1,))
assert_size_stride(primals_7, (4, 4, 3, 3), (36, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4,), (1,), torch.int64)
get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0[grid(4)](buf0, 4, XBLOCK=4,
num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_reflection_pad2d_1[grid(256)](primals_1, buf1, 256,
XBLOCK=128, num_warps=4, num_stages=1)
buf4 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_mv_2[grid(1)](primals_4,
primals_2, buf4, 1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf7 = empty_strided_cuda((), (), torch.float32)
buf30 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_dot_linalg_vector_norm_mv_3[grid(1)](primals_4
, buf4, buf7, buf30, 1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf8 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 1, 1), torch.float32)
triton_poi_fused_div_4[grid(16)](primals_4, buf7, buf8, 16, XBLOCK=
16, num_warps=1, num_stages=1)
buf9 = extern_kernels.convolution(buf1, buf8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 4, 4, 4), (64, 16, 4, 1))
buf10 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 16, 16), torch.
float32)
buf13 = reinterpret_tensor(buf11, (1, 16, 1, 1), (16, 1, 1, 1), 0)
del buf11
triton_per_fused__native_batch_norm_legit_5[grid(16)](buf13, buf9,
buf10, 16, 16, XBLOCK=1, num_warps=2, num_stages=1)
buf14 = empty_strided_cuda((4, 4, 6, 6), (144, 36, 6, 1), torch.float32
)
triton_poi_fused_reflection_pad2d_6[grid(576)](buf9, buf10, buf13,
buf14, 576, XBLOCK=128, num_warps=4, num_stages=1)
buf15 = empty_strided_cuda((36,), (1,), torch.float32)
buf16 = buf7
del buf7
buf17 = buf16
del buf16
buf44 = empty_strided_cuda((36,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_mv_7[grid(1)](buf17,
primals_7, primals_5, buf15, buf44, 1, 36, XBLOCK=1, num_warps=
2, num_stages=1)
buf18 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_div_mv_8[grid(4)](primals_7, buf15, buf17, buf18,
4, 36, XBLOCK=1, num_warps=2, num_stages=1)
buf20 = empty_strided_cuda((4,), (1,), torch.float32)
triton_per_fused_add_div_linalg_vector_norm_9[grid(1)](buf18, buf20,
1, 4, XBLOCK=1, num_warps=2, num_stages=1)
buf21 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_dot_10[grid(1)](buf20, buf18, buf21, 1, 4, XBLOCK=
1, num_warps=2, num_stages=1)
del buf18
buf22 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
triton_poi_fused_div_11[grid(144)](primals_7, buf21, buf22, 144,
XBLOCK=128, num_warps=4, num_stages=1)
buf23 = extern_kernels.convolution(buf14, buf22, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf23, (4, 4, 4, 4), (64, 16, 4, 1))
buf24 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 1, 1), torch.float32)
buf25 = empty_strided_cuda((1, 16, 1, 1), (16, 1, 16, 16), torch.
float32)
buf27 = reinterpret_tensor(buf25, (1, 16, 1, 1), (16, 1, 1, 1), 0)
del buf25
buf28 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf29 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_per_fused__native_batch_norm_legit_add_relu_threshold_backward_12[
grid(16)](buf27, buf23, primals_1, buf24, buf28, buf29, 16, 16,
XBLOCK=1, num_warps=2, num_stages=1)
del primals_1
buf31 = torch.ops.aten.set_.source_Tensor(primals_2, buf30)
assert_size_stride(buf31, (4,), (1,))
del primals_2
buf35 = torch.ops.aten.set_.source_Tensor(primals_3, buf4)
assert_size_stride(buf35, (4,), (1,))
del primals_3
buf39 = torch.ops.aten.set_.source_Tensor(primals_5, buf20)
assert_size_stride(buf39, (4,), (1,))
del buf15
buf45 = torch.ops.aten.set_.source_Tensor(primals_6, buf44)
assert_size_stride(buf45, (36,), (1,))
del primals_6
return (buf28, buf8, buf22, primals_4, primals_5, primals_7, buf0, buf1,
buf4, buf8, buf9, buf10, buf13, buf14, buf17, buf20, buf21, buf22,
buf23, buf24, buf27, buf29)
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super(LayerNorm, self).__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super(SpectralNorm, self).__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super(Conv2dLayer, self).__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.activation = nn.LeakyReLU(0.2, inplace=True)
elif activation == 'prelu':
self.activation = nn.PReLU()
elif activation == 'selu':
self.activation = nn.SELU(inplace=True)
elif activation == 'tanh':
self.activation = nn.Tanh()
elif activation == 'sigmoid':
self.activation = nn.Sigmoid()
elif activation == 'none':
self.activation = None
else:
assert 0, 'Unsupported activation: {}'.format(activation)
if sn:
self.conv2d = SpectralNorm(nn.Conv2d(in_channels, out_channels,
kernel_size, stride, padding=0, dilation=dilation, bias=False))
else:
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=0, dilation=dilation, bias=False)
def forward(self, x):
x = self.pad(x)
x = self.conv2d(x)
if self.norm:
x = self.norm(x)
if self.activation:
x = self.activation(x)
return x
class BasicBlock_INNew(nn.Module):
def __init__(self, inplanes, planes, stride=1, first=False):
super(BasicBlock_INNew, self).__init__()
self.expansion = 4
self.stride = stride
self.first = first
self.conv1 = Conv2dLayer(inplanes, planes, 1, stride, 0, pad_type=
'reflect', activation='lrelu', norm='in')
self.conv2 = Conv2dLayer(planes, planes, 3, 1, 1, pad_type=
'reflect', activation='lrelu', norm='in')
if stride != 1 or first is True:
self.downsample = Conv2dLayer(inplanes, planes * self.expansion,
1, stride, 0, pad_type='reflect', activation='lrelu', norm='bn'
)
self.relu = nn.ReLU(inplace=True)
def forward(self, input_0):
primals_2 = self.conv1.conv2d.module.weight_u
primals_3 = self.conv1.conv2d.module.weight_v
primals_4 = self.conv1.conv2d.module.weight_bar
primals_5 = self.conv2.conv2d.module.weight_u
primals_6 = self.conv2.conv2d.module.weight_v
primals_7 = self.conv2.conv2d.module.weight_bar
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| LenKerr/Semantic-Colorization-GAN | BasicBlock_IN | false | 5,544 | [
"MIT"
] | 1 | 2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | https://github.com/LenKerr/Semantic-Colorization-GAN/tree/2ce52406ca6fc92e69692b451b1c9ae66ba3b76f | import torch
import torch.nn as nn
from torch.nn import Parameter
def l2normalize(v, eps=1e-12):
return v / (v.norm() + eps)
class LayerNorm(nn.Module):
def __init__(self, num_features, eps=1e-08, affine=True):
super().__init__()
self.num_features = num_features
self.affine = affine
self.eps = eps
if self.affine:
self.gamma = Parameter(torch.Tensor(num_features).uniform_())
self.beta = Parameter(torch.zeros(num_features))
def forward(self, x):
shape = [-1] + [1] * (x.dim() - 1)
if x.size(0) == 1:
mean = x.view(-1).mean().view(*shape)
std = x.view(-1).std().view(*shape)
else:
mean = x.view(x.size(0), -1).mean(1).view(*shape)
std = x.view(x.size(0), -1).std(1).view(*shape)
x = (x - mean) / (std + self.eps)
if self.affine:
shape = [1, -1] + [1] * (x.dim() - 2)
x = x * self.gamma.view(*shape) + self.beta.view(*shape)
return x
class SpectralNorm(nn.Module):
def __init__(self, module, name='weight', power_iterations=1):
super().__init__()
self.module = module
self.name = name
self.power_iterations = power_iterations
if not self._made_params():
self._make_params()
def _update_u_v(self):
u = getattr(self.module, self.name + '_u')
v = getattr(self.module, self.name + '_v')
w = getattr(self.module, self.name + '_bar')
height = w.data.shape[0]
for _ in range(self.power_iterations):
v.data = l2normalize(torch.mv(torch.t(w.view(height, -1).data),
u.data))
u.data = l2normalize(torch.mv(w.view(height, -1).data, v.data))
sigma = u.dot(w.view(height, -1).mv(v))
setattr(self.module, self.name, w / sigma.expand_as(w))
def _made_params(self):
try:
getattr(self.module, self.name + '_u')
getattr(self.module, self.name + '_v')
getattr(self.module, self.name + '_bar')
return True
except AttributeError:
return False
def _make_params(self):
w = getattr(self.module, self.name)
height = w.data.shape[0]
width = w.view(height, -1).data.shape[1]
u = Parameter(w.data.new(height).normal_(0, 1), requires_grad=False)
v = Parameter(w.data.new(width).normal_(0, 1), requires_grad=False)
u.data = l2normalize(u.data)
v.data = l2normalize(v.data)
w_bar = Parameter(w.data)
del self.module._parameters[self.name]
self.module.register_parameter(self.name + '_u', u)
self.module.register_parameter(self.name + '_v', v)
self.module.register_parameter(self.name + '_bar', w_bar)
def forward(self, *args):
self._update_u_v()
return self.module.forward(*args)
class Conv2dLayer(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1,
padding=0, dilation=1, pad_type='zero', activation='lrelu', norm=
'none', sn=True):
super().__init__()
if pad_type == 'reflect':
self.pad = nn.ReflectionPad2d(padding)
elif pad_type == 'replicate':
self.pad = nn.ReplicationPad2d(padding)
elif pad_type == 'zero':
self.pad = nn.ZeroPad2d(padding)
else:
assert 0, 'Unsupported padding type: {}'.format(pad_type)
if norm == 'bn':
self.norm = nn.BatchNorm2d(out_channels)
elif norm == 'in':
self.norm = nn.InstanceNorm2d(out_channels)
elif norm == 'ln':
self.norm = LayerNorm(out_channels)
elif norm == 'none':
self.norm = None
else:
assert 0, 'Unsupported normalization: {}'.format(norm)
if activation == 'relu':
self.activation = nn.ReLU(inplace=True)
elif activation == 'lrelu':
self.acti
# ... truncated (>4000 chars) for memory efficiency |
Downsample | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/cu/ccutvo2v4333pq6xhrg2zryqqwthm7dmmuqprvva2xdwiodpz5jn.py
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# conv2d => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_2, %primals_3, [2, 2], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_0 = async_compile.triton('triton_poi_fused_convolution_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 4) % 4
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x3), tmp2, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_1, primals_2, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 2, 2), (16, 4, 2, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_0.run(buf1, primals_3, 64, grid=grid(64), stream=stream0)
del primals_3
return (buf1, primals_1, primals_2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
def avg_pool_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D average pooling module.
"""
if dims == 1:
return nn.AvgPool1d(*args, **kwargs)
elif dims == 2:
return nn.AvgPool2d(*args, **kwargs)
elif dims == 3:
return nn.AvgPool3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class Downsample(nn.Module):
"""
A downsampling layer with an optional convolution.
:param channels: channels in the inputs and outputs.
:param use_conv: a bool determining if a convolution is applied.
:param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then
downsampling occurs in the inner-two dimensions.
"""
def __init__(self, channels, use_conv, dims=2, out_channels=None):
super().__init__()
self.channels = channels
self.out_channels = out_channels or channels
self.use_conv = use_conv
self.dims = dims
stride = 2 if dims != 3 else (1, 2, 2)
if use_conv:
self.op = conv_nd(dims, self.channels, self.out_channels, 3,
stride=stride, padding=1)
else:
assert self.channels == self.out_channels
self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride)
def forward(self, x):
assert x.shape[1] == self.channels
return self.op(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'channels': 4, 'use_conv': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_1, primals_2, stride=(2,
2), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 2, 2), (16, 4, 2, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(64)](buf1, primals_3, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_3
return buf1, primals_1, primals_2
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
def avg_pool_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D average pooling module.
"""
if dims == 1:
return nn.AvgPool1d(*args, **kwargs)
elif dims == 2:
return nn.AvgPool2d(*args, **kwargs)
elif dims == 3:
return nn.AvgPool3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class DownsampleNew(nn.Module):
"""
A downsampling layer with an optional convolution.
:param channels: channels in the inputs and outputs.
:param use_conv: a bool determining if a convolution is applied.
:param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then
downsampling occurs in the inner-two dimensions.
"""
def __init__(self, channels, use_conv, dims=2, out_channels=None):
super().__init__()
self.channels = channels
self.out_channels = out_channels or channels
self.use_conv = use_conv
self.dims = dims
stride = 2 if dims != 3 else (1, 2, 2)
if use_conv:
self.op = conv_nd(dims, self.channels, self.out_channels, 3,
stride=stride, padding=1)
else:
assert self.channels == self.out_channels
self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride)
def forward(self, input_0):
primals_2 = self.op.weight
primals_3 = self.op.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| Liujingxiu23/guided-diffusion | Downsample | false | 5,545 | [
"MIT"
] | 1 | 0ba878e517b276c45d1195eb29f6f5f72659a05b | https://github.com/Liujingxiu23/guided-diffusion/tree/0ba878e517b276c45d1195eb29f6f5f72659a05b | import torch
import torch.nn as nn
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
def avg_pool_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D average pooling module.
"""
if dims == 1:
return nn.AvgPool1d(*args, **kwargs)
elif dims == 2:
return nn.AvgPool2d(*args, **kwargs)
elif dims == 3:
return nn.AvgPool3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class Model(nn.Module):
"""
A downsampling layer with an optional convolution.
:param channels: channels in the inputs and outputs.
:param use_conv: a bool determining if a convolution is applied.
:param dims: determines if the signal is 1D, 2D, or 3D. If 3D, then
downsampling occurs in the inner-two dimensions.
"""
def __init__(self, channels, use_conv, dims=2, out_channels=None):
super().__init__()
self.channels = channels
self.out_channels = out_channels or channels
self.use_conv = use_conv
self.dims = dims
stride = 2 if dims != 3 else (1, 2, 2)
if use_conv:
self.op = conv_nd(dims, self.channels, self.out_channels, 3,
stride=stride, padding=1)
else:
assert self.channels == self.out_channels
self.op = avg_pool_nd(dims, kernel_size=stride, stride=stride)
def forward(self, x):
assert x.shape[1] == self.channels
return self.op(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
AdditiveAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/vu/cvuv73ijvl45rys2kvk2dud7shg4nznn622fzyldpf7pmxppx3o5.py
# Topologically Sorted Source Nodes: [tanh], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# tanh => tanh
# Graph fragment:
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%view_1,), kwargs = {})
triton_poi_fused_tanh_0 = async_compile.triton('triton_poi_fused_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_tanh_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + (x2), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dm/cdmkcxuzpnailvibeivaikqdr4zvashgzwju7qijhq5aizlo3aor.py
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weights => amax, clone, exp, sub
# Graph fragment:
# %clone : [num_users=2] = call_function[target=torch.ops.aten.clone.default](args = (%permute_2,), kwargs = {memory_format: torch.contiguous_format})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%clone, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clone, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 4
x2 = (xindex // 16)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (16*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (4 + x0 + (16*x2)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (8 + x0 + (16*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (12 + x0 + (16*x2)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x3), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ot/cotghubvxq6i3kdxijgdhem4ntfmdc5cn3al7gv2kuqi2pbqiyxe.py
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# att_weights => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_2 = async_compile.triton('triton_poi_fused__softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (y0 + (16*y1)), ymask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (4 + y0 + (16*y1)), ymask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (8 + y0 + (16*y1)), ymask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (12 + y0 + (16*y1)), ymask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2 + (4*y3)), tmp8, xmask & ymask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [tanh], Original ATen: [aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_tanh_0.run(buf1, primals_2, 64, grid=grid(64), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [add_score], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4, 4), (16, 1, 4), torch.float32)
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf2, buf3, 64, grid=grid(64), stream=stream0)
buf4 = reinterpret_tensor(buf2, (4, 4, 4), (16, 4, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [att_weights], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf3, buf4, 16, 4, grid=grid(16, 4), stream=stream0)
buf5 = reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [att_out], Original ATen: [aten.bmm]
extern_kernels.bmm(buf4, primals_3, out=buf5)
return (buf5, primals_3, buf1, buf4, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class AdditiveAttention(nn.Module):
def __init__(self, in_features, att_hidden, out_features, bias=True):
super(AdditiveAttention, self).__init__()
self.out_size = out_features
self.linear1 = nn.Linear(in_features=in_features, out_features=
att_hidden, bias=bias)
self.tanh = nn.Tanh()
self.linear2 = nn.Linear(in_features=att_hidden, out_features=
out_features, bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, inputs, mask=None):
"""
:param inputs: (bz, seq_len, in_features)
:param mask: (bz, seq_len) 填充为0
:return:
"""
add_score = self.linear2(self.tanh(self.linear1(inputs)))
add_score = add_score.transpose(1, 2)
if mask is not None:
pad_mask = mask == 0
add_score = add_score.masked_fill(pad_mask[:, None, :], -
1000000000.0)
att_weights = self.softmax(add_score)
att_out = torch.bmm(att_weights, inputs)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'att_hidden': 4, 'out_features': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 4
x2 = xindex // 16
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (4 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (8 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (12 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x3, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK:
tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (y0 + 16 * y1), ymask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (4 + y0 + 16 * y1), ymask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (8 + y0 + 16 * y1), ymask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (12 + y0 + 16 * y1), ymask, eviction_policy=
'evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2 + 4 * y3), tmp8, xmask & ymask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_tanh_0[grid(64)](buf1, primals_2, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4, 4), (16, 1, 4), torch.float32)
triton_poi_fused__softmax_1[grid(64)](buf2, buf3, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf4 = reinterpret_tensor(buf2, (4, 4, 4), (16, 4, 1), 0)
del buf2
triton_poi_fused__softmax_2[grid(16, 4)](buf3, buf4, 16, 4, XBLOCK=
4, YBLOCK=16, num_warps=1, num_stages=1)
buf5 = reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0)
del buf3
extern_kernels.bmm(buf4, primals_3, out=buf5)
return buf5, primals_3, buf1, buf4, primals_4
class AdditiveAttentionNew(nn.Module):
def __init__(self, in_features, att_hidden, out_features, bias=True):
super(AdditiveAttentionNew, self).__init__()
self.out_size = out_features
self.linear1 = nn.Linear(in_features=in_features, out_features=
att_hidden, bias=bias)
self.tanh = nn.Tanh()
self.linear2 = nn.Linear(in_features=att_hidden, out_features=
out_features, bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, input_0):
primals_1 = self.linear1.weight
primals_2 = self.linear1.bias
primals_4 = self.linear2.weight
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| LindgeW/DomainAdaption4DependencyParsing | AdditiveAttention | false | 5,546 | [
"Apache-2.0"
] | 1 | 5de136a37d8fe730e4235ed95bf923763fe21ea6 | https://github.com/LindgeW/DomainAdaption4DependencyParsing/tree/5de136a37d8fe730e4235ed95bf923763fe21ea6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_features, att_hidden, out_features, bias=True):
super().__init__()
self.out_size = out_features
self.linear1 = nn.Linear(in_features=in_features, out_features=
att_hidden, bias=bias)
self.tanh = nn.Tanh()
self.linear2 = nn.Linear(in_features=att_hidden, out_features=
out_features, bias=False)
self.softmax = nn.Softmax(dim=-1)
def forward(self, inputs, mask=None):
"""
:param inputs: (bz, seq_len, in_features)
:param mask: (bz, seq_len) 填充为0
:return:
"""
add_score = self.linear2(self.tanh(self.linear1(inputs)))
add_score = add_score.transpose(1, 2)
if mask is not None:
pad_mask = mask == 0
add_score = add_score.masked_fill(pad_mask[:, None, :], -
1000000000.0)
att_weights = self.softmax(add_score)
att_out = torch.bmm(att_weights, inputs)
return att_out
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
myConv2d | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/sr/csrhhqsexdcor6gq6tz4dawxblhadgekinzxxkt33uwojltligp6.py
# Topologically Sorted Source Nodes: [autograd_function_apply], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# autograd_function_apply => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_0 = async_compile.triton('triton_poi_fused_convolution_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [autograd_function_apply], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
del primals_1
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [autograd_function_apply], Original ATen: [aten.convolution]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_0.run(buf1, primals_2, 16, grid=grid(16), stream=stream0)
del primals_2
return (buf1, reinterpret_tensor(primals_3, (4, 4, 4, 4), (16, 64, 4, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class myConv2dFunction(torch.autograd.Function):
@staticmethod
def forward(ctx, input, weight, bias):
ctx.save_for_backward(input, weight, bias)
return F.conv2d(input, weight, bias)
@staticmethod
def backward(ctx, grad_output):
input, weight, _bias = ctx.saved_tensors
_out_channels, _in_channels, kernel_height, kernel_width = list(weight
.size())
grad_input = F.conv2d(grad_output, torch.Tensor.rot90(weight, 2, [2,
3]).transpose(0, 1), padding=(kernel_width - 1, kernel_height - 1))
grad_weight = F.conv2d(input.transpose(0, 1), grad_output.transpose
(0, 1)).transpose(0, 1)
grad_bias = grad_output.sum([0, 2, 3])
return grad_input, grad_weight, grad_bias
class myConv2d(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size):
super(myConv2d, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.kenerl_size = kernel_size
sqrtk = math.sqrt(1 / (in_channels * kernel_size[0] * kernel_size[1]))
self.weight = nn.Parameter(torch.Tensor(out_channels, in_channels,
kernel_size[0], kernel_size[1]))
self.weight.data.uniform_(-sqrtk, sqrtk)
self.bias = nn.Parameter(torch.Tensor(out_channels))
self.bias.data.uniform_(-sqrtk, sqrtk)
def forward(self, input):
return myConv2dFunction.apply(input, self.weight, self.bias)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': [4, 4]}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import math
import torch.nn as nn
import torch.nn.functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
del primals_1
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(16)](buf1, primals_2, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_2
return buf1, reinterpret_tensor(primals_3, (4, 4, 4, 4), (16, 64, 4, 1), 0)
class myConv2dFunction(torch.autograd.Function):
@staticmethod
def forward(ctx, input, weight, bias):
ctx.save_for_backward(input, weight, bias)
return F.conv2d(input, weight, bias)
@staticmethod
def backward(ctx, grad_output):
input, weight, _bias = ctx.saved_tensors
_out_channels, _in_channels, kernel_height, kernel_width = list(weight
.size())
grad_input = F.conv2d(grad_output, torch.Tensor.rot90(weight, 2, [2,
3]).transpose(0, 1), padding=(kernel_width - 1, kernel_height - 1))
grad_weight = F.conv2d(input.transpose(0, 1), grad_output.transpose
(0, 1)).transpose(0, 1)
grad_bias = grad_output.sum([0, 2, 3])
return grad_input, grad_weight, grad_bias
class myConv2dNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size):
super(myConv2dNew, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.kenerl_size = kernel_size
sqrtk = math.sqrt(1 / (in_channels * kernel_size[0] * kernel_size[1]))
self.weight = nn.Parameter(torch.Tensor(out_channels, in_channels,
kernel_size[0], kernel_size[1]))
self.weight.data.uniform_(-sqrtk, sqrtk)
self.bias = nn.Parameter(torch.Tensor(out_channels))
self.bias.data.uniform_(-sqrtk, sqrtk)
def forward(self, input_0):
primals_1 = self.weight
primals_2 = self.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| LogCreative/custom-tensor | myConv2d | false | 5,547 | [
"MIT"
] | 1 | 63eccf82821b4d4076a4fdfc7380ee72333360f1 | https://github.com/LogCreative/custom-tensor/tree/63eccf82821b4d4076a4fdfc7380ee72333360f1 | import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class myConv2dFunction(torch.autograd.Function):
@staticmethod
def forward(ctx, input, weight, bias):
ctx.save_for_backward(input, weight, bias)
return F.conv2d(input, weight, bias)
@staticmethod
def backward(ctx, grad_output):
input, weight, _bias = ctx.saved_tensors
_out_channels, _in_channels, kernel_height, kernel_width = list(weight
.size())
grad_input = F.conv2d(grad_output, torch.Tensor.rot90(weight, 2, [2,
3]).transpose(0, 1), padding=(kernel_width - 1, kernel_height - 1))
grad_weight = F.conv2d(input.transpose(0, 1), grad_output.transpose
(0, 1)).transpose(0, 1)
grad_bias = grad_output.sum([0, 2, 3])
return grad_input, grad_weight, grad_bias
class Model(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size):
super().__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.kenerl_size = kernel_size
sqrtk = math.sqrt(1 / (in_channels * kernel_size[0] * kernel_size[1]))
self.weight = nn.Parameter(torch.Tensor(out_channels, in_channels,
kernel_size[0], kernel_size[1]))
self.weight.data.uniform_(-sqrtk, sqrtk)
self.bias = nn.Parameter(torch.Tensor(out_channels))
self.bias.data.uniform_(-sqrtk, sqrtk)
def forward(self, input):
return myConv2dFunction.apply(input, self.weight, self.bias)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
FCNet | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/tf/ctfobpckmiv3kkga3a6gzs6unuclcnxpb4xc2h5r3udgxgix4ip5.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# y => relu
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_3), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, primals_3, 16, grid=grid(16), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf1, reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [y_1], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf3, primals_5, 16, grid=grid(16), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [y_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, buf3, reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf4)
del primals_7
return (buf4, primals_1, buf1, buf3, primals_6, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import numpy as np
from torch.autograd import Variable
import torch.nn as nn
import torch.nn.functional as F
import torch.optim
class BasicNet:
def __init__(self, optimizer_fn, gpu, LSTM=False):
self.gpu = gpu and torch.cuda.is_available()
self.LSTM = LSTM
if self.gpu:
self
self.FloatTensor = torch.FloatTensor
else:
self.FloatTensor = torch.FloatTensor
def to_torch_variable(self, x, dtype='float32'):
if isinstance(x, Variable):
return x
if not isinstance(x, torch.FloatTensor):
x = torch.from_numpy(np.asarray(x, dtype=dtype))
if self.gpu:
x = x
return Variable(x)
def reset(self, terminal):
if not self.LSTM:
return
if terminal:
self.h.data.zero_()
self.c.data.zero_()
self.h = Variable(self.h.data)
self.c = Variable(self.c.data)
class VanillaNet(BasicNet):
def predict(self, x, to_numpy=False):
y = self.forward(x)
if to_numpy:
if type(y) is list:
y = [y_.cpu().data.numpy() for y_ in y]
else:
y = y.cpu().data.numpy()
return y
class FCNet(nn.Module, VanillaNet):
def __init__(self, dims, optimizer_fn=None, gpu=True):
super(FCNet, self).__init__()
self.fc1 = nn.Linear(dims[0], dims[1])
self.fc2 = nn.Linear(dims[1], dims[2])
self.fc3 = nn.Linear(dims[2], dims[3])
BasicNet.__init__(self, optimizer_fn, gpu)
def forward(self, x):
x = self.to_torch_variable(x)
x = x.view(x.size(0), -1)
y = F.relu(self.fc1(x))
y = F.relu(self.fc2(y))
y = self.fc3(y)
return y
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'dims': [4, 4, 4, 4]}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import numpy as np
from torch.autograd import Variable
import torch.nn as nn
import torch.optim
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4),
(1, 4), 0), out=buf0)
del primals_2
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(16)](buf1, primals_3, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf1, reinterpret_tensor(primals_4, (4, 4), (1, 4
), 0), out=buf2)
buf3 = buf2
del buf2
triton_poi_fused_relu_0[grid(16)](buf3, primals_5, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, buf3, reinterpret_tensor(primals_6,
(4, 4), (1, 4), 0), alpha=1, beta=1, out=buf4)
del primals_7
return buf4, primals_1, buf1, buf3, primals_6, primals_4
class BasicNet:
def __init__(self, optimizer_fn, gpu, LSTM=False):
self.gpu = gpu and torch.cuda.is_available()
self.LSTM = LSTM
if self.gpu:
self
self.FloatTensor = torch.FloatTensor
else:
self.FloatTensor = torch.FloatTensor
def to_torch_variable(self, x, dtype='float32'):
if isinstance(x, Variable):
return x
if not isinstance(x, torch.FloatTensor):
x = torch.from_numpy(np.asarray(x, dtype=dtype))
if self.gpu:
x = x
return Variable(x)
def reset(self, terminal):
if not self.LSTM:
return
if terminal:
self.h.data.zero_()
self.c.data.zero_()
self.h = Variable(self.h.data)
self.c = Variable(self.c.data)
class VanillaNet(BasicNet):
def predict(self, x, to_numpy=False):
y = self.forward(x)
if to_numpy:
if type(y) is list:
y = [y_.cpu().data.numpy() for y_ in y]
else:
y = y.cpu().data.numpy()
return y
class FCNetNew(nn.Module, VanillaNet):
def __init__(self, dims, optimizer_fn=None, gpu=True):
super(FCNetNew, self).__init__()
self.fc1 = nn.Linear(dims[0], dims[1])
self.fc2 = nn.Linear(dims[1], dims[2])
self.fc3 = nn.Linear(dims[2], dims[3])
BasicNet.__init__(self, optimizer_fn, gpu)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_3 = self.fc1.bias
primals_2 = self.fc2.weight
primals_5 = self.fc2.bias
primals_4 = self.fc3.weight
primals_7 = self.fc3.bias
primals_6 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| G-Flor/deeprl | FCNet | false | 5,548 | [
"Apache-2.0"
] | 1 | aeae2c5d585e5853dc638968b1f090eb60abd351 | https://github.com/G-Flor/deeprl/tree/aeae2c5d585e5853dc638968b1f090eb60abd351 | import torch
import numpy as np
from torch.autograd import Variable
import torch.nn as nn
import torch.nn.functional as F
import torch.optim
class BasicNet:
def __init__(self, optimizer_fn, gpu, LSTM=False):
self.gpu = gpu and torch.cuda.is_available()
self.LSTM = LSTM
if self.gpu:
self
self.FloatTensor = torch.FloatTensor
else:
self.FloatTensor = torch.FloatTensor
def to_torch_variable(self, x, dtype='float32'):
if isinstance(x, Variable):
return x
if not isinstance(x, torch.FloatTensor):
x = torch.from_numpy(np.asarray(x, dtype=dtype))
if self.gpu:
x = x
return Variable(x)
def reset(self, terminal):
if not self.LSTM:
return
if terminal:
self.h.data.zero_()
self.c.data.zero_()
self.h = Variable(self.h.data)
self.c = Variable(self.c.data)
class VanillaNet(BasicNet):
def predict(self, x, to_numpy=False):
y = self.forward(x)
if to_numpy:
if type(y) is list:
y = [y_.cpu().data.numpy() for y_ in y]
else:
y = y.cpu().data.numpy()
return y
class Model(nn.Module, VanillaNet):
def __init__(self, dims, optimizer_fn=None, gpu=True):
super().__init__()
self.fc1 = nn.Linear(dims[0], dims[1])
self.fc2 = nn.Linear(dims[1], dims[2])
self.fc3 = nn.Linear(dims[2], dims[3])
BasicNet.__init__(self, optimizer_fn, gpu)
def forward(self, x):
x = self.to_torch_variable(x)
x = x.view(x.size(0), -1)
y = F.relu(self.fc1(x))
y = F.relu(self.fc2(y))
y = self.fc3(y)
return y
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return []
|
Net | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/nc/cncwsucylpsg2zmlivjfxu6vbd64ztxjndlsix2ysjtby3xohgk4.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# x_1 => tanh
# Graph fragment:
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%view_1,), kwargs = {})
triton_poi_fused_tanh_0 = async_compile.triton('triton_poi_fused_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_tanh_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + (x2), tmp3, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_tanh_0.run(buf1, primals_2, 256, grid=grid(256), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del primals_5
return (reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from numpy import *
class Net(torch.nn.Module):
def __init__(self, n_feature, n_hidden, n_output):
super(Net, self).__init__()
self.hidden = torch.nn.Linear(n_feature, n_hidden)
self.predict = torch.nn.Linear(n_hidden, n_output)
def forward(self, x):
x = self.hidden(x)
x = torch.tanh(x)
x = self.predict(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'n_feature': 4, 'n_hidden': 4, 'n_output': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
from numpy import *
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_tanh_0[grid(256)](buf1, primals_2, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del primals_5
return reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, primals_4
class NetNew(torch.nn.Module):
def __init__(self, n_feature, n_hidden, n_output):
super(NetNew, self).__init__()
self.hidden = torch.nn.Linear(n_feature, n_hidden)
self.predict = torch.nn.Linear(n_hidden, n_output)
def forward(self, input_0):
primals_1 = self.hidden.weight
primals_2 = self.hidden.bias
primals_4 = self.predict.weight
primals_5 = self.predict.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| LishudaNoBug/learning_PyTorch | Net | false | 5,549 | [
"MIT"
] | 1 | 1026035a9cb3d70e2fe97363b532e63db3ca136d | https://github.com/LishudaNoBug/learning_PyTorch/tree/1026035a9cb3d70e2fe97363b532e63db3ca136d | import torch
from numpy import *
class Model(torch.nn.Module):
def __init__(self, n_feature, n_hidden, n_output):
super().__init__()
self.hidden = torch.nn.Linear(n_feature, n_hidden)
self.predict = torch.nn.Linear(n_hidden, n_output)
def forward(self, x):
x = self.hidden(x)
x = torch.tanh(x)
x = self.predict(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
DiceLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/gd/cgdtd7ki7lurypoeyfwjebdfquygdeupjef4ltfbbbdk5u7owcpl.py
# Topologically Sorted Source Nodes: [intersection, sum_1, sum_2, sum_3], Original ATen: [aten.mul, aten.sum]
# Source node to ATen node mapping:
# intersection => mul
# sum_1 => sum_1
# sum_2 => sum_2
# sum_3 => sum_3
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %view_1), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%view, [1]), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%view_1, [1]), kwargs = {})
triton_per_fused_mul_sum_0 = async_compile.triton('triton_per_fused_mul_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mul_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 3, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (64*x0)), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + (r1 + (64*x0)), xmask, other=0.0)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp10 = tl.where(xmask, tmp8, 0)
tmp11 = tl.sum(tmp10, 1)[:, None]
tmp12 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp14 = tl.where(xmask, tmp12, 0)
tmp15 = tl.sum(tmp14, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp7, xmask)
tl.store(out_ptr1 + (x0), tmp11, xmask)
tl.store(out_ptr2 + (x0), tmp15, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/vq/cvqiixp4wmb73ig2cla6idbqq7i6vd5n3qmdluadrv32f52pdgw3.py
# Topologically Sorted Source Nodes: [add, mul_1, add_1, add_2, score, sum_4, truediv_1, score_1], Original ATen: [aten.add, aten.mul, aten.div, aten.sum, aten.rsub]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# add_2 => add_2
# mul_1 => mul_1
# score => div
# score_1 => sub
# sum_4 => sum_4
# truediv_1 => div_1
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_1, 1.0), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_2, %sum_3), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, 1.0), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_1, %add_2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%div,), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_4, 4), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %div_1), kwargs = {})
triton_per_fused_add_div_mul_rsub_sum_1 = async_compile.triton('triton_per_fused_add_div_mul_rsub_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {4: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mul_rsub_sum_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_mul_rsub_sum_1(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp5 = tl.load(in_ptr1 + (r0), None)
tmp6 = tl.load(in_ptr2 + (r0), None)
tmp1 = 1.0
tmp2 = tmp0 + tmp1
tmp3 = 2.0
tmp4 = tmp2 * tmp3
tmp7 = tmp5 + tmp6
tmp8 = tmp7 + tmp1
tmp9 = tmp4 / tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.sum(tmp10, 1)[:, None]
tmp13 = 0.25
tmp14 = tmp12 * tmp13
tmp15 = tmp1 - tmp14
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp15, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, ), (1, ), torch.float32)
buf1 = empty_strided_cuda((4, ), (1, ), torch.float32)
buf2 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [intersection, sum_1, sum_2, sum_3], Original ATen: [aten.mul, aten.sum]
stream0 = get_raw_stream(0)
triton_per_fused_mul_sum_0.run(arg1_1, arg0_1, buf0, buf1, buf2, 4, 64, grid=grid(4), stream=stream0)
del arg0_1
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [add, mul_1, add_1, add_2, score, sum_4, truediv_1, score_1], Original ATen: [aten.add, aten.mul, aten.div, aten.sum, aten.rsub]
triton_per_fused_add_div_mul_rsub_sum_1.run(buf4, buf0, buf1, buf2, 1, 4, grid=grid(1), stream=stream0)
del buf0
del buf1
del buf2
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from typing import *
class DiceLoss(nn.Module):
def __init__(self, smooth: 'float'=1.0):
super().__init__()
self.smooth = smooth
def forward(self, logits, targets):
num = targets.size(0)
probs = torch.sigmoid(logits)
m1, m2 = probs.view(num, -1), targets.view(num, -1)
intersection = m1 * m2
score = 2.0 * (intersection.sum(1) + self.smooth) / (m1.sum(1) + m2
.sum(1) + self.smooth)
score = 1 - score.sum() / num
return score
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from typing import *
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1,
out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 4
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 64 * x0), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + (r1 + 64 * x0), xmask, other=0.0)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp10 = tl.where(xmask, tmp8, 0)
tmp11 = tl.sum(tmp10, 1)[:, None]
tmp12 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp14 = tl.where(xmask, tmp12, 0)
tmp15 = tl.sum(tmp14, 1)[:, None]
tl.store(out_ptr0 + x0, tmp7, xmask)
tl.store(out_ptr1 + x0, tmp11, xmask)
tl.store(out_ptr2 + x0, tmp15, xmask)
@triton.jit
def triton_per_fused_add_div_mul_rsub_sum_1(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp5 = tl.load(in_ptr1 + r0, None)
tmp6 = tl.load(in_ptr2 + r0, None)
tmp1 = 1.0
tmp2 = tmp0 + tmp1
tmp3 = 2.0
tmp4 = tmp2 * tmp3
tmp7 = tmp5 + tmp6
tmp8 = tmp7 + tmp1
tmp9 = tmp4 / tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.sum(tmp10, 1)[:, None]
tmp13 = 0.25
tmp14 = tmp12 * tmp13
tmp15 = tmp1 - tmp14
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp15, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4,), (1,), torch.float32)
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
buf2 = empty_strided_cuda((4,), (1,), torch.float32)
get_raw_stream(0)
triton_per_fused_mul_sum_0[grid(4)](arg1_1, arg0_1, buf0, buf1,
buf2, 4, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
triton_per_fused_add_div_mul_rsub_sum_1[grid(1)](buf4, buf0, buf1,
buf2, 1, 4, XBLOCK=1, num_warps=2, num_stages=1)
del buf0
del buf1
del buf2
return buf4,
class DiceLossNew(nn.Module):
def __init__(self, smooth: 'float'=1.0):
super().__init__()
self.smooth = smooth
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| Lleyton-Ariton/landwatch | DiceLoss | false | 5,550 | [
"MIT"
] | 1 | 21e86e899d33d0ee349cf9bf87c6c13ebdab82fa | https://github.com/Lleyton-Ariton/landwatch/tree/21e86e899d33d0ee349cf9bf87c6c13ebdab82fa | import torch
import torch.nn as nn
from typing import *
class Model(nn.Module):
def __init__(self, smooth: 'float'=1.0):
super().__init__()
self.smooth = smooth
def forward(self, logits, targets):
num = targets.size(0)
probs = torch.sigmoid(logits)
m1, m2 = probs.view(num, -1), targets.view(num, -1)
intersection = m1 * m2
score = 2.0 * (intersection.sum(1) + self.smooth) / (m1.sum(1) + m2
.sum(1) + self.smooth)
score = 1 - score.sum() / num
return score
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
KL_Loss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/l3/cl3mqwaki56dc4zcxfjjgkbopnejxzhksqm6egdinynmjrsrw2qw.py
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# softmax => exp_1
# Graph fragment:
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, 1), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor, [1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor, %amax_default), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor, 1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp3 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tmp16 = tl_math.exp(tmp15)
tl.store(out_ptr0 + (x3), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/pf/cpfkvifhrhobwuxls65xhwdpkryeblqmmtghouii4lp3rhe3crx4.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %mul_tensor_1 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 1), kwargs = {})
# %amax_default_1 : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor_1, [1], True), kwargs = {})
# %sub_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor_1, %amax_default_1), kwargs = {})
# %div_tensor_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor_1, 1), kwargs = {})
triton_poi_fused_1 = async_compile.triton('triton_poi_fused_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp3 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tl.store(out_ptr0 + (x3), tmp15, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/d5/cd5jvj2xtf5vbrydrjr7gjura3e2lbc3fjgbn4anltegd6a6wh3n.py
# Topologically Sorted Source Nodes: [softmax, teacher_outputs, kl_div, output_batch, loss], Original ATen: [aten._softmax, aten.add, aten.xlogy, aten._log_softmax, aten.mul, aten.sub, aten.sum, aten.div]
# Source node to ATen node mapping:
# kl_div => div_3, eq, full_default, full_default_1, isnan, log_1, mul, mul_1, sub_3, sum_3, where, where_1
# loss => mul_2
# output_batch => exp, log, sub_1, sum_1
# softmax => div_2, sum_2
# teacher_outputs => add
# Graph fragment:
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_1, [1], True), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_1, %sum_2), kwargs = {})
# %add : [num_users=5] = call_function[target=torch.ops.aten.add.Tensor](args = (%div_2, 1e-07), kwargs = {})
# %isnan : [num_users=1] = call_function[target=torch.ops.aten.isnan.default](args = (%add,), kwargs = {})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], nan), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %eq : [num_users=1] = call_function[target=torch.ops.aten.eq.Scalar](args = (%add, 0), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %log_1 : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%add,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, %log_1), kwargs = {})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq, %full_default, %mul_1), kwargs = {})
# %where_1 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%isnan, %full_default_1, %where), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor_1,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1], True), kwargs = {})
# %log : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%sum_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%div_tensor_1, %log), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, %sub_1), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%where_1, %mul), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%sub_3,), kwargs = {})
# %div_3 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_3, 4), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_3, 1), kwargs = {})
triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2 = async_compile.triton('triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 10, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r3 = rindex
r0 = rindex % 16
r2 = (rindex // 64)
tmp0 = tl.load(in_ptr0 + (r3), None)
tmp1 = tl.load(in_ptr0 + (r0 + (64*r2)), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (16 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (32 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (48 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (r3), None)
tmp20 = tl.load(in_ptr1 + (r0 + (64*r2)), None, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (16 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr1 + (32 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp28 = tl.load(in_ptr1 + (48 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp9 = 1e-07
tmp10 = tmp8 + tmp9
tmp11 = libdevice.isnan(tmp10).to(tl.int1)
tmp12 = 0.0
tmp13 = tmp10 == tmp12
tmp14 = tl_math.log(tmp10)
tmp15 = tmp10 * tmp14
tmp16 = tl.where(tmp13, tmp12, tmp15)
tmp17 = float("nan")
tmp18 = tl.where(tmp11, tmp17, tmp16)
tmp21 = tl_math.exp(tmp20)
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp27 + tmp29
tmp31 = tl_math.log(tmp30)
tmp32 = tmp19 - tmp31
tmp33 = tmp10 * tmp32
tmp34 = tmp18 - tmp33
tmp35 = tl.broadcast_to(tmp34, [RBLOCK])
tmp37 = triton_helpers.promote_to_tensor(tl.sum(tmp35, 0))
tmp38 = 0.25
tmp39 = tmp37 * tmp38
tmp40 = 1.0
tmp41 = tmp39 * tmp40
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp41, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(arg1_1, buf0, 256, grid=grid(256), stream=stream0)
del arg1_1
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(arg0_1, buf2, 256, grid=grid(256), stream=stream0)
del arg0_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [softmax, teacher_outputs, kl_div, output_batch, loss], Original ATen: [aten._softmax, aten.add, aten.xlogy, aten._log_softmax, aten.mul, aten.sub, aten.sum, aten.div]
triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2.run(buf4, buf0, buf2, 1, 256, grid=grid(1), stream=stream0)
del buf0
del buf2
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
import torch.nn.functional as F
import torch.utils
class KL_Loss(nn.Module):
def __init__(self, temperature=1):
super(KL_Loss, self).__init__()
self.T = temperature
def forward(self, output_batch, teacher_outputs):
output_batch = F.log_softmax(output_batch / self.T, dim=1)
teacher_outputs = F.softmax(teacher_outputs / self.T, dim=1) + 10 ** -7
loss = self.T * self.T * nn.KLDivLoss(reduction='batchmean')(
output_batch, teacher_outputs)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch import nn
import torch.utils
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp3 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tmp16 = tl_math.exp(tmp15)
tl.store(out_ptr0 + x3, tmp16, xmask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp3 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tl.store(out_ptr0 + x3, tmp15, xmask)
@triton.jit
def triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2(
in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r3 = rindex
r0 = rindex % 16
r2 = rindex // 64
tmp0 = tl.load(in_ptr0 + r3, None)
tmp1 = tl.load(in_ptr0 + (r0 + 64 * r2), None, eviction_policy='evict_last'
)
tmp2 = tl.load(in_ptr0 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp19 = tl.load(in_ptr1 + r3, None)
tmp20 = tl.load(in_ptr1 + (r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr1 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp28 = tl.load(in_ptr1 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp9 = 1e-07
tmp10 = tmp8 + tmp9
tmp11 = libdevice.isnan(tmp10).to(tl.int1)
tmp12 = 0.0
tmp13 = tmp10 == tmp12
tmp14 = tl_math.log(tmp10)
tmp15 = tmp10 * tmp14
tmp16 = tl.where(tmp13, tmp12, tmp15)
tmp17 = float('nan')
tmp18 = tl.where(tmp11, tmp17, tmp16)
tmp21 = tl_math.exp(tmp20)
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp27 + tmp29
tmp31 = tl_math.log(tmp30)
tmp32 = tmp19 - tmp31
tmp33 = tmp10 * tmp32
tmp34 = tmp18 - tmp33
tmp35 = tl.broadcast_to(tmp34, [RBLOCK])
tmp37 = triton_helpers.promote_to_tensor(tl.sum(tmp35, 0))
tmp38 = 0.25
tmp39 = tmp37 * tmp38
tmp40 = 1.0
tmp41 = tmp39 * tmp40
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp41, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(256)](arg1_1, buf0, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del arg1_1
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_1[grid(256)](arg0_1, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
triton_per_fused__log_softmax__softmax_add_div_mul_sub_sum_xlogy_2[grid
(1)](buf4, buf0, buf2, 1, 256, num_warps=2, num_stages=1)
del buf0
del buf2
return buf4,
class KL_LossNew(nn.Module):
def __init__(self, temperature=1):
super(KL_LossNew, self).__init__()
self.T = temperature
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| Little0o0/FedML | KL_Loss | false | 5,551 | [
"Apache-2.0"
] | 1 | 720015c90fcfec88d465a81b1e8fb45676dce9fb | https://github.com/Little0o0/FedML/tree/720015c90fcfec88d465a81b1e8fb45676dce9fb | import torch
from torch import nn
import torch.nn.functional as F
import torch.utils
class Model(nn.Module):
def __init__(self, temperature=1):
super().__init__()
self.T = temperature
def forward(self, output_batch, teacher_outputs):
output_batch = F.log_softmax(output_batch / self.T, dim=1)
teacher_outputs = F.softmax(teacher_outputs / self.T, dim=1) + 10 ** -7
loss = self.T * self.T * nn.KLDivLoss(reduction='batchmean')(
output_batch, teacher_outputs)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
CE_Loss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/jh/cjhafiazvhnqzahvgpeyzzxgeb5atp7ebiv4plitnblial63qxb6.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %mul_tensor_1 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 1), kwargs = {})
# %amax_default_1 : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor_1, [1], True), kwargs = {})
# %sub_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor_1, %amax_default_1), kwargs = {})
# %div_tensor_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor_1, 1), kwargs = {})
triton_poi_fused_0 = async_compile.triton('triton_poi_fused_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp3 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tl.store(out_ptr0 + (x3), tmp15, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/k4/ck4jcdo7pholu6j7hyu5a5wtavlli6c52bnxxxqwlta7ah2dps4s.py
# Topologically Sorted Source Nodes: [teacher_outputs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# teacher_outputs => exp_1
# Graph fragment:
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, 1), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor, [1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor, %amax_default), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor, 1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor,), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp3 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tmp16 = tl_math.exp(tmp15)
tl.store(out_ptr0 + (x3), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/mn/cmnlidmjpz2yihyn6sanrop2yyzhcy2ynecgwnndb5sydmmmfwas.py
# Topologically Sorted Source Nodes: [output_batch, teacher_outputs, mul, sum_1, mul_1, loss], Original ATen: [aten._log_softmax, aten._softmax, aten.mul, aten.sum, aten.div]
# Source node to ATen node mapping:
# loss => div_3
# mul => mul
# mul_1 => mul_1
# output_batch => exp, log, sub_1, sum_1
# sum_1 => sum_3
# teacher_outputs => div_2, sum_2
# Graph fragment:
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor_1,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1], True), kwargs = {})
# %log : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%sum_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%div_tensor_1, %log), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_1, [1], True), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_1, %sum_2), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %div_2), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%mul,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_3, -1), kwargs = {})
# %div_3 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_1, 4), kwargs = {})
triton_per_fused__log_softmax__softmax_div_mul_sum_2 = async_compile.triton('triton_per_fused__log_softmax__softmax_div_mul_sum_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__log_softmax__softmax_div_mul_sum_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 10, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__log_softmax__softmax_div_mul_sum_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r3 = rindex
r0 = rindex % 16
r2 = (rindex // 64)
tmp0 = tl.load(in_ptr0 + (r3), None)
tmp1 = tl.load(in_ptr0 + (r0 + (64*r2)), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (16 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (32 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (48 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (r3), None)
tmp15 = tl.load(in_ptr1 + (r0 + (64*r2)), None, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (16 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr1 + (32 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (48 + r0 + (64*r2)), None, eviction_policy='evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp5 + tmp7
tmp10 = tl_math.exp(tmp9)
tmp11 = tmp8 + tmp10
tmp12 = tl_math.log(tmp11)
tmp13 = tmp0 - tmp12
tmp17 = tmp15 + tmp16
tmp19 = tmp17 + tmp18
tmp21 = tmp19 + tmp20
tmp22 = tmp14 / tmp21
tmp23 = tmp13 * tmp22
tmp24 = tl.broadcast_to(tmp23, [RBLOCK])
tmp26 = triton_helpers.promote_to_tensor(tl.sum(tmp24, 0))
tmp27 = -1.0
tmp28 = tmp26 * tmp27
tmp29 = 0.25
tmp30 = tmp28 * tmp29
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp30, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [teacher_outputs], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(arg1_1, buf1, 256, grid=grid(256), stream=stream0)
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [output_batch, teacher_outputs, mul, sum_1, mul_1, loss], Original ATen: [aten._log_softmax, aten._softmax, aten.mul, aten.sum, aten.div]
triton_per_fused__log_softmax__softmax_div_mul_sum_2.run(buf4, buf0, buf1, 1, 256, grid=grid(1), stream=stream0)
del buf0
del buf1
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
import torch.nn.functional as F
import torch.utils
class CE_Loss(nn.Module):
def __init__(self, temperature=1):
super(CE_Loss, self).__init__()
self.T = temperature
def forward(self, output_batch, teacher_outputs):
output_batch = F.log_softmax(output_batch / self.T, dim=1)
teacher_outputs = F.softmax(teacher_outputs / self.T, dim=1)
loss = -self.T * self.T * torch.sum(torch.mul(output_batch,
teacher_outputs)) / teacher_outputs.size(0)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch import nn
import torch.utils
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp3 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tl.store(out_ptr0 + x3, tmp15, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp3 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = tmp14 * tmp1
tmp16 = tl_math.exp(tmp15)
tl.store(out_ptr0 + x3, tmp16, xmask)
@triton.jit
def triton_per_fused__log_softmax__softmax_div_mul_sum_2(in_out_ptr0,
in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r3 = rindex
r0 = rindex % 16
r2 = rindex // 64
tmp0 = tl.load(in_ptr0 + r3, None)
tmp1 = tl.load(in_ptr0 + (r0 + 64 * r2), None, eviction_policy='evict_last'
)
tmp3 = tl.load(in_ptr0 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp14 = tl.load(in_ptr1 + r3, None)
tmp15 = tl.load(in_ptr1 + (r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp16 = tl.load(in_ptr1 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp18 = tl.load(in_ptr1 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp20 = tl.load(in_ptr1 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp5 + tmp7
tmp10 = tl_math.exp(tmp9)
tmp11 = tmp8 + tmp10
tmp12 = tl_math.log(tmp11)
tmp13 = tmp0 - tmp12
tmp17 = tmp15 + tmp16
tmp19 = tmp17 + tmp18
tmp21 = tmp19 + tmp20
tmp22 = tmp14 / tmp21
tmp23 = tmp13 * tmp22
tmp24 = tl.broadcast_to(tmp23, [RBLOCK])
tmp26 = triton_helpers.promote_to_tensor(tl.sum(tmp24, 0))
tmp27 = -1.0
tmp28 = tmp26 * tmp27
tmp29 = 0.25
tmp30 = tmp28 * tmp29
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp30, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(256)](arg0_1, buf0, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_1[grid(256)](arg1_1, buf1, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
triton_per_fused__log_softmax__softmax_div_mul_sum_2[grid(1)](buf4,
buf0, buf1, 1, 256, num_warps=2, num_stages=1)
del buf0
del buf1
return buf4,
class CE_LossNew(nn.Module):
def __init__(self, temperature=1):
super(CE_LossNew, self).__init__()
self.T = temperature
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| Little0o0/FedML | CE_Loss | false | 5,552 | [
"Apache-2.0"
] | 1 | 720015c90fcfec88d465a81b1e8fb45676dce9fb | https://github.com/Little0o0/FedML/tree/720015c90fcfec88d465a81b1e8fb45676dce9fb | import torch
from torch import nn
import torch.nn.functional as F
import torch.utils
class Model(nn.Module):
def __init__(self, temperature=1):
super().__init__()
self.T = temperature
def forward(self, output_batch, teacher_outputs):
output_batch = F.log_softmax(output_batch / self.T, dim=1)
teacher_outputs = F.softmax(teacher_outputs / self.T, dim=1)
loss = -self.T * self.T * torch.sum(torch.mul(output_batch,
teacher_outputs)) / teacher_outputs.size(0)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
SENet | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/nc/cncwsucylpsg2zmlivjfxu6vbd64ztxjndlsix2ysjtby3xohgk4.py
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# out_1 => tanh
# Graph fragment:
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%view_1,), kwargs = {})
triton_poi_fused_tanh_0 = async_compile.triton('triton_poi_fused_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_tanh_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + (x2), tmp3, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_tanh_0.run(buf1, primals_2, 256, grid=grid(256), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [out_3], Original ATen: [aten.tanh]
triton_poi_fused_tanh_0.run(buf3, primals_5, 256, grid=grid(256), stream=stream0)
del primals_5
return (buf3, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, buf3, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.utils.data
class SENet(nn.Module):
"""support estimation network"""
def __init__(self, input_size: 'int', hidden_size: 'int', output_dims:
'int') ->None:
super(SENet, self).__init__()
self.l_1 = nn.Linear(input_size, hidden_size)
self.l_2 = nn.Linear(hidden_size, output_dims)
self.act = nn.Tanh()
def forward(self, x: 'torch.Tensor') ->torch.Tensor:
out = self.l_1(x)
out = self.act(out)
out = self.l_2(out)
out = self.act(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_size': 4, 'hidden_size': 4, 'output_dims': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_tanh_0[grid(256)](buf1, primals_2, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf2
triton_poi_fused_tanh_0[grid(256)](buf3, primals_5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_5
return buf3, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf1, buf3, primals_4
class SENetNew(nn.Module):
"""support estimation network"""
def __init__(self, input_size: 'int', hidden_size: 'int', output_dims:
'int') ->None:
super(SENetNew, self).__init__()
self.l_1 = nn.Linear(input_size, hidden_size)
self.l_2 = nn.Linear(hidden_size, output_dims)
self.act = nn.Tanh()
def forward(self, input_0):
primals_1 = self.l_1.weight
primals_2 = self.l_1.bias
primals_4 = self.l_2.weight
primals_5 = self.l_2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| L-Net-1992/DI-engine | SENet | false | 5,553 | [
"Apache-2.0"
] | 1 | 06803b4e18fa64bbed0fd1d44952242c0c063b0f | https://github.com/L-Net-1992/DI-engine/tree/06803b4e18fa64bbed0fd1d44952242c0c063b0f | import torch
import torch.nn as nn
import torch.utils.data
class Model(nn.Module):
"""support estimation network"""
def __init__(self, input_size: 'int', hidden_size: 'int', output_dims:
'int') ->None:
super().__init__()
self.l_1 = nn.Linear(input_size, hidden_size)
self.l_2 = nn.Linear(hidden_size, output_dims)
self.act = nn.Tanh()
def forward(self, x: 'torch.Tensor') ->torch.Tensor:
out = self.l_1(x)
out = self.act(out)
out = self.l_2(out)
out = self.act(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
MAPELoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/57/c577ica3v5sskq3ryzudbubbezlo2d426ruphdt5ra7ufqarb4m5.py
# Topologically Sorted Source Nodes: [sub, abs_1, abs_2, add, truediv, mean], Original ATen: [aten.sub, aten.abs, aten.add, aten.div, aten.mean]
# Source node to ATen node mapping:
# abs_1 => abs_1
# abs_2 => abs_2
# add => add
# mean => mean
# sub => sub
# truediv => div
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%sub,), kwargs = {})
# %abs_2 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%arg1_1,), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%abs_2, 0.01), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%abs_1, %add), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%div,), kwargs = {})
triton_per_fused_abs_add_div_mean_sub_0 = async_compile.triton('triton_per_fused_abs_add_div_mean_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_abs_add_div_mean_sub_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_abs_add_div_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = tl_math.abs(tmp1)
tmp5 = 0.01
tmp6 = tmp4 + tmp5
tmp7 = tmp3 / tmp6
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = 256.0
tmp12 = tmp10 / tmp11
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp12, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [sub, abs_1, abs_2, add, truediv, mean], Original ATen: [aten.sub, aten.abs, aten.add, aten.div, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_abs_add_div_mean_sub_0.run(buf1, arg0_1, arg1_1, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class MAPELoss(nn.Module):
def forward(self, input, target):
return (torch.abs(input - target) / (torch.abs(target) + 0.01)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_add_div_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = tl_math.abs(tmp1)
tmp5 = 0.01
tmp6 = tmp4 + tmp5
tmp7 = tmp3 / tmp6
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = 256.0
tmp12 = tmp10 / tmp11
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp12, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_add_div_mean_sub_0[grid(1)](buf1, arg0_1,
arg1_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
class MAPELossNew(nn.Module):
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| LongerVision/oidn | MAPELoss | false | 5,554 | [
"Apache-2.0"
] | 1 | 2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | https://github.com/LongerVision/oidn/tree/2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | import torch
import torch.nn as nn
class Model(nn.Module):
def forward(self, input, target):
return (torch.abs(input - target) / (torch.abs(target) + 0.01)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
MultiheadAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/qw/cqw7yoyglmtjad3kirznl5odetqfs3k6pjtnfdbzklyhsdvuvgft.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# multi_head_attention_forward => mul
# Graph fragment:
# %mul : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_3, 1.0), kwargs = {})
triton_poi_fused_mul_0 = async_compile.triton('triton_poi_fused_mul_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/hz/chzi3aam26mikdhljz5x7jlqazm7kpktzeptsf36thgfhsg7ub6a.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%bmm, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%bmm, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/em/cem6qbxwbiqnjqybzk5arf2obt5uggy4qs7otwwpovvnrhvdc6h4.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_2 = async_compile.triton('triton_poi_fused__softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rh/crhjfwyl6xoj5ylcsbbh6lp2vlegits2zkdej3b3wb2q4ddfnejv.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# multi_head_attention_forward => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_7,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_3 = async_compile.triton('triton_poi_fused_clone_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 4
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x1)), xmask & ymask)
tl.store(out_ptr0 + (x1 + (4*y0)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5z/c5zy7julai2lhuinuwjgyl62nx7cyws6ni5poe5jzp7qn532rcgh.py
# Topologically Sorted Source Nodes: [add], Original ATen: [aten.add]
# Source node to ATen node mapping:
# add => add
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_1, %squeeze), kwargs = {})
triton_poi_fused_add_4 = async_compile.triton('triton_poi_fused_add_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_4', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_4(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_out_ptr0 + (x2), xmask)
tmp2 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (12, 4), (4, 1))
assert_size_stride(primals_3, (12, ), (1, ))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.addmm]
extern_kernels.addmm(reinterpret_tensor(primals_3, (4, ), (1, ), 4), primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 16), alpha=1, beta=1, out=buf1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.addmm]
extern_kernels.addmm(reinterpret_tensor(primals_3, (4, ), (1, ), 8), primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 32), alpha=1, beta=1, out=buf2)
del primals_2
buf3 = reinterpret_tensor(buf0, (4, 4, 1), (1, 4, 16), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_0.run(buf3, primals_3, 16, grid=grid(16), stream=stream0)
del primals_3
buf4 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.bmm]
extern_kernels.bmm(buf3, reinterpret_tensor(buf1, (4, 1, 4), (1, 1, 4), 0), out=buf4)
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf4, buf5, 64, grid=grid(64), stream=stream0)
buf6 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf5, buf6, 64, grid=grid(64), stream=stream0)
del buf5
buf7 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.bmm]
extern_kernels.bmm(buf6, reinterpret_tensor(buf2, (4, 4, 1), (1, 4, 1), 0), out=buf7)
buf8 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.clone]
triton_poi_fused_clone_3.run(buf7, buf8, 4, 4, grid=grid(4, 4), stream=stream0)
buf9 = reinterpret_tensor(buf7, (4, 4), (4, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf8, (4, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf9)
buf10 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [add], Original ATen: [aten.add]
triton_poi_fused_add_4.run(buf10, primals_1, primals_5, 16, grid=grid(16), stream=stream0)
del primals_5
return (buf10, primals_1, buf6, reinterpret_tensor(buf8, (4, 4), (4, 1), 0), primals_4, reinterpret_tensor(buf2, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf3, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf1, (4, 4, 1), (1, 4, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((12, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((12, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class MultiheadAttention(nn.Module):
"""A warpper for torch.nn.MultiheadAttention.
This module implements MultiheadAttention with residual connection.
Args:
embed_dims (int): The embedding dimension.
num_heads (int): Parallel attention heads. Same as
`nn.MultiheadAttention`.
attn_drop (float): A Dropout layer on attn_output_weights. Default 0.0.
proj_drop (float): The drop out rate after attention. Default 0.0.
"""
def __init__(self, embed_dims, num_heads, attn_drop=0.0, proj_drop=0.0):
super(MultiheadAttention, self).__init__()
assert embed_dims % num_heads == 0, f'embed_dims must be divisible by num_heads. got {embed_dims} and {num_heads}.'
self.embed_dims = embed_dims
self.num_heads = num_heads
self.attn = nn.MultiheadAttention(embed_dims, num_heads, attn_drop)
self.dropout = nn.Dropout(proj_drop)
def forward(self, x, key=None, value=None, residual=None, query_pos=
None, key_pos=None, attn_mask=None, key_padding_mask=None):
"""Forward function for `MultiheadAttention`.
Args:
x (Tensor): The input query with shape [num_query, bs,
embed_dims]. Same in `nn.MultiheadAttention.forward`.
key (Tensor): The key tensor with shape [num_key, bs,
embed_dims]. Same in `nn.MultiheadAttention.forward`.
Default None. If None, the `query` will be used.
value (Tensor): The value tensor with same shape as `key`.
Same in `nn.MultiheadAttention.forward`. Default None.
If None, the `key` will be used.
residual (Tensor): The tensor used for addition, with the
same shape as `x`. Default None. If None, `x` will be used.
query_pos (Tensor): The positional encoding for query, with
the same shape as `x`. Default None. If not None, it will
be added to `x` before forward function.
key_pos (Tensor): The positional encoding for `key`, with the
same shape as `key`. Default None. If not None, it will
be added to `key` before forward function. If None, and
`query_pos` has the same shape as `key`, then `query_pos`
will be used for `key_pos`.
attn_mask (Tensor): ByteTensor mask with shape [num_query,
num_key]. Same in `nn.MultiheadAttention.forward`.
Default None.
key_padding_mask (Tensor): ByteTensor with shape [bs, num_key].
Same in `nn.MultiheadAttention.forward`. Default None.
Returns:
Tensor: forwarded results with shape [num_query, bs, embed_dims].
"""
query = x
if key is None:
key = query
if value is None:
value = key
if residual is None:
residual = x
if key_pos is None:
if query_pos is not None and key is not None:
if query_pos.shape == key.shape:
key_pos = query_pos
if query_pos is not None:
query = query + query_pos
if key_pos is not None:
key = key + key_pos
out = self.attn(query, key, value=value, attn_mask=attn_mask,
key_padding_mask=key_padding_mask)[0]
return residual + self.dropout(out)
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'embed_dims': 4, 'num_heads': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(in_out_ptr0 + x2, tmp4, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 4
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x1), xmask & ymask)
tl.store(out_ptr0 + (x1 + 4 * y0), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_4(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK:
tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_out_ptr0 + x2, xmask)
tmp2 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (12, 4), (4, 1))
assert_size_stride(primals_3, (12,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4),
(1, 4), 0), out=buf0)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(reinterpret_tensor(primals_3, (4,), (1,), 4),
primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 16),
alpha=1, beta=1, out=buf1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(reinterpret_tensor(primals_3, (4,), (1,), 8),
primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 32),
alpha=1, beta=1, out=buf2)
del primals_2
buf3 = reinterpret_tensor(buf0, (4, 4, 1), (1, 4, 16), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_mul_0[grid(16)](buf3, primals_3, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_3
buf4 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf3, reinterpret_tensor(buf1, (4, 1, 4), (1, 1,
4), 0), out=buf4)
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_1[grid(64)](buf4, buf5, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf6 = buf4
del buf4
triton_poi_fused__softmax_2[grid(64)](buf5, buf6, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf5
buf7 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
extern_kernels.bmm(buf6, reinterpret_tensor(buf2, (4, 4, 1), (1, 4,
1), 0), out=buf7)
buf8 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
triton_poi_fused_clone_3[grid(4, 4)](buf7, buf8, 4, 4, XBLOCK=4,
YBLOCK=4, num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf7, (4, 4), (4, 1), 0)
del buf7
extern_kernels.mm(reinterpret_tensor(buf8, (4, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf9)
buf10 = buf9
del buf9
triton_poi_fused_add_4[grid(16)](buf10, primals_1, primals_5, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_5
return buf10, primals_1, buf6, reinterpret_tensor(buf8, (4, 4), (4, 1), 0
), primals_4, reinterpret_tensor(buf2, (4, 1, 4), (1, 1, 4), 0
), reinterpret_tensor(buf3, (4, 1, 4), (1, 1, 4), 0
), reinterpret_tensor(buf1, (4, 4, 1), (1, 4, 1), 0)
class MultiheadAttentionNew(nn.Module):
"""A warpper for torch.nn.MultiheadAttention.
This module implements MultiheadAttention with residual connection.
Args:
embed_dims (int): The embedding dimension.
num_heads (int): Parallel attention heads. Same as
`nn.MultiheadAttention`.
attn_drop (float): A Dropout layer on attn_output_weights. Default 0.0.
proj_drop (float): The drop out rate after attention. Default 0.0.
"""
def __init__(self, embed_dims, num_heads, attn_drop=0.0, proj_drop=0.0):
super(MultiheadAttentionNew, self).__init__()
assert embed_dims % num_heads == 0, f'embed_dims must be divisible by num_heads. got {embed_dims} and {num_heads}.'
self.embed_dims = embed_dims
self.num_heads = num_heads
self.attn = nn.MultiheadAttention(embed_dims, num_heads, attn_drop)
self.dropout = nn.Dropout(proj_drop)
def forward(self, input_0):
primals_2 = self.attn.in_proj_weight
primals_3 = self.attn.in_proj_bias
primals_1 = self.attn.out_proj.weight
primals_5 = self.attn.out_proj.bias
primals_4 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| LiuXiaoxuanPKU/actnn-mmcls | MultiheadAttention | false | 5,555 | [
"Apache-2.0"
] | 1 | c97d1116d54ddb3f9b1e51baebe25ffb2b3f7b75 | https://github.com/LiuXiaoxuanPKU/actnn-mmcls/tree/c97d1116d54ddb3f9b1e51baebe25ffb2b3f7b75 | import torch
import torch.nn as nn
class Model(nn.Module):
"""A warpper for torch.nn.MultiheadAttention.
This module implements MultiheadAttention with residual connection.
Args:
embed_dims (int): The embedding dimension.
num_heads (int): Parallel attention heads. Same as
`nn.MultiheadAttention`.
attn_drop (float): A Dropout layer on attn_output_weights. Default 0.0.
proj_drop (float): The drop out rate after attention. Default 0.0.
"""
def __init__(self, embed_dims, num_heads, attn_drop=0.0, proj_drop=0.0):
super().__init__()
assert embed_dims % num_heads == 0, f'embed_dims must be divisible by num_heads. got {embed_dims} and {num_heads}.'
self.embed_dims = embed_dims
self.num_heads = num_heads
self.attn = nn.MultiheadAttention(embed_dims, num_heads, attn_drop)
self.dropout = nn.Dropout(proj_drop)
def forward(self, x, key=None, value=None, residual=None, query_pos=
None, key_pos=None, attn_mask=None, key_padding_mask=None):
"""Forward function for `MultiheadAttention`.
Args:
x (Tensor): The input query with shape [num_query, bs,
embed_dims]. Same in `nn.MultiheadAttention.forward`.
key (Tensor): The key tensor with shape [num_key, bs,
embed_dims]. Same in `nn.MultiheadAttention.forward`.
Default None. If None, the `query` will be used.
value (Tensor): The value tensor with same shape as `key`.
Same in `nn.MultiheadAttention.forward`. Default None.
If None, the `key` will be used.
residual (Tensor): The tensor used for addition, with the
same shape as `x`. Default None. If None, `x` will be used.
query_pos (Tensor): The positional encoding for query, with
the same shape as `x`. Default None. If not None, it will
be added to `x` before forward function.
key_pos (Tensor): The positional encoding for `key`, with the
same shape as `key`. Default None. If not None, it will
be added to `key` before forward function. If None, and
`query_pos` has the same shape as `key`, then `query_pos`
will be used for `key_pos`.
attn_mask (Tensor): ByteTensor mask with shape [num_query,
num_key]. Same in `nn.MultiheadAttention.forward`.
Default None.
key_padding_mask (Tensor): ByteTensor with shape [bs, num_key].
Same in `nn.MultiheadAttention.forward`. Default None.
Returns:
Tensor: forwarded results with shape [num_query, bs, embed_dims].
"""
query = x
if key is None:
key = query
if value is None:
value = key
if residual is None:
residual = x
if key_pos is None:
if query_pos is not None and key is not None:
if query_pos.shape == key.shape:
key_pos = query_pos
if query_pos is not None:
query = query + query_pos
if key_pos is not None:
key = key + key_pos
out = self.attn(query, key, value=value, attn_mask=attn_mask,
key_padding_mask=key_padding_mask)[0]
return residual + self.dropout(out)
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [4, 4]
|
SMAPELoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/l5/cl5owl63sm4p4y6vlb33daka3yidfydimcncohmywgjaogtb5o5f.py
# Topologically Sorted Source Nodes: [sub, abs_1, abs_2, abs_3, add, add_1, truediv, mean], Original ATen: [aten.sub, aten.abs, aten.add, aten.div, aten.mean]
# Source node to ATen node mapping:
# abs_1 => abs_1
# abs_2 => abs_2
# abs_3 => abs_3
# add => add
# add_1 => add_1
# mean => mean
# sub => sub
# truediv => div
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%sub,), kwargs = {})
# %abs_2 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%arg0_1,), kwargs = {})
# %abs_3 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%arg1_1,), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%abs_2, %abs_3), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add, 0.01), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%abs_1, %add_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%div,), kwargs = {})
triton_per_fused_abs_add_div_mean_sub_0 = async_compile.triton('triton_per_fused_abs_add_div_mean_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_abs_add_div_mean_sub_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_abs_add_div_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = tl_math.abs(tmp0)
tmp5 = tl_math.abs(tmp1)
tmp6 = tmp4 + tmp5
tmp7 = 0.01
tmp8 = tmp6 + tmp7
tmp9 = tmp3 / tmp8
tmp10 = tl.broadcast_to(tmp9, [RBLOCK])
tmp12 = triton_helpers.promote_to_tensor(tl.sum(tmp10, 0))
tmp13 = 256.0
tmp14 = tmp12 / tmp13
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp14, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [sub, abs_1, abs_2, abs_3, add, add_1, truediv, mean], Original ATen: [aten.sub, aten.abs, aten.add, aten.div, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_abs_add_div_mean_sub_0.run(buf1, arg0_1, arg1_1, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class SMAPELoss(nn.Module):
def forward(self, input, target):
return (torch.abs(input - target) / (torch.abs(input) + torch.abs(
target) + 0.01)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_add_div_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = tl_math.abs(tmp0)
tmp5 = tl_math.abs(tmp1)
tmp6 = tmp4 + tmp5
tmp7 = 0.01
tmp8 = tmp6 + tmp7
tmp9 = tmp3 / tmp8
tmp10 = tl.broadcast_to(tmp9, [RBLOCK])
tmp12 = triton_helpers.promote_to_tensor(tl.sum(tmp10, 0))
tmp13 = 256.0
tmp14 = tmp12 / tmp13
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp14, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_add_div_mean_sub_0[grid(1)](buf1, arg0_1,
arg1_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
class SMAPELossNew(nn.Module):
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| LongerVision/oidn | SMAPELoss | false | 5,556 | [
"Apache-2.0"
] | 1 | 2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | https://github.com/LongerVision/oidn/tree/2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | import torch
import torch.nn as nn
class Model(nn.Module):
def forward(self, input, target):
return (torch.abs(input - target) / (torch.abs(input) + torch.abs(
target) + 0.01)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GradientLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/mc/cmcqirvp2mlpcb2uj7frf6qjesljm7rlah5n553gwdx35dyjovhe.py
# Topologically Sorted Source Nodes: [cat, cat_1, sub_4, abs_1, mean], Original ATen: [aten.cat, aten.sub, aten.abs, aten.mean]
# Source node to ATen node mapping:
# abs_1 => abs_1
# cat => cat
# cat_1 => cat_1
# mean => mean
# sub_4 => sub_4
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%sub, %sub_1], -3), kwargs = {})
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%sub_2, %sub_3], -3), kwargs = {})
# %sub_4 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%cat, %cat_1), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%sub_4,), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%abs_1,), kwargs = {})
triton_per_fused_abs_cat_mean_sub_0 = async_compile.triton('triton_per_fused_abs_cat_mean_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 512],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_abs_cat_mean_sub_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 8, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_abs_cat_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 288
RBLOCK: tl.constexpr = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = rindex < rnumel
r2 = (rindex // 9) % 8
r0 = rindex % 3
r1 = (rindex // 3) % 3
r3 = (rindex // 72)
tmp0 = r2
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (tl.broadcast_to(4 + r0 + (4*r1) + (16*r2) + (64*r3), [RBLOCK])), rmask & tmp4, other=0.0)
tmp6 = tl.load(in_ptr0 + (tl.broadcast_to(r0 + (4*r1) + (16*r2) + (64*r3), [RBLOCK])), rmask & tmp4, other=0.0)
tmp7 = tmp5 - tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tmp11 = tl.full([1], 8, tl.int64)
tmp12 = tmp0 < tmp11
tmp13 = tl.load(in_ptr0 + (tl.broadcast_to(1 + r0 + (4*r1) + (16*((-4) + r2)) + (64*r3), [RBLOCK])), rmask & tmp10, other=0.0)
tmp14 = tl.load(in_ptr0 + (tl.broadcast_to(r0 + (4*r1) + (16*((-4) + r2)) + (64*r3), [RBLOCK])), rmask & tmp10, other=0.0)
tmp15 = tmp13 - tmp14
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp10, tmp15, tmp16)
tmp18 = tl.where(tmp4, tmp9, tmp17)
tmp19 = tl.load(in_ptr1 + (tl.broadcast_to(4 + r0 + (4*r1) + (16*r2) + (64*r3), [RBLOCK])), rmask & tmp4, other=0.0)
tmp20 = tl.load(in_ptr1 + (tl.broadcast_to(r0 + (4*r1) + (16*r2) + (64*r3), [RBLOCK])), rmask & tmp4, other=0.0)
tmp21 = tmp19 - tmp20
tmp22 = tl.full(tmp21.shape, 0.0, tmp21.dtype)
tmp23 = tl.where(tmp4, tmp21, tmp22)
tmp24 = tl.load(in_ptr1 + (tl.broadcast_to(1 + r0 + (4*r1) + (16*((-4) + r2)) + (64*r3), [RBLOCK])), rmask & tmp10, other=0.0)
tmp25 = tl.load(in_ptr1 + (tl.broadcast_to(r0 + (4*r1) + (16*((-4) + r2)) + (64*r3), [RBLOCK])), rmask & tmp10, other=0.0)
tmp26 = tmp24 - tmp25
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp10, tmp26, tmp27)
tmp29 = tl.where(tmp4, tmp23, tmp28)
tmp30 = tmp18 - tmp29
tmp31 = tl_math.abs(tmp30)
tmp32 = tl.broadcast_to(tmp31, [RBLOCK])
tmp34 = tl.where(rmask, tmp32, 0)
tmp35 = triton_helpers.promote_to_tensor(tl.sum(tmp34, 0))
tmp36 = 288.0
tmp37 = tmp35 / tmp36
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp37, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [cat, cat_1, sub_4, abs_1, mean], Original ATen: [aten.cat, aten.sub, aten.abs, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_abs_cat_mean_sub_0.run(buf1, arg0_1, arg1_1, 1, 288, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def gradient(input):
input0 = input[..., :-1, :-1]
didy = input[..., 1:, :-1] - input0
didx = input[..., :-1, 1:] - input0
return torch.cat((didy, didx), -3)
class GradientLoss(nn.Module):
def forward(self, input, target):
return torch.abs(gradient(input) - gradient(target)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_cat_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, rnumel):
XBLOCK: tl.constexpr = 1
rnumel = 288
RBLOCK: tl.constexpr = 512
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
rmask = rindex < rnumel
r2 = rindex // 9 % 8
r0 = rindex % 3
r1 = rindex // 3 % 3
r3 = rindex // 72
tmp0 = r2
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + tl.broadcast_to(4 + r0 + 4 * r1 + 16 * r2 + 64 *
r3, [RBLOCK]), rmask & tmp4, other=0.0)
tmp6 = tl.load(in_ptr0 + tl.broadcast_to(r0 + 4 * r1 + 16 * r2 + 64 *
r3, [RBLOCK]), rmask & tmp4, other=0.0)
tmp7 = tmp5 - tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp13 = tl.load(in_ptr0 + tl.broadcast_to(1 + r0 + 4 * r1 + 16 * (-4 +
r2) + 64 * r3, [RBLOCK]), rmask & tmp10, other=0.0)
tmp14 = tl.load(in_ptr0 + tl.broadcast_to(r0 + 4 * r1 + 16 * (-4 + r2) +
64 * r3, [RBLOCK]), rmask & tmp10, other=0.0)
tmp15 = tmp13 - tmp14
tmp16 = tl.full(tmp15.shape, 0.0, tmp15.dtype)
tmp17 = tl.where(tmp10, tmp15, tmp16)
tmp18 = tl.where(tmp4, tmp9, tmp17)
tmp19 = tl.load(in_ptr1 + tl.broadcast_to(4 + r0 + 4 * r1 + 16 * r2 +
64 * r3, [RBLOCK]), rmask & tmp4, other=0.0)
tmp20 = tl.load(in_ptr1 + tl.broadcast_to(r0 + 4 * r1 + 16 * r2 + 64 *
r3, [RBLOCK]), rmask & tmp4, other=0.0)
tmp21 = tmp19 - tmp20
tmp22 = tl.full(tmp21.shape, 0.0, tmp21.dtype)
tmp23 = tl.where(tmp4, tmp21, tmp22)
tmp24 = tl.load(in_ptr1 + tl.broadcast_to(1 + r0 + 4 * r1 + 16 * (-4 +
r2) + 64 * r3, [RBLOCK]), rmask & tmp10, other=0.0)
tmp25 = tl.load(in_ptr1 + tl.broadcast_to(r0 + 4 * r1 + 16 * (-4 + r2) +
64 * r3, [RBLOCK]), rmask & tmp10, other=0.0)
tmp26 = tmp24 - tmp25
tmp27 = tl.full(tmp26.shape, 0.0, tmp26.dtype)
tmp28 = tl.where(tmp10, tmp26, tmp27)
tmp29 = tl.where(tmp4, tmp23, tmp28)
tmp30 = tmp18 - tmp29
tmp31 = tl_math.abs(tmp30)
tmp32 = tl.broadcast_to(tmp31, [RBLOCK])
tmp34 = tl.where(rmask, tmp32, 0)
tmp35 = triton_helpers.promote_to_tensor(tl.sum(tmp34, 0))
tmp36 = 288.0
tmp37 = tmp35 / tmp36
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp37, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_cat_mean_sub_0[grid(1)](buf1, arg0_1, arg1_1,
1, 288, num_warps=4, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def gradient(input):
input0 = input[..., :-1, :-1]
didy = input[..., 1:, :-1] - input0
didx = input[..., :-1, 1:] - input0
return torch.cat((didy, didx), -3)
class GradientLossNew(nn.Module):
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| LongerVision/oidn | GradientLoss | false | 5,557 | [
"Apache-2.0"
] | 1 | 2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | https://github.com/LongerVision/oidn/tree/2f9e59f8b747b217f78c5c274f4f2bff347a03a7 | import torch
import torch.nn as nn
def gradient(input):
input0 = input[..., :-1, :-1]
didy = input[..., 1:, :-1] - input0
didx = input[..., :-1, 1:] - input0
return torch.cat((didy, didx), -3)
class Model(nn.Module):
def forward(self, input, target):
return torch.abs(gradient(input) - gradient(target)).mean()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
ResHead | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/l3/cl35tzbhrd24dhunkbb6gjs54aklpyr46oikqhoylcgmkcmhujil.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# x => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%primals_1, [-1, -2], True), kwargs = {})
triton_per_fused_mean_0 = async_compile.triton('triton_per_fused_mean_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[16, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mean_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mean_0(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 16
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (16*x0)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tmp5 = 16.0
tmp6 = tmp4 / tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + (x0), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 16, 16), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_mean_0.run(buf1, primals_1, 16, 16, grid=grid(16), stream=stream0)
del primals_1
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_3, reinterpret_tensor(buf1, (4, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del primals_2
del primals_3
return (buf2, reinterpret_tensor(buf1, (4, 4), (4, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from torch.nn import Module
import torch
import torch.nn as nn
import torch.utils.data
def gap2d(_w_in):
"""Helper for building a gap2d layer."""
return nn.AdaptiveAvgPool2d((1, 1))
def gap2d_cx(cx, _w_in):
"""Accumulates complexity of gap2d into cx = (h, w, flops, params, acts)."""
flops, params, acts = cx['flops'], cx['params'], cx['acts']
return {'h': 1, 'w': 1, 'flops': flops, 'params': params, 'acts': acts}
def linear(w_in, w_out, *, bias=False):
"""Helper for building a linear layer."""
return nn.Linear(w_in, w_out, bias=bias)
def linear_cx(cx, w_in, w_out, *, bias=False, num_locations=1):
"""Accumulates complexity of linear into cx = (h, w, flops, params, acts)."""
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
flops += w_in * w_out * num_locations + (w_out * num_locations if bias else
0)
params += w_in * w_out + (w_out if bias else 0)
acts += w_out * num_locations
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class ResHead(Module):
"""ResNet head: AvgPool, 1x1."""
def __init__(self, w_in, num_classes):
super(ResHead, self).__init__()
self.avg_pool = gap2d(w_in)
self.fc = linear(w_in, num_classes, bias=True)
def forward(self, x):
x = self.avg_pool(x)
x = x.view(x.size(0), -1)
x = self.fc(x)
return x
@staticmethod
def complexity(cx, w_in, num_classes):
cx = gap2d_cx(cx, w_in)
cx = linear_cx(cx, w_in, num_classes, bias=True)
return cx
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'w_in': 4, 'num_classes': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch.nn import Module
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_mean_0(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK:
tl.constexpr):
xnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 16 * x0), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tmp5 = 16.0
tmp6 = tmp4 / tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 16, 16), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_mean_0[grid(16)](buf1, primals_1, 16, 16, XBLOCK=8,
num_warps=2, num_stages=1)
del primals_1
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(buf1, (4, 4), (4,
1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), alpha
=1, beta=1, out=buf2)
del primals_2
del primals_3
return buf2, reinterpret_tensor(buf1, (4, 4), (4, 1), 0)
def gap2d(_w_in):
"""Helper for building a gap2d layer."""
return nn.AdaptiveAvgPool2d((1, 1))
def gap2d_cx(cx, _w_in):
"""Accumulates complexity of gap2d into cx = (h, w, flops, params, acts)."""
flops, params, acts = cx['flops'], cx['params'], cx['acts']
return {'h': 1, 'w': 1, 'flops': flops, 'params': params, 'acts': acts}
def linear(w_in, w_out, *, bias=False):
"""Helper for building a linear layer."""
return nn.Linear(w_in, w_out, bias=bias)
def linear_cx(cx, w_in, w_out, *, bias=False, num_locations=1):
"""Accumulates complexity of linear into cx = (h, w, flops, params, acts)."""
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
flops += w_in * w_out * num_locations + (w_out * num_locations if bias else
0)
params += w_in * w_out + (w_out if bias else 0)
acts += w_out * num_locations
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class ResHeadNew(Module):
"""ResNet head: AvgPool, 1x1."""
def __init__(self, w_in, num_classes):
super(ResHeadNew, self).__init__()
self.avg_pool = gap2d(w_in)
self.fc = linear(w_in, num_classes, bias=True)
@staticmethod
def complexity(cx, w_in, num_classes):
cx = gap2d_cx(cx, w_in)
cx = linear_cx(cx, w_in, num_classes, bias=True)
return cx
def forward(self, input_0):
primals_2 = self.fc.weight
primals_3 = self.fc.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| MAC-AutoML/XCompression | ResHead | false | 5,558 | [
"MIT"
] | 1 | 9f76eb3ccfb3057110ecf12aa48dec00a4667a25 | https://github.com/MAC-AutoML/XCompression/tree/9f76eb3ccfb3057110ecf12aa48dec00a4667a25 | from torch.nn import Module
import torch
import torch.nn as nn
import torch.utils.data
def gap2d(_w_in):
"""Helper for building a gap2d layer."""
return nn.AdaptiveAvgPool2d((1, 1))
def gap2d_cx(cx, _w_in):
"""Accumulates complexity of gap2d into cx = (h, w, flops, params, acts)."""
flops, params, acts = cx['flops'], cx['params'], cx['acts']
return {'h': 1, 'w': 1, 'flops': flops, 'params': params, 'acts': acts}
def linear(w_in, w_out, *, bias=False):
"""Helper for building a linear layer."""
return nn.Linear(w_in, w_out, bias=bias)
def linear_cx(cx, w_in, w_out, *, bias=False, num_locations=1):
"""Accumulates complexity of linear into cx = (h, w, flops, params, acts)."""
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
flops += w_in * w_out * num_locations + (w_out * num_locations if bias else
0)
params += w_in * w_out + (w_out if bias else 0)
acts += w_out * num_locations
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class Model(Module):
"""ResNet head: AvgPool, 1x1."""
def __init__(self, w_in, num_classes):
super().__init__()
self.avg_pool = gap2d(w_in)
self.fc = linear(w_in, num_classes, bias=True)
def forward(self, x):
x = self.avg_pool(x)
x = x.view(x.size(0), -1)
x = self.fc(x)
return x
@staticmethod
def complexity(cx, w_in, num_classes):
cx = gap2d_cx(cx, w_in)
cx = linear_cx(cx, w_in, num_classes, bias=True)
return cx
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
MaxPool | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ud/cude6zl4nio2ly5l3l5cwlmxkoqtt4qkekbvrzk6nz7rpwc6ypf3.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# x => getitem
# Graph fragment:
# %getitem : [num_users=1] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets, 0), kwargs = {})
triton_poi_fused_max_pool2d_with_indices_0 = async_compile.triton('triton_poi_fused_max_pool2d_with_indices_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_max_pool2d_with_indices_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 16, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 3) % 3
x0 = xindex % 3
x2 = (xindex // 9)
x4 = xindex
tmp0 = (-1) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = (-1) + x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp10 & xmask, other=float("-inf"))
tmp12 = x0
tmp13 = tmp12 >= tmp1
tmp14 = tmp12 < tmp3
tmp15 = tmp13 & tmp14
tmp16 = tmp5 & tmp15
tmp17 = tl.load(in_ptr0 + ((-4) + x0 + (4*x1) + (16*x2)), tmp16 & xmask, other=float("-inf"))
tmp18 = triton_helpers.maximum(tmp17, tmp11)
tmp19 = 1 + x0
tmp20 = tmp19 >= tmp1
tmp21 = tmp19 < tmp3
tmp22 = tmp20 & tmp21
tmp23 = tmp5 & tmp22
tmp24 = tl.load(in_ptr0 + ((-3) + x0 + (4*x1) + (16*x2)), tmp23 & xmask, other=float("-inf"))
tmp25 = triton_helpers.maximum(tmp24, tmp18)
tmp26 = 2 + x0
tmp27 = tmp26 >= tmp1
tmp28 = tmp26 < tmp3
tmp29 = tmp27 & tmp28
tmp30 = tmp5 & tmp29
tmp31 = tl.load(in_ptr0 + ((-2) + x0 + (4*x1) + (16*x2)), tmp30 & xmask, other=float("-inf"))
tmp32 = triton_helpers.maximum(tmp31, tmp25)
tmp33 = x1
tmp34 = tmp33 >= tmp1
tmp35 = tmp33 < tmp3
tmp36 = tmp34 & tmp35
tmp37 = tmp36 & tmp9
tmp38 = tl.load(in_ptr0 + ((-1) + x0 + (4*x1) + (16*x2)), tmp37 & xmask, other=float("-inf"))
tmp39 = triton_helpers.maximum(tmp38, tmp32)
tmp40 = tmp36 & tmp15
tmp41 = tl.load(in_ptr0 + (x0 + (4*x1) + (16*x2)), tmp40 & xmask, other=float("-inf"))
tmp42 = triton_helpers.maximum(tmp41, tmp39)
tmp43 = tmp36 & tmp22
tmp44 = tl.load(in_ptr0 + (1 + x0 + (4*x1) + (16*x2)), tmp43 & xmask, other=float("-inf"))
tmp45 = triton_helpers.maximum(tmp44, tmp42)
tmp46 = tmp36 & tmp29
tmp47 = tl.load(in_ptr0 + (2 + x0 + (4*x1) + (16*x2)), tmp46 & xmask, other=float("-inf"))
tmp48 = triton_helpers.maximum(tmp47, tmp45)
tmp49 = 1 + x1
tmp50 = tmp49 >= tmp1
tmp51 = tmp49 < tmp3
tmp52 = tmp50 & tmp51
tmp53 = tmp52 & tmp9
tmp54 = tl.load(in_ptr0 + (3 + x0 + (4*x1) + (16*x2)), tmp53 & xmask, other=float("-inf"))
tmp55 = triton_helpers.maximum(tmp54, tmp48)
tmp56 = tmp52 & tmp15
tmp57 = tl.load(in_ptr0 + (4 + x0 + (4*x1) + (16*x2)), tmp56 & xmask, other=float("-inf"))
tmp58 = triton_helpers.maximum(tmp57, tmp55)
tmp59 = tmp52 & tmp22
tmp60 = tl.load(in_ptr0 + (5 + x0 + (4*x1) + (16*x2)), tmp59 & xmask, other=float("-inf"))
tmp61 = triton_helpers.maximum(tmp60, tmp58)
tmp62 = tmp52 & tmp29
tmp63 = tl.load(in_ptr0 + (6 + x0 + (4*x1) + (16*x2)), tmp62 & xmask, other=float("-inf"))
tmp64 = triton_helpers.maximum(tmp63, tmp61)
tmp65 = 2 + x1
tmp66 = tmp65 >= tmp1
tmp67 = tmp65 < tmp3
tmp68 = tmp66 & tmp67
tmp69 = tmp68 & tmp9
tmp70 = tl.load(in_ptr0 + (7 + x0 + (4*x1) + (16*x2)), tmp69 & xmask, other=float("-inf"))
tmp71 = triton_helpers.maximum(tmp70, tmp64)
tmp72 = tmp68 & tmp15
tmp73 = tl.load(in_ptr0 + (8 + x0 + (4*x1) + (16*x2)), tmp72 & xmask, other=float("-inf"))
tmp74 = triton_helpers.maximum(tmp73, tmp71)
tmp75 = tmp68 & tmp22
tmp76 = tl.load(in_ptr0 + (9 + x0 + (4*x1) + (16*x2)), tmp75 & xmask, other=float("-inf"))
tmp77 = triton_helpers.maximum(tmp76, tmp74)
tmp78 = tmp68 & tmp29
tmp79 = tl.load(in_ptr0 + (10 + x0 + (4*x1) + (16*x2)), tmp78 & xmask, other=float("-inf"))
tmp80 = triton_helpers.maximum(tmp79, tmp77)
tl.store(out_ptr0 + (x4), tmp80, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.max_pool2d_with_indices]
stream0 = get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0.run(arg0_1, buf0, 144, grid=grid(144), stream=stream0)
del arg0_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.optim
import torch.utils.data
class MaxPool(nn.Module):
def __init__(self, kernel_size, stride=1, padding=1, zero_pad=False):
super(MaxPool, self).__init__()
self.zero_pad = nn.ZeroPad2d((1, 0, 1, 0)) if zero_pad else None
self.pool = nn.MaxPool2d(kernel_size, stride=stride, padding=padding)
def forward(self, x):
if self.zero_pad:
x = self.zero_pad(x)
x = self.pool(x)
if self.zero_pad:
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'kernel_size': 4}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
import torch.optim
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 3 % 3
x0 = xindex % 3
x2 = xindex // 9
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = -1 + x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp10 & xmask,
other=float('-inf'))
tmp12 = x0
tmp13 = tmp12 >= tmp1
tmp14 = tmp12 < tmp3
tmp15 = tmp13 & tmp14
tmp16 = tmp5 & tmp15
tmp17 = tl.load(in_ptr0 + (-4 + x0 + 4 * x1 + 16 * x2), tmp16 & xmask,
other=float('-inf'))
tmp18 = triton_helpers.maximum(tmp17, tmp11)
tmp19 = 1 + x0
tmp20 = tmp19 >= tmp1
tmp21 = tmp19 < tmp3
tmp22 = tmp20 & tmp21
tmp23 = tmp5 & tmp22
tmp24 = tl.load(in_ptr0 + (-3 + x0 + 4 * x1 + 16 * x2), tmp23 & xmask,
other=float('-inf'))
tmp25 = triton_helpers.maximum(tmp24, tmp18)
tmp26 = 2 + x0
tmp27 = tmp26 >= tmp1
tmp28 = tmp26 < tmp3
tmp29 = tmp27 & tmp28
tmp30 = tmp5 & tmp29
tmp31 = tl.load(in_ptr0 + (-2 + x0 + 4 * x1 + 16 * x2), tmp30 & xmask,
other=float('-inf'))
tmp32 = triton_helpers.maximum(tmp31, tmp25)
tmp33 = x1
tmp34 = tmp33 >= tmp1
tmp35 = tmp33 < tmp3
tmp36 = tmp34 & tmp35
tmp37 = tmp36 & tmp9
tmp38 = tl.load(in_ptr0 + (-1 + x0 + 4 * x1 + 16 * x2), tmp37 & xmask,
other=float('-inf'))
tmp39 = triton_helpers.maximum(tmp38, tmp32)
tmp40 = tmp36 & tmp15
tmp41 = tl.load(in_ptr0 + (x0 + 4 * x1 + 16 * x2), tmp40 & xmask, other
=float('-inf'))
tmp42 = triton_helpers.maximum(tmp41, tmp39)
tmp43 = tmp36 & tmp22
tmp44 = tl.load(in_ptr0 + (1 + x0 + 4 * x1 + 16 * x2), tmp43 & xmask,
other=float('-inf'))
tmp45 = triton_helpers.maximum(tmp44, tmp42)
tmp46 = tmp36 & tmp29
tmp47 = tl.load(in_ptr0 + (2 + x0 + 4 * x1 + 16 * x2), tmp46 & xmask,
other=float('-inf'))
tmp48 = triton_helpers.maximum(tmp47, tmp45)
tmp49 = 1 + x1
tmp50 = tmp49 >= tmp1
tmp51 = tmp49 < tmp3
tmp52 = tmp50 & tmp51
tmp53 = tmp52 & tmp9
tmp54 = tl.load(in_ptr0 + (3 + x0 + 4 * x1 + 16 * x2), tmp53 & xmask,
other=float('-inf'))
tmp55 = triton_helpers.maximum(tmp54, tmp48)
tmp56 = tmp52 & tmp15
tmp57 = tl.load(in_ptr0 + (4 + x0 + 4 * x1 + 16 * x2), tmp56 & xmask,
other=float('-inf'))
tmp58 = triton_helpers.maximum(tmp57, tmp55)
tmp59 = tmp52 & tmp22
tmp60 = tl.load(in_ptr0 + (5 + x0 + 4 * x1 + 16 * x2), tmp59 & xmask,
other=float('-inf'))
tmp61 = triton_helpers.maximum(tmp60, tmp58)
tmp62 = tmp52 & tmp29
tmp63 = tl.load(in_ptr0 + (6 + x0 + 4 * x1 + 16 * x2), tmp62 & xmask,
other=float('-inf'))
tmp64 = triton_helpers.maximum(tmp63, tmp61)
tmp65 = 2 + x1
tmp66 = tmp65 >= tmp1
tmp67 = tmp65 < tmp3
tmp68 = tmp66 & tmp67
tmp69 = tmp68 & tmp9
tmp70 = tl.load(in_ptr0 + (7 + x0 + 4 * x1 + 16 * x2), tmp69 & xmask,
other=float('-inf'))
tmp71 = triton_helpers.maximum(tmp70, tmp64)
tmp72 = tmp68 & tmp15
tmp73 = tl.load(in_ptr0 + (8 + x0 + 4 * x1 + 16 * x2), tmp72 & xmask,
other=float('-inf'))
tmp74 = triton_helpers.maximum(tmp73, tmp71)
tmp75 = tmp68 & tmp22
tmp76 = tl.load(in_ptr0 + (9 + x0 + 4 * x1 + 16 * x2), tmp75 & xmask,
other=float('-inf'))
tmp77 = triton_helpers.maximum(tmp76, tmp74)
tmp78 = tmp68 & tmp29
tmp79 = tl.load(in_ptr0 + (10 + x0 + 4 * x1 + 16 * x2), tmp78 & xmask,
other=float('-inf'))
tmp80 = triton_helpers.maximum(tmp79, tmp77)
tl.store(out_ptr0 + x4, tmp80, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0[grid(144)](arg0_1, buf0,
144, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class MaxPoolNew(nn.Module):
def __init__(self, kernel_size, stride=1, padding=1, zero_pad=False):
super(MaxPoolNew, self).__init__()
self.zero_pad = nn.ZeroPad2d((1, 0, 1, 0)) if zero_pad else None
self.pool = nn.MaxPool2d(kernel_size, stride=stride, padding=padding)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| LongKt7/Face_Recognize_Pytorch | MaxPool | false | 5,559 | [
"MIT"
] | 1 | baa02e633d379abe1001c8b8acb942617177329c | https://github.com/LongKt7/Face_Recognize_Pytorch/tree/baa02e633d379abe1001c8b8acb942617177329c | import torch
import torch.nn as nn
import torch.optim
import torch.utils.data
class Model(nn.Module):
def __init__(self, kernel_size, stride=1, padding=1, zero_pad=False):
super().__init__()
self.zero_pad = nn.ZeroPad2d((1, 0, 1, 0)) if zero_pad else None
self.pool = nn.MaxPool2d(kernel_size, stride=stride, padding=padding)
def forward(self, x):
if self.zero_pad:
x = self.zero_pad(x)
x = self.pool(x)
if self.zero_pad:
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
UsBlock_nounpool | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/rl/crlzvualg2za225vkja5fjpnxss76chgi2glfm2hpmsgp2hm7zo3.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_1 => relu
# Graph fragment:
# %relu : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%squeeze,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%squeeze_5, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 64)
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(1, 1, 1), padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf2, 256, grid=grid(256), stream=stream0)
del primals_2
return (buf1, primals_1, reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 3, 3, 3), (108, 27, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class UsBlock_nounpool(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super(UsBlock_nounpool, self).__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
x = self.conv(x)
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 64
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1,
4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(1, 1,
1), padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_2, buf2, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
return buf1, primals_1, reinterpret_tensor(primals_3, (1, 4, 4, 4, 4),
(256, 64, 16, 4, 1), 0), buf2
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class UsBlock_nounpoolNew(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super(UsBlock_nounpoolNew, self).__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, input_0):
primals_1 = self.conv.weight
primals_2 = self.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| MATHplus-Young-Academy/P2-Cardiac-Motion | UsBlock_nounpool | false | 5,560 | [
"Apache-2.0"
] | 1 | 844995e8e5760f981c425d13c0bd7f2f3bb8baec | https://github.com/MATHplus-Young-Academy/P2-Cardiac-Motion/tree/844995e8e5760f981c425d13c0bd7f2f3bb8baec | import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class Model(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super().__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
x = self.conv(x)
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
WDV29LayerNormalization | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/hp/chpdwpegv6lvistek2wqgimtufecqvfp6grp5rpblk5yjicjzqd2.py
# Topologically Sorted Source Nodes: [layer_norm], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# layer_norm => add_2, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_7, [3]), kwargs = {correction: 0, keepdim: True})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
triton_poi_fused_native_layer_norm_0 = async_compile.triton('triton_poi_fused_native_layer_norm_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_native_layer_norm_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + (x0), tmp8, xmask)
tl.store(out_ptr1 + (x0), tmp23, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/el/celxf6tvonm4hnojqioclt5acflsj3tnnwh4e7fon3fxf7kiua46.py
# Topologically Sorted Source Nodes: [tanh, mul, weight, tanh_1, mul_1, bias, layer_norm], Original ATen: [aten.tanh, aten.mul, aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# bias => add_1
# layer_norm => add_2, add_3, mul_2, mul_3, rsqrt, sub, var_mean
# mul => mul
# mul_1 => mul_1
# tanh => tanh
# tanh_1 => tanh_1
# weight => add
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%primals_1,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_3, %tanh), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_4), kwargs = {})
# %tanh_1 : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%primals_2,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_5, %tanh_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_6), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_7, [3]), kwargs = {correction: 0, keepdim: True})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_7, %getitem_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_2, %add), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_3, %add_1), kwargs = {})
triton_poi_fused_add_mul_native_layer_norm_tanh_1 = async_compile.triton('triton_poi_fused_add_mul_native_layer_norm_tanh_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: '*fp32', 8: '*fp32', 9: '*fp32', 10: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_native_layer_norm_tanh_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 9, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_native_layer_norm_tanh_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, in_ptr6, in_ptr7, in_ptr8, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr6 + (x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr7 + (x0), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr8 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp7 = libdevice.tanh(tmp6)
tmp8 = tmp5 * tmp7
tmp10 = tmp8 + tmp9
tmp11 = tmp4 * tmp10
tmp14 = libdevice.tanh(tmp13)
tmp15 = tmp12 * tmp14
tmp17 = tmp15 + tmp16
tmp18 = tmp11 + tmp17
tl.store(out_ptr0 + (x2), tmp18, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, ), (1, ))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf1 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [layer_norm], Original ATen: [aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_poi_fused_native_layer_norm_0.run(primals_7, buf0, buf1, 64, grid=grid(64), stream=stream0)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh, mul, weight, tanh_1, mul_1, bias, layer_norm], Original ATen: [aten.tanh, aten.mul, aten.add, aten.native_layer_norm]
triton_poi_fused_add_mul_native_layer_norm_tanh_1.run(primals_7, buf0, buf1, primals_3, primals_1, primals_4, primals_5, primals_2, primals_6, buf2, 256, grid=grid(256), stream=stream0)
del buf0
del buf1
del primals_3
del primals_4
del primals_5
del primals_6
return (buf2, primals_1, primals_2, primals_7, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import numbers
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class WDV29LayerNormalization(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, normalized_shape, wd_require_gradient=False, eps=
1e-05, elementwise_affine=True):
super(WDV29LayerNormalization, self).__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.eps = eps
self.elementwise_affine = elementwise_affine
if self.elementwise_affine:
self.weight = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.bias = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_bias_weight = Parameter(torch.Tensor(*normalized_shape))
self.tanh_bias_bias = Parameter(torch.Tensor(*normalized_shape))
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
if self.elementwise_affine:
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
init.ones_(self.weight)
init.zeros_(self.bias)
def forward(self, input):
weight = self.weight
bias = self.bias
weight = self.tanh_weight_weight * torch.tanh(weight
) + self.tanh_bias_weight
bias = self.tanh_weight_bias * torch.tanh(bias) + self.tanh_bias_bias
return F.layer_norm(input, self.normalized_shape, weight, bias,
self.eps)
def extra_repr(self):
return (
'{normalized_shape}, eps={eps}, elementwise_affine={elementwise_affine}'
.format(**self.__dict__))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'normalized_shape': 4}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import numbers
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + x0, tmp8, xmask)
tl.store(out_ptr1 + x0, tmp23, xmask)
@triton.jit
def triton_poi_fused_add_mul_native_layer_norm_tanh_1(in_ptr0, in_ptr1,
in_ptr2, in_ptr3, in_ptr4, in_ptr5, in_ptr6, in_ptr7, in_ptr8, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr6 + x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr7 + x0, xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr8 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp7 = libdevice.tanh(tmp6)
tmp8 = tmp5 * tmp7
tmp10 = tmp8 + tmp9
tmp11 = tmp4 * tmp10
tmp14 = libdevice.tanh(tmp13)
tmp15 = tmp12 * tmp14
tmp17 = tmp15 + tmp16
tmp18 = tmp11 + tmp17
tl.store(out_ptr0 + x2, tmp18, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4,), (1,))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf1 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(64)](primals_7, buf0,
buf1, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_mul_native_layer_norm_tanh_1[grid(256)](primals_7,
buf0, buf1, primals_3, primals_1, primals_4, primals_5,
primals_2, primals_6, buf2, 256, XBLOCK=128, num_warps=4,
num_stages=1)
del buf0
del buf1
del primals_3
del primals_4
del primals_5
del primals_6
return buf2, primals_1, primals_2, primals_7
class WDV29LayerNormalizationNew(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, normalized_shape, wd_require_gradient=False, eps=
1e-05, elementwise_affine=True):
super(WDV29LayerNormalizationNew, self).__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.eps = eps
self.elementwise_affine = elementwise_affine
if self.elementwise_affine:
self.weight = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.bias = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_bias_weight = Parameter(torch.Tensor(*normalized_shape))
self.tanh_bias_bias = Parameter(torch.Tensor(*normalized_shape))
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
if self.elementwise_affine:
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
init.ones_(self.weight)
init.zeros_(self.bias)
def extra_repr(self):
return (
'{normalized_shape}, eps={eps}, elementwise_affine={elementwise_affine}'
.format(**self.__dict__))
def forward(self, input_0):
primals_1 = self.weight
primals_2 = self.tanh_weight_weight
primals_3 = self.tanh_weight_bias
primals_4 = self.bias
primals_5 = self.tanh_bias_weight
primals_6 = self.tanh_bias_bias
primals_7 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| Lollipop321/weight-distillation | WDV29LayerNormalization | false | 5,561 | [
"BSD-3-Clause"
] | 1 | cfc76ec58e3e88094dde1825287b2968f9718431 | https://github.com/Lollipop321/weight-distillation/tree/cfc76ec58e3e88094dde1825287b2968f9718431 | import numbers
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class Model(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, normalized_shape, wd_require_gradient=False, eps=
1e-05, elementwise_affine=True):
super().__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.eps = eps
self.elementwise_affine = elementwise_affine
if self.elementwise_affine:
self.weight = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.bias = Parameter(torch.Tensor(*normalized_shape),
requires_grad=wd_require_gradient)
self.tanh_bias_weight = Parameter(torch.Tensor(*normalized_shape))
self.tanh_bias_bias = Parameter(torch.Tensor(*normalized_shape))
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parame
# ... truncated (>4000 chars) for memory efficiency |
WDV29Linear | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/rb/crbx6wdzdbdjlck75irsth4iy6smxxjgxz3q6l4of2njw7sqx3ug.py
# Topologically Sorted Source Nodes: [tanh, mul, weight], Original ATen: [aten.tanh, aten.mul, aten.add]
# Source node to ATen node mapping:
# mul => mul
# tanh => tanh
# weight => add
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%primals_1,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_2, %tanh), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_3), kwargs = {})
triton_poi_fused_add_mul_tanh_0 = async_compile.triton('triton_poi_fused_add_mul_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_tanh_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_tanh_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp4 = tl.load(in_ptr2 + (x0), xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + (x0), tmp5, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/j2/cj2txmrvo6fsowrygo4hnb7eq7omgzwo3kyzcghwvhkhqu5qxr4y.py
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias], Original ATen: [aten.tanh, aten.mul, aten.add]
# Source node to ATen node mapping:
# bias => add_1
# mul_1 => mul_1
# tanh_1 => tanh_1
# Graph fragment:
# %tanh_1 : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%primals_4,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_5, %tanh_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_6), kwargs = {})
triton_poi_fused_add_mul_tanh_1 = async_compile.triton('triton_poi_fused_add_mul_tanh_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_tanh_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_tanh_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp4 = tl.load(in_ptr2 + (x0), xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + (x0), tmp5, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh, mul, weight], Original ATen: [aten.tanh, aten.mul, aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mul_tanh_0.run(primals_2, primals_1, primals_3, buf0, 16, grid=grid(16), stream=stream0)
del primals_2
del primals_3
buf1 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias], Original ATen: [aten.tanh, aten.mul, aten.add]
triton_poi_fused_add_mul_tanh_1.run(primals_5, primals_4, primals_6, buf1, 4, grid=grid(4), stream=stream0)
del primals_5
del primals_6
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias, linear], Original ATen: [aten.tanh, aten.mul, aten.add, aten.addmm]
extern_kernels.addmm(buf1, reinterpret_tensor(primals_7, (64, 4), (4, 1), 0), reinterpret_tensor(buf0, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del buf0
del buf1
return (reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0), primals_1, primals_4, reinterpret_tensor(primals_7, (64, 4), (4, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class WDV29Linear(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, in_features, out_features, wd_require_gradient=False,
bias=True):
super(WDV29Linear, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.Tensor(out_features, in_features),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.bias = Parameter(torch.Tensor(out_features), requires_grad
=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight)
bound = 1 / math.sqrt(fan_in)
init.uniform_(self.bias, -bound, bound)
def forward(self, input):
weight = self.weight
weight = self.tanh_weight_weight * torch.tanh(weight
) + self.tanh_bias_weight
bias = self.bias
bias = self.tanh_weight_bias * torch.tanh(bias) + self.tanh_bias_bias
return F.linear(input, weight, bias)
def extra_repr(self):
return 'in_features={}, out_features={}, bias={}'.format(self.
in_features, self.out_features, self.bias is not None)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import math
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_tanh_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp4 = tl.load(in_ptr2 + x0, xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + x0, tmp5, xmask)
@triton.jit
def triton_poi_fused_add_mul_tanh_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp4 = tl.load(in_ptr2 + x0, xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + x0, tmp5, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_tanh_0[grid(16)](primals_2, primals_1,
primals_3, buf0, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_2
del primals_3
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
triton_poi_fused_add_mul_tanh_1[grid(4)](primals_5, primals_4,
primals_6, buf1, 4, XBLOCK=4, num_warps=1, num_stages=1)
del primals_5
del primals_6
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(buf1, reinterpret_tensor(primals_7, (64, 4), (
4, 1), 0), reinterpret_tensor(buf0, (4, 4), (1, 4), 0), alpha=1,
beta=1, out=buf2)
del buf0
del buf1
return reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0
), primals_1, primals_4, reinterpret_tensor(primals_7, (64, 4), (4,
1), 0)
class WDV29LinearNew(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, in_features, out_features, wd_require_gradient=False,
bias=True):
super(WDV29LinearNew, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.Tensor(out_features, in_features),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.bias = Parameter(torch.Tensor(out_features), requires_grad
=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight)
bound = 1 / math.sqrt(fan_in)
init.uniform_(self.bias, -bound, bound)
def extra_repr(self):
return 'in_features={}, out_features={}, bias={}'.format(self.
in_features, self.out_features, self.bias is not None)
def forward(self, input_0):
primals_1 = self.weight
primals_2 = self.tanh_weight_weight
primals_3 = self.tanh_bias_weight
primals_4 = self.bias
primals_5 = self.tanh_weight_bias
primals_6 = self.tanh_bias_bias
primals_7 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| Lollipop321/weight-distillation | WDV29Linear | false | 5,562 | [
"BSD-3-Clause"
] | 1 | cfc76ec58e3e88094dde1825287b2968f9718431 | https://github.com/Lollipop321/weight-distillation/tree/cfc76ec58e3e88094dde1825287b2968f9718431 | import math
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class Model(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, in_features, out_features, wd_require_gradient=False,
bias=True):
super().__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.Tensor(out_features, in_features),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.bias = Parameter(torch.Tensor(out_features), requires_grad
=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight)
bound = 1 / math.sqrt(fan_in)
init.uniform_(self.bias, -bound, bound)
def forward(self, input):
weight = self.weight
weight = self.tanh_weight_weight * torch.tanh(weight
) + self.tanh_bias_weight
bias = self.bias
bias = self.tanh_weight_bias * torch.tanh(bias) + self.tanh_bias_bias
return F.linear(input, weight, bias)
def extra_repr(self):
return 'in_features={}, out_features={}, bias={}'.format(self.
in_features, self.out_features, self.bias is not None)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
WDV52LayerNormalization | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/hp/chpdwpegv6lvistek2wqgimtufecqvfp6grp5rpblk5yjicjzqd2.py
# Topologically Sorted Source Nodes: [layer_norm], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# layer_norm => add_2, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_13, [3]), kwargs = {correction: 0, keepdim: True})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
triton_poi_fused_native_layer_norm_0 = async_compile.triton('triton_poi_fused_native_layer_norm_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_native_layer_norm_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + (x0), tmp8, xmask)
tl.store(out_ptr1 + (x0), tmp23, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/el/celxf6tvonm4hnojqioclt5acflsj3tnnwh4e7fon3fxf7kiua46.py
# Topologically Sorted Source Nodes: [tanh, mul, weight_wd_6, tanh_1, mul_1, bias_wd_6, layer_norm], Original ATen: [aten.tanh, aten.mul, aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# bias_wd_6 => add_1
# layer_norm => add_2, add_3, mul_2, mul_3, rsqrt, sub, var_mean
# mul => mul
# mul_1 => mul_1
# tanh => tanh
# tanh_1 => tanh_1
# weight_wd_6 => add
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%squeeze,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_5, %tanh), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_6), kwargs = {})
# %tanh_1 : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%squeeze_1,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_11, %tanh_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_12), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_13, [3]), kwargs = {correction: 0, keepdim: True})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_2,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_13, %getitem_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_2, %add), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_3, %add_1), kwargs = {})
triton_poi_fused_add_mul_native_layer_norm_tanh_1 = async_compile.triton('triton_poi_fused_add_mul_native_layer_norm_tanh_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: '*fp32', 8: '*fp32', 9: '*fp32', 10: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_native_layer_norm_tanh_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 9, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_native_layer_norm_tanh_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, in_ptr6, in_ptr7, in_ptr8, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr6 + (x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr7 + (x0), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr8 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp7 = libdevice.tanh(tmp6)
tmp8 = tmp5 * tmp7
tmp10 = tmp8 + tmp9
tmp11 = tmp4 * tmp10
tmp14 = libdevice.tanh(tmp13)
tmp15 = tmp12 * tmp14
tmp17 = tmp15 + tmp16
tmp18 = tmp11 + tmp17
tl.store(out_ptr0 + (x2), tmp18, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 1), (1, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, 1), (1, 1))
assert_size_stride(primals_11, (4, ), (1, ))
assert_size_stride(primals_12, (4, ), (1, ))
assert_size_stride(primals_13, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), primals_2, out=buf0)
del primals_2
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_3], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf0, (4, 4), (1, 4), 0), primals_3, out=buf1)
buf2 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_4], Original ATen: [aten.mm]
extern_kernels.mm(buf1, primals_4, out=buf2)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_7, (4, 4), (1, 4), 0), primals_8, out=buf3)
del primals_8
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_3], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf3, (4, 4), (1, 4), 0), primals_9, out=buf4)
buf5 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_4], Original ATen: [aten.mm]
extern_kernels.mm(buf4, primals_10, out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [layer_norm], Original ATen: [aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_poi_fused_native_layer_norm_0.run(primals_13, buf6, buf7, 64, grid=grid(64), stream=stream0)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh, mul, weight_wd_6, tanh_1, mul_1, bias_wd_6, layer_norm], Original ATen: [aten.tanh, aten.mul, aten.add, aten.native_layer_norm]
triton_poi_fused_add_mul_native_layer_norm_tanh_1.run(primals_13, buf6, buf7, primals_5, buf2, primals_6, primals_11, buf5, primals_12, buf8, 256, grid=grid(256), stream=stream0)
del buf6
del buf7
del primals_12
del primals_6
return (buf8, primals_3, primals_5, primals_9, primals_11, primals_13, buf2, buf5, reinterpret_tensor(buf4, (4, 4), (1, 4), 0), reinterpret_tensor(primals_10, (1, 4), (1, 1), 0), buf3, primals_7, reinterpret_tensor(buf1, (4, 4), (1, 4), 0), reinterpret_tensor(primals_4, (1, 4), (1, 1), 0), buf0, primals_1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 1), (1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 1), (1, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import numbers
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class WDV52LayerNormalization(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, ts, normalized_shape, wd_normalized_shape,
wd_decoder_layers, wd_require_gradient=False, eps=1e-05,
elementwise_affine=True):
super(WDV52LayerNormalization, self).__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.ts = ts
self.eps = eps
self.elementwise_affine = elementwise_affine
self.wd_decoder_layers = wd_decoder_layers
if self.elementwise_affine:
self.out_wd_weight = Parameter(torch.Tensor(wd_normalized_shape,
*normalized_shape))
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_normalized_shape, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.out_wd_bias = Parameter(torch.Tensor(wd_normalized_shape,
*normalized_shape))
self.ly_wd_bias = Parameter(torch.Tensor(ts, ts))
self.wd_bias = Parameter(torch.Tensor(ts, 1))
self.bias = Parameter(torch.Tensor(wd_normalized_shape, ts),
requires_grad=wd_require_gradient)
self.tanh_bias_weight = Parameter(torch.Tensor(*normalized_shape))
self.tanh_bias_bias = Parameter(torch.Tensor(*normalized_shape))
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
if self.elementwise_affine:
torch.nn.init.xavier_normal_(self.out_wd_weight)
torch.nn.init.xavier_normal_(self.ly_wd_weight)
torch.nn.init.xavier_normal_(self.out_wd_bias)
torch.nn.init.xavier_normal_(self.ly_wd_bias)
torch.nn.init.constant_(self.wd_weight, 1 / self.ts)
torch.nn.init.constant_(self.wd_bias, 1 / self.ts)
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
init.ones_(self.weight)
init.zeros_(self.bias)
def forward(self, input):
weight = self.weight
weight_wd = torch.transpose(weight, 0, 1)
weight_wd = torch.matmul(weight_wd, self.out_wd_weight)
weight_wd = torch.transpose(weight_wd, 0, 1)
weight_wd = torch.matmul(weight_wd, self.ly_wd_weight)
weight_wd = torch.matmul(weight_wd, self.wd_weight)
weight_wd = weight_wd.squeeze(-1)
weight_wd = self.tanh_weight_weight * torch.tanh(weight_wd
) + self.tanh_bias_weight
bias = self.bias
bias_wd = torch.transpose(bias, 0, 1)
bias_wd = torch.matmul(bias_wd, self.out_wd_bias)
bias_wd = torch.transpose(bias_wd, 0, 1)
bias_wd = torch.matmul(bias_wd, self.ly_wd_bias)
bias_wd = torch.matmul(bias_wd, self.wd_bias)
bias_wd = bias_wd.squeeze(-1)
bias_wd = self.tanh_weight_bias * torch.tanh(bias_wd
) + self.tanh_bias_bias
return F.layer_norm(input, self.normalized_shape, weight_wd,
bias_wd, self.eps)
def extra_repr(self):
return (
'{normalized_shape}, eps={eps}, elementwise_affine={elementwise_affine}'
.format(**self.__dict__))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'ts': 4, 'normalized_shape': 4, 'wd_normalized_shape': 4,
'wd_decoder_layers': 1}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import numbers
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + x0, tmp8, xmask)
tl.store(out_ptr1 + x0, tmp23, xmask)
@triton.jit
def triton_poi_fused_add_mul_native_layer_norm_tanh_1(in_ptr0, in_ptr1,
in_ptr2, in_ptr3, in_ptr4, in_ptr5, in_ptr6, in_ptr7, in_ptr8, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr6 + x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr7 + x0, xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr8 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp7 = libdevice.tanh(tmp6)
tmp8 = tmp5 * tmp7
tmp10 = tmp8 + tmp9
tmp11 = tmp4 * tmp10
tmp14 = libdevice.tanh(tmp13)
tmp15 = tmp12 * tmp14
tmp17 = tmp15 + tmp16
tmp18 = tmp11 + tmp17
tl.store(out_ptr0 + x2, tmp18, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 1), (1, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, 1), (1, 1))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (4,), (1,))
assert_size_stride(primals_13, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0),
primals_2, out=buf0)
del primals_2
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (4, 4), (1, 4), 0),
primals_3, out=buf1)
buf2 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.mm(buf1, primals_4, out=buf2)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_7, (4, 4), (1, 4), 0),
primals_8, out=buf3)
del primals_8
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (4, 4), (1, 4), 0),
primals_9, out=buf4)
buf5 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.mm(buf4, primals_10, out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(64)](primals_13, buf6,
buf7, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_mul_native_layer_norm_tanh_1[grid(256)](primals_13
, buf6, buf7, primals_5, buf2, primals_6, primals_11, buf5,
primals_12, buf8, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf6
del buf7
del primals_12
del primals_6
return (buf8, primals_3, primals_5, primals_9, primals_11, primals_13,
buf2, buf5, reinterpret_tensor(buf4, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_10, (1, 4), (1, 1), 0), buf3, primals_7,
reinterpret_tensor(buf1, (4, 4), (1, 4), 0), reinterpret_tensor(
primals_4, (1, 4), (1, 1), 0), buf0, primals_1)
class WDV52LayerNormalizationNew(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, ts, normalized_shape, wd_normalized_shape,
wd_decoder_layers, wd_require_gradient=False, eps=1e-05,
elementwise_affine=True):
super(WDV52LayerNormalizationNew, self).__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.ts = ts
self.eps = eps
self.elementwise_affine = elementwise_affine
self.wd_decoder_layers = wd_decoder_layers
if self.elementwise_affine:
self.out_wd_weight = Parameter(torch.Tensor(wd_normalized_shape,
*normalized_shape))
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_normalized_shape, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.out_wd_bias = Parameter(torch.Tensor(wd_normalized_shape,
*normalized_shape))
self.ly_wd_bias = Parameter(torch.Tensor(ts, ts))
self.wd_bias = Parameter(torch.Tensor(ts, 1))
self.bias = Parameter(torch.Tensor(wd_normalized_shape, ts),
requires_grad=wd_require_gradient)
self.tanh_bias_weight = Parameter(torch.Tensor(*normalized_shape))
self.tanh_bias_bias = Parameter(torch.Tensor(*normalized_shape))
else:
self.register_parameter('weight', None)
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
if self.elementwise_affine:
torch.nn.init.xavier_normal_(self.out_wd_weight)
torch.nn.init.xavier_normal_(self.ly_wd_weight)
torch.nn.init.xavier_normal_(self.out_wd_bias)
torch.nn.init.xavier_normal_(self.ly_wd_bias)
torch.nn.init.constant_(self.wd_weight, 1 / self.ts)
torch.nn.init.constant_(self.wd_bias, 1 / self.ts)
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
init.ones_(self.weight)
init.zeros_(self.bias)
def extra_repr(self):
return (
'{normalized_shape}, eps={eps}, elementwise_affine={elementwise_affine}'
.format(**self.__dict__))
def forward(self, input_0):
primals_1 = self.out_wd_weight
primals_2 = self.ly_wd_weight
primals_4 = self.wd_weight
primals_3 = self.weight
primals_5 = self.tanh_weight_weight
primals_6 = self.tanh_weight_bias
primals_7 = self.out_wd_bias
primals_8 = self.ly_wd_bias
primals_10 = self.wd_bias
primals_9 = self.bias
primals_11 = self.tanh_bias_weight
primals_12 = self.tanh_bias_bias
primals_13 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13])
return output[0]
| Lollipop321/weight-distillation | WDV52LayerNormalization | false | 5,563 | [
"BSD-3-Clause"
] | 1 | cfc76ec58e3e88094dde1825287b2968f9718431 | https://github.com/Lollipop321/weight-distillation/tree/cfc76ec58e3e88094dde1825287b2968f9718431 | import numbers
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class Model(torch.nn.Module):
"""Applies Layer Normalization over a mini-batch of inputs as described in
the paper `Layer Normalization`_ .
.. math::
y = \\frac{x - \\mathrm{E}[x]}{ \\sqrt{\\mathrm{Var}[x] + \\epsilon}} * \\gamma + \\beta
The mean and standard-deviation are calculated separately over the last
certain number dimensions which have to be of the shape specified by
:attr:`normalized_shape`.
:math:`\\gamma` and :math:`\\beta` are learnable affine transform parameters of
:attr:`normalized_shape` if :attr:`elementwise_affine` is ``True``.
.. note::
Unlike Batch Normalization and Instance Normalization, which applies
scalar scale and bias for each entire channel/plane with the
:attr:`affine` option, Layer Normalization applies per-element scale and
bias with :attr:`elementwise_affine`.
This layer uses statistics computed from input data in both training and
evaluation modes.
Args:
normalized_shape (int or list or torch.Size): input shape from an expected input
of size
.. math::
[* \\times \\text{normalized\\_shape}[0] \\times \\text{normalized\\_shape}[1]
\\times \\ldots \\times \\text{normalized\\_shape}[-1]]
If a single integer is used, it is treated as a singleton list, and this module will
normalize over the last dimension which is expected to be of that specific size.
eps: a value added to the denominator for numerical stability. Default: 1e-5
elementwise_affine: a boolean value that when set to ``True``, this module
has learnable per-element affine parameters initialized to ones (for weights)
and zeros (for biases). Default: ``True``.
Shape:
- Input: :math:`(N, *)`
- Output: :math:`(N, *)` (same shape as input)
Examples::
>>> input = torch.randn(20, 5, 10, 10)
>>> # With Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:])
>>> # Without Learnable Parameters
>>> m = nn.LayerNorm(input.size()[1:], elementwise_affine=False)
>>> # Normalize over last two dimensions
>>> m = nn.LayerNorm([10, 10])
>>> # Normalize over last dimension of size 10
>>> m = nn.LayerNorm(10)
>>> # Activating the module
>>> output = m(input)
.. _`Layer Normalization`: https://arxiv.org/abs/1607.06450
"""
__constants__ = ['normalized_shape', 'weight', 'bias', 'eps',
'elementwise_affine']
def __init__(self, ts, normalized_shape, wd_normalized_shape,
wd_decoder_layers, wd_require_gradient=False, eps=1e-05,
elementwise_affine=True):
super().__init__()
if isinstance(normalized_shape, numbers.Integral):
normalized_shape = normalized_shape,
self.normalized_shape = tuple(normalized_shape)
self.ts = ts
self.eps = eps
self.elementwise_affine = elementwise_affine
self.wd_decoder_layers = wd_decoder_layers
if self.elementwise_affine:
self.out_wd_weight = Parameter(torch.Tensor(wd_normalized_shape,
*normalized_shape))
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_normalized_shape, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(*normalized_shape)
)
self.tanh_weight_bias = Parameter(torch.Tensor(*normalized_shape))
self.out_wd_bias = Parameter(torch.Tensor(wd_normalized_shape,
# ... truncated (>4000 chars) for memory efficiency |
UsBlockRes | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/7y/c7y5ocwgrsytsbdmcmeke6s3x47svfjto6zudqpi4u7tnys7gcnh.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.add]
# Source node to ATen node mapping:
# x_1 => add
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%squeeze, %primals_4), kwargs = {})
triton_poi_fused_add_0 = async_compile.triton('triton_poi_fused_add_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 2048
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x1 = (xindex // 512)
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (x2), None)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tl.store(in_out_ptr0 + (x2), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4e/c4eof5cmaf74zvhv7uwyqomvkalvly4gkan7imrv54mxkkvb64lk.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_3 => relu
# Graph fragment:
# %relu : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%squeeze_1,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%squeeze_6, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_relu_threshold_backward_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2048
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x1 = (xindex // 512)
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, None)
tl.store(out_ptr0 + (x2), tmp6, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 2, 2, 2), (32, 8, 4, 2, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 8, 8, 8), (512, 64, 8, 1))
assert_size_stride(primals_5, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_6, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(2, 2, 2), padding=(0, 0, 0), dilation=(1, 1, 1), transposed=True, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 8, 8, 8), (2048, 512, 64, 8, 1))
buf1 = reinterpret_tensor(buf0, (4, 8, 8, 8), (512, 64, 8, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_0.run(buf1, primals_2, primals_4, 2048, grid=grid(2048), stream=stream0)
del primals_2
del primals_4
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (1, 4, 8, 8, 8), (0, 512, 64, 8, 1), 0), primals_5, stride=(1, 1, 1), padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf2, (1, 4, 8, 8, 8), (2048, 512, 64, 8, 1))
buf3 = reinterpret_tensor(buf2, (4, 8, 8, 8), (512, 64, 8, 1), 0); del buf2 # reuse
buf4 = empty_strided_cuda((4, 8, 8, 8), (512, 64, 8, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_1.run(buf3, primals_6, buf4, 2048, grid=grid(2048), stream=stream0)
del primals_6
return (buf3, primals_1, primals_5, reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), reinterpret_tensor(buf1, (1, 4, 8, 8, 8), (2048, 512, 64, 8, 1), 0), buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 2, 2, 2), (32, 8, 4, 2, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 8, 8, 8), (512, 64, 8, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 4, 3, 3, 3), (108, 27, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
def conv1x1(in_channels, out_channels):
return nn.Conv3d(in_channels, out_channels, kernel_size=1)
def upconv2x2(in_channels, out_channels, mode='transpose'):
if mode == 'transpose':
return nn.ConvTranspose3d(in_channels, out_channels, kernel_size=2,
stride=2)
else:
return nn.Sequential(nn.Upsample(mode='trilinear', scale_factor=2),
conv1x1(in_channels, out_channels))
class UsBlockRes(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super(UsBlockRes, self).__init__()
self.upconv = upconv2x2(in_channels, out_channels, mode=up_mode)
self.conv = conv3x3(out_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, before_pool, x):
x = self.upconv(x)
x = x + before_pool
x = self.conv(x)
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 8, 8, 8]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK:
tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x1 = xindex // 512
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + x2, None)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x1 = xindex // 512
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 2, 2, 2), (32, 8, 4, 2, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 8, 8, 8), (512, 64, 8, 1))
assert_size_stride(primals_5, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_6, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1,
4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(2, 2,
2), padding=(0, 0, 0), dilation=(1, 1, 1), transposed=True,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 8, 8, 8), (2048, 512, 64, 8, 1))
buf1 = reinterpret_tensor(buf0, (4, 8, 8, 8), (512, 64, 8, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_add_0[grid(2048)](buf1, primals_2, primals_4, 2048,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
del primals_4
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (1, 4, 8,
8, 8), (0, 512, 64, 8, 1), 0), primals_5, stride=(1, 1, 1),
padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf2, (1, 4, 8, 8, 8), (2048, 512, 64, 8, 1))
buf3 = reinterpret_tensor(buf2, (4, 8, 8, 8), (512, 64, 8, 1), 0)
del buf2
buf4 = empty_strided_cuda((4, 8, 8, 8), (512, 64, 8, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_1[grid(2048)](buf3,
primals_6, buf4, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_6
return buf3, primals_1, primals_5, reinterpret_tensor(primals_3, (1, 4,
4, 4, 4), (256, 64, 16, 4, 1), 0), reinterpret_tensor(buf1, (1, 4,
8, 8, 8), (2048, 512, 64, 8, 1), 0), buf4
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
def conv1x1(in_channels, out_channels):
return nn.Conv3d(in_channels, out_channels, kernel_size=1)
def upconv2x2(in_channels, out_channels, mode='transpose'):
if mode == 'transpose':
return nn.ConvTranspose3d(in_channels, out_channels, kernel_size=2,
stride=2)
else:
return nn.Sequential(nn.Upsample(mode='trilinear', scale_factor=2),
conv1x1(in_channels, out_channels))
class UsBlockResNew(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super(UsBlockResNew, self).__init__()
self.upconv = upconv2x2(in_channels, out_channels, mode=up_mode)
self.conv = conv3x3(out_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, input_0, input_1):
primals_1 = self.upconv.weight
primals_2 = self.upconv.bias
primals_5 = self.conv.weight
primals_6 = self.conv.bias
primals_4 = input_0
primals_3 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| MATHplus-Young-Academy/P2-Cardiac-Motion | UsBlockRes | false | 5,564 | [
"Apache-2.0"
] | 1 | 844995e8e5760f981c425d13c0bd7f2f3bb8baec | https://github.com/MATHplus-Young-Academy/P2-Cardiac-Motion/tree/844995e8e5760f981c425d13c0bd7f2f3bb8baec | import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
def conv1x1(in_channels, out_channels):
return nn.Conv3d(in_channels, out_channels, kernel_size=1)
def upconv2x2(in_channels, out_channels, mode='transpose'):
if mode == 'transpose':
return nn.ConvTranspose3d(in_channels, out_channels, kernel_size=2,
stride=2)
else:
return nn.Sequential(nn.Upsample(mode='trilinear', scale_factor=2),
conv1x1(in_channels, out_channels))
class Model(nn.Module):
def __init__(self, in_channels, out_channels, up_mode='transpose'):
super().__init__()
self.upconv = upconv2x2(in_channels, out_channels, mode=up_mode)
self.conv = conv3x3(out_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
def forward(self, before_pool, x):
x = self.upconv(x)
x = x + before_pool
x = self.conv(x)
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 8, 8, 8]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
ContinousRotReprDecoder | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/hd/chd4rxbi63w5iepye5w4zqxwpml3db4bngunfmfplycvf2qoeigi.py
# Topologically Sorted Source Nodes: [b1, mul, dot_prod], Original ATen: [aten.div, aten.mul, aten.sum]
# Source node to ATen node mapping:
# b1 => div
# dot_prod => sum_2
# mul => mul
# Graph fragment:
# %div : [num_users=5] = call_function[target=torch.ops.aten.div.Tensor](args = (%select, %expand), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %select_1), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1], True), kwargs = {})
triton_poi_fused_div_mul_sum_0 = async_compile.triton('triton_poi_fused_div_mul_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_mul_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_mul_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (6*x0), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (2 + (6*x0)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (4 + (6*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (1 + (6*x0)), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (3 + (6*x0)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr0 + (5 + (6*x0)), xmask, eviction_policy='evict_last')
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = 1e-12
tmp10 = triton_helpers.maximum(tmp8, tmp9)
tmp11 = tmp0 / tmp10
tmp13 = tmp11 * tmp12
tmp14 = tmp2 / tmp10
tmp16 = tmp14 * tmp15
tmp17 = tmp13 + tmp16
tmp18 = tmp5 / tmp10
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tl.store(out_ptr0 + (x0), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/or/corli2hhegxx5f3pgn3genmxh3u3xzhbqybdatbsgwql66rinl3a.py
# Topologically Sorted Source Nodes: [b1, mul, dot_prod, mul_1, sub], Original ATen: [aten.div, aten.mul, aten.sum, aten.sub]
# Source node to ATen node mapping:
# b1 => div
# dot_prod => sum_2
# mul => mul
# mul_1 => mul_1
# sub => sub
# Graph fragment:
# %div : [num_users=5] = call_function[target=torch.ops.aten.div.Tensor](args = (%select, %expand), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %select_1), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1], True), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_2, %div), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%select_2, %mul_1), kwargs = {})
triton_poi_fused_div_mul_sub_sum_1 = async_compile.triton('triton_poi_fused_div_mul_sub_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_mul_sub_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_mul_sub_sum_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 3)
tmp0 = tl.load(in_ptr0 + (1 + (2*x2)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (2*x2), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (6*x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (2 + (6*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (4 + (6*x1)), xmask, eviction_policy='evict_last')
tmp4 = tmp3 * tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = libdevice.sqrt(tmp10)
tmp12 = 1e-12
tmp13 = triton_helpers.maximum(tmp11, tmp12)
tmp14 = tmp2 / tmp13
tmp15 = tmp1 * tmp14
tmp16 = tmp0 - tmp15
tl.store(out_ptr0 + (x2), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/6u/c6ujw2gsz4fiffrddbz6pylpsw3ourbw2ir2dtfko3n5r3gxebkj.py
# Topologically Sorted Source Nodes: [b1, b2, b3], Original ATen: [aten.div, aten.linalg_cross]
# Source node to ATen node mapping:
# b1 => div
# b2 => div_1
# b3 => index, index_1, index_2, index_3, mul_2, mul_3
# Graph fragment:
# %div : [num_users=5] = call_function[target=torch.ops.aten.div.Tensor](args = (%select, %expand), kwargs = {})
# %div_1 : [num_users=3] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, %expand_1), kwargs = {})
# %index : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%div, [None, %remainder]), kwargs = {})
# %index_1 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%div_1, [None, %remainder_1]), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index, %index_1), kwargs = {})
# %index_2 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%div, [None, %remainder_2]), kwargs = {})
# %index_3 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%div_1, [None, %remainder_3]), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index_2, %index_3), kwargs = {})
triton_poi_fused_div_linalg_cross_2 = async_compile.triton('triton_poi_fused_div_linalg_cross_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_linalg_cross_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 10, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_linalg_cross_2(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 3
x1 = (xindex // 3)
x2 = xindex
tmp0 = tl.load(in_ptr0 + ((2*((1 + x0) % 3)) + (6*x1)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (6*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + (6*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (4 + (6*x1)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + ((3*x1) + ((2 + x0) % 3)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (3*x1), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (1 + (3*x1)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (2 + (3*x1)), xmask, eviction_policy='evict_last')
tmp26 = tl.load(in_ptr0 + ((2*((2 + x0) % 3)) + (6*x1)), xmask, eviction_policy='evict_last')
tmp28 = tl.load(in_ptr1 + ((3*x1) + ((1 + x0) % 3)), xmask)
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp9 = libdevice.sqrt(tmp8)
tmp10 = 1e-12
tmp11 = triton_helpers.maximum(tmp9, tmp10)
tmp12 = tmp0 / tmp11
tmp15 = tmp14 * tmp14
tmp17 = tmp16 * tmp16
tmp18 = tmp15 + tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = libdevice.sqrt(tmp21)
tmp23 = triton_helpers.maximum(tmp22, tmp10)
tmp24 = tmp13 / tmp23
tmp25 = tmp12 * tmp24
tmp27 = tmp26 / tmp11
tmp29 = tmp28 / tmp23
tmp30 = tmp27 * tmp29
tl.store(out_ptr0 + (x2), tmp25, xmask)
tl.store(out_ptr1 + (x2), tmp30, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/6w/c6woiqqq2vb4nax2oe4lql3ogk6yfdkg35mdxnytclkct4emsqlb.py
# Topologically Sorted Source Nodes: [stack], Original ATen: [aten.stack]
# Source node to ATen node mapping:
# stack => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%unsqueeze, %unsqueeze_1, %unsqueeze_2], -1), kwargs = {})
triton_poi_fused_stack_3 = async_compile.triton('triton_poi_fused_stack_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_stack_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 10, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_stack_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 36
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 3
x3 = (xindex // 3)
x2 = (xindex // 9)
x5 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2*x3), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (6*x2), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp7 = tmp6 * tmp6
tmp8 = tl.load(in_ptr0 + (2 + (6*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = tl.load(in_ptr0 + (4 + (6*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = libdevice.sqrt(tmp13)
tmp15 = 1e-12
tmp16 = triton_helpers.maximum(tmp14, tmp15)
tmp17 = tmp5 / tmp16
tmp18 = tl.full(tmp17.shape, 0.0, tmp17.dtype)
tmp19 = tl.where(tmp4, tmp17, tmp18)
tmp20 = tmp0 >= tmp3
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp0 < tmp21
tmp23 = tmp20 & tmp22
tmp24 = tl.load(in_ptr1 + (x3), tmp23 & xmask, eviction_policy='evict_last', other=0.0)
tmp25 = tl.load(in_ptr1 + (3*x2), tmp23 & xmask, eviction_policy='evict_last', other=0.0)
tmp26 = tmp25 * tmp25
tmp27 = tl.load(in_ptr1 + (1 + (3*x2)), tmp23 & xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tmp27 * tmp27
tmp29 = tmp26 + tmp28
tmp30 = tl.load(in_ptr1 + (2 + (3*x2)), tmp23 & xmask, eviction_policy='evict_last', other=0.0)
tmp31 = tmp30 * tmp30
tmp32 = tmp29 + tmp31
tmp33 = libdevice.sqrt(tmp32)
tmp34 = triton_helpers.maximum(tmp33, tmp15)
tmp35 = tmp24 / tmp34
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp23, tmp35, tmp36)
tmp38 = tmp0 >= tmp21
tmp39 = tl.full([1], 3, tl.int64)
tmp40 = tmp0 < tmp39
tmp41 = tl.load(in_ptr2 + (x3), tmp38 & xmask, eviction_policy='evict_last', other=0.0)
tmp42 = tl.load(in_ptr3 + (x3), tmp38 & xmask, eviction_policy='evict_last', other=0.0)
tmp43 = tmp41 - tmp42
tmp44 = tl.full(tmp43.shape, 0.0, tmp43.dtype)
tmp45 = tl.where(tmp38, tmp43, tmp44)
tmp46 = tl.where(tmp23, tmp37, tmp45)
tmp47 = tl.where(tmp4, tmp19, tmp46)
tl.store(out_ptr0 + (x5), tmp47, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 3, 2), (6, 2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
# Topologically Sorted Source Nodes: [b1, mul, dot_prod], Original ATen: [aten.div, aten.mul, aten.sum]
stream0 = get_raw_stream(0)
triton_poi_fused_div_mul_sum_0.run(arg0_1, buf0, 4, grid=grid(4), stream=stream0)
buf1 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
# Topologically Sorted Source Nodes: [b1, mul, dot_prod, mul_1, sub], Original ATen: [aten.div, aten.mul, aten.sum, aten.sub]
triton_poi_fused_div_mul_sub_sum_1.run(arg0_1, buf0, buf1, 12, grid=grid(12), stream=stream0)
del buf0
buf2 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
buf3 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
# Topologically Sorted Source Nodes: [b1, b2, b3], Original ATen: [aten.div, aten.linalg_cross]
triton_poi_fused_div_linalg_cross_2.run(arg0_1, buf1, buf2, buf3, 12, grid=grid(12), stream=stream0)
buf4 = empty_strided_cuda((4, 3, 3), (9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [stack], Original ATen: [aten.stack]
triton_poi_fused_stack_3.run(arg0_1, buf1, buf2, buf3, buf4, 36, grid=grid(36), stream=stream0)
del arg0_1
del buf1
del buf2
del buf3
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 3, 2), (6, 2, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class ContinousRotReprDecoder(nn.Module):
def __init__(self):
super(ContinousRotReprDecoder, self).__init__()
def forward(self, module_input):
reshaped_input = module_input.view(-1, 3, 2)
b1 = F.normalize(reshaped_input[:, :, 0], dim=1)
dot_prod = torch.sum(b1 * reshaped_input[:, :, 1], dim=1, keepdim=True)
b2 = F.normalize(reshaped_input[:, :, 1] - dot_prod * b1, dim=-1)
b3 = torch.cross(b1, b2, dim=1)
return torch.stack([b1, b2, b3], dim=-1)
def get_inputs():
return [torch.rand([4, 3, 2])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_div_mul_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 6 * x0, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (2 + 6 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (4 + 6 * x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (1 + 6 * x0), xmask, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr0 + (3 + 6 * x0), xmask, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr0 + (5 + 6 * x0), xmask, eviction_policy='evict_last'
)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = 1e-12
tmp10 = triton_helpers.maximum(tmp8, tmp9)
tmp11 = tmp0 / tmp10
tmp13 = tmp11 * tmp12
tmp14 = tmp2 / tmp10
tmp16 = tmp14 * tmp15
tmp17 = tmp13 + tmp16
tmp18 = tmp5 / tmp10
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tl.store(out_ptr0 + x0, tmp21, xmask)
@triton.jit
def triton_poi_fused_div_mul_sub_sum_1(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 3
tmp0 = tl.load(in_ptr0 + (1 + 2 * x2), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + 2 * x2, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + 6 * x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (2 + 6 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (4 + 6 * x1), xmask, eviction_policy='evict_last')
tmp4 = tmp3 * tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = libdevice.sqrt(tmp10)
tmp12 = 1e-12
tmp13 = triton_helpers.maximum(tmp11, tmp12)
tmp14 = tmp2 / tmp13
tmp15 = tmp1 * tmp14
tmp16 = tmp0 - tmp15
tl.store(out_ptr0 + x2, tmp16, xmask)
@triton.jit
def triton_poi_fused_div_linalg_cross_2(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 3
x1 = xindex // 3
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 * ((1 + x0) % 3) + 6 * x1), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + 6 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 6 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (4 + 6 * x1), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (3 * x1 + (2 + x0) % 3), xmask,
eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + 3 * x1, xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (1 + 3 * x1), xmask, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr1 + (2 + 3 * x1), xmask, eviction_policy='evict_last'
)
tmp26 = tl.load(in_ptr0 + (2 * ((2 + x0) % 3) + 6 * x1), xmask,
eviction_policy='evict_last')
tmp28 = tl.load(in_ptr1 + (3 * x1 + (1 + x0) % 3), xmask)
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp9 = libdevice.sqrt(tmp8)
tmp10 = 1e-12
tmp11 = triton_helpers.maximum(tmp9, tmp10)
tmp12 = tmp0 / tmp11
tmp15 = tmp14 * tmp14
tmp17 = tmp16 * tmp16
tmp18 = tmp15 + tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = libdevice.sqrt(tmp21)
tmp23 = triton_helpers.maximum(tmp22, tmp10)
tmp24 = tmp13 / tmp23
tmp25 = tmp12 * tmp24
tmp27 = tmp26 / tmp11
tmp29 = tmp28 / tmp23
tmp30 = tmp27 * tmp29
tl.store(out_ptr0 + x2, tmp25, xmask)
tl.store(out_ptr1 + x2, tmp30, xmask)
@triton.jit
def triton_poi_fused_stack_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 36
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 3
x3 = xindex // 3
x2 = xindex // 9
x5 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + 2 * x3, tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + 6 * x2, tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp7 = tmp6 * tmp6
tmp8 = tl.load(in_ptr0 + (2 + 6 * x2), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = tl.load(in_ptr0 + (4 + 6 * x2), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = libdevice.sqrt(tmp13)
tmp15 = 1e-12
tmp16 = triton_helpers.maximum(tmp14, tmp15)
tmp17 = tmp5 / tmp16
tmp18 = tl.full(tmp17.shape, 0.0, tmp17.dtype)
tmp19 = tl.where(tmp4, tmp17, tmp18)
tmp20 = tmp0 >= tmp3
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp0 < tmp21
tmp23 = tmp20 & tmp22
tmp24 = tl.load(in_ptr1 + x3, tmp23 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp25 = tl.load(in_ptr1 + 3 * x2, tmp23 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp26 = tmp25 * tmp25
tmp27 = tl.load(in_ptr1 + (1 + 3 * x2), tmp23 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp28 = tmp27 * tmp27
tmp29 = tmp26 + tmp28
tmp30 = tl.load(in_ptr1 + (2 + 3 * x2), tmp23 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp31 = tmp30 * tmp30
tmp32 = tmp29 + tmp31
tmp33 = libdevice.sqrt(tmp32)
tmp34 = triton_helpers.maximum(tmp33, tmp15)
tmp35 = tmp24 / tmp34
tmp36 = tl.full(tmp35.shape, 0.0, tmp35.dtype)
tmp37 = tl.where(tmp23, tmp35, tmp36)
tmp38 = tmp0 >= tmp21
tl.full([1], 3, tl.int64)
tmp41 = tl.load(in_ptr2 + x3, tmp38 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp42 = tl.load(in_ptr3 + x3, tmp38 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp43 = tmp41 - tmp42
tmp44 = tl.full(tmp43.shape, 0.0, tmp43.dtype)
tmp45 = tl.where(tmp38, tmp43, tmp44)
tmp46 = tl.where(tmp23, tmp37, tmp45)
tmp47 = tl.where(tmp4, tmp19, tmp46)
tl.store(out_ptr0 + x5, tmp47, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 3, 2), (6, 2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_mul_sum_0[grid(4)](arg0_1, buf0, 4, XBLOCK=4,
num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
triton_poi_fused_div_mul_sub_sum_1[grid(12)](arg0_1, buf0, buf1, 12,
XBLOCK=16, num_warps=1, num_stages=1)
del buf0
buf2 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
buf3 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
triton_poi_fused_div_linalg_cross_2[grid(12)](arg0_1, buf1, buf2,
buf3, 12, XBLOCK=16, num_warps=1, num_stages=1)
buf4 = empty_strided_cuda((4, 3, 3), (9, 3, 1), torch.float32)
triton_poi_fused_stack_3[grid(36)](arg0_1, buf1, buf2, buf3, buf4,
36, XBLOCK=64, num_warps=1, num_stages=1)
del arg0_1
del buf1
del buf2
del buf3
return buf4,
class ContinousRotReprDecoderNew(nn.Module):
def __init__(self):
super(ContinousRotReprDecoderNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| LuckyDC/human_body_prior | ContinousRotReprDecoder | false | 5,565 | [
"Xnet",
"X11"
] | 1 | 6a46613b4cbd9c62d888359f1435cec501643af3 | https://github.com/LuckyDC/human_body_prior/tree/6a46613b4cbd9c62d888359f1435cec501643af3 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, module_input):
reshaped_input = module_input.view(-1, 3, 2)
b1 = F.normalize(reshaped_input[:, :, 0], dim=1)
dot_prod = torch.sum(b1 * reshaped_input[:, :, 1], dim=1, keepdim=True)
b2 = F.normalize(reshaped_input[:, :, 1] - dot_prod * b1, dim=-1)
b3 = torch.cross(b1, b2, dim=1)
return torch.stack([b1, b2, b3], dim=-1)
def get_inputs():
return [torch.rand([4, 3, 2])]
def get_init_inputs():
return []
|
BCEWithLogitsLoss2d | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/5u/c5u7c3xfjaazdtxwasop4c2d3vlwaagqtikdsnpoz56tcxjbvfvx.py
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits], Original ATen: [aten.binary_cross_entropy_with_logits]
# Source node to ATen node mapping:
# binary_cross_entropy_with_logits => abs_1, exp, full_default, log1p, mean, minimum, mul, neg, sub, sub_1, sub_2
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %view_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %view), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %minimum : [num_users=1] = call_function[target=torch.ops.aten.minimum.default](args = (%full_default, %view), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%view,), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%abs_1,), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%neg,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%minimum, %log1p), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul, %sub_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sub_2,), kwargs = {})
triton_per_fused_binary_cross_entropy_with_logits_0 = async_compile.triton('triton_per_fused_binary_cross_entropy_with_logits_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_binary_cross_entropy_with_logits_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_binary_cross_entropy_with_logits_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp3 = tl.load(in_ptr1 + (r0), None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 256.0
tmp17 = tmp15 / tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp17, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits], Original ATen: [aten.binary_cross_entropy_with_logits]
stream0 = get_raw_stream(0)
triton_per_fused_binary_cross_entropy_with_logits_0.run(buf1, arg1_1, arg0_1, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import numpy as np
import torch.nn as nn
class BCEWithLogitsLoss2d(nn.Module):
"""Computationally stable version of 2D BCE loss
"""
def __init__(self, weight=None, reduction='elementwise_mean'):
super(BCEWithLogitsLoss2d, self).__init__()
if isinstance(weight, np.ndarray):
weight = torch.from_numpy(weight)
self.bce_loss = nn.BCEWithLogitsLoss(weight, reduction)
def forward(self, logits, targets):
logits_flat = logits.view(-1)
targets_flat = targets.view(-1)
return self.bce_loss(logits_flat, targets_flat)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_binary_cross_entropy_with_logits_0(in_out_ptr0,
in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp3 = tl.load(in_ptr1 + r0, None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 256.0
tmp17 = tmp15 / tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp17, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_binary_cross_entropy_with_logits_0[grid(1)](buf1,
arg1_1, arg0_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
class BCEWithLogitsLoss2dNew(nn.Module):
"""Computationally stable version of 2D BCE loss
"""
def __init__(self, weight=None, reduction='elementwise_mean'):
super(BCEWithLogitsLoss2dNew, self).__init__()
if isinstance(weight, np.ndarray):
weight = torch.from_numpy(weight)
self.bce_loss = nn.BCEWithLogitsLoss(weight, reduction)
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MIPT-Oulu/3D-Histo-Grading | BCEWithLogitsLoss2d | false | 5,566 | [
"MIT"
] | 1 | b779a154d0e5b104fc152c8952124768fb7b1dc6 | https://github.com/MIPT-Oulu/3D-Histo-Grading/tree/b779a154d0e5b104fc152c8952124768fb7b1dc6 | import torch
import numpy as np
import torch.nn as nn
class Model(nn.Module):
"""Computationally stable version of 2D BCE loss
"""
def __init__(self, weight=None, reduction='elementwise_mean'):
super().__init__()
if isinstance(weight, np.ndarray):
weight = torch.from_numpy(weight)
self.bce_loss = nn.BCEWithLogitsLoss(weight, reduction)
def forward(self, logits, targets):
logits_flat = logits.view(-1)
targets_flat = targets.view(-1)
return self.bce_loss(logits_flat, targets_flat)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
ViTStemPatchify | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/sr/csrhhqsexdcor6gq6tz4dawxblhadgekinzxxkt33uwojltligp6.py
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# conv2d => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [4, 4], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_0 = async_compile.triton('triton_poi_fused_convolution_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(4, 4), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_0.run(buf1, primals_2, 16, grid=grid(16), stream=stream0)
del primals_2
return (buf1, primals_1, primals_3, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from torch.nn import Module
import torch
import torch.nn as nn
import torch.utils.data
def patchify2d(w_in, w_out, k, *, bias=True):
"""Helper for building a patchify layer as used by ViT models."""
return nn.Conv2d(w_in, w_out, k, stride=k, padding=0, bias=bias)
def patchify2d_cx(cx, w_in, w_out, k, *, bias=True):
"""Accumulates complexity of patchify2d into cx = (h, w, flops, params, acts)."""
err_str = 'Only kernel sizes divisible by the input size are supported.'
assert cx['h'] % k == 0 and cx['w'] % k == 0, err_str
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
h, w = h // k, w // k
flops += k * k * w_in * w_out * h * w + (w_out * h * w if bias else 0)
params += k * k * w_in * w_out + (w_out if bias else 0)
acts += w_out * h * w
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class ViTStemPatchify(Module):
"""The patchify vision transformer stem as per https://arxiv.org/abs/2010.11929."""
def __init__(self, w_in, w_out, k):
super(ViTStemPatchify, self).__init__()
self.patchify = patchify2d(w_in, w_out, k, bias=True)
def forward(self, x):
return self.patchify(x)
@staticmethod
def complexity(cx, w_in, w_out, k):
return patchify2d_cx(cx, w_in, w_out, k, bias=True)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'w_in': 4, 'w_out': 4, 'k': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch.nn import Module
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(4,
4), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(16)](buf1, primals_2, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_2
return buf1, primals_1, primals_3
def patchify2d(w_in, w_out, k, *, bias=True):
"""Helper for building a patchify layer as used by ViT models."""
return nn.Conv2d(w_in, w_out, k, stride=k, padding=0, bias=bias)
def patchify2d_cx(cx, w_in, w_out, k, *, bias=True):
"""Accumulates complexity of patchify2d into cx = (h, w, flops, params, acts)."""
err_str = 'Only kernel sizes divisible by the input size are supported.'
assert cx['h'] % k == 0 and cx['w'] % k == 0, err_str
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
h, w = h // k, w // k
flops += k * k * w_in * w_out * h * w + (w_out * h * w if bias else 0)
params += k * k * w_in * w_out + (w_out if bias else 0)
acts += w_out * h * w
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class ViTStemPatchifyNew(Module):
"""The patchify vision transformer stem as per https://arxiv.org/abs/2010.11929."""
def __init__(self, w_in, w_out, k):
super(ViTStemPatchifyNew, self).__init__()
self.patchify = patchify2d(w_in, w_out, k, bias=True)
@staticmethod
def complexity(cx, w_in, w_out, k):
return patchify2d_cx(cx, w_in, w_out, k, bias=True)
def forward(self, input_0):
primals_1 = self.patchify.weight
primals_2 = self.patchify.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| MAC-AutoML/XCompression | ViTStemPatchify | false | 5,567 | [
"MIT"
] | 1 | 9f76eb3ccfb3057110ecf12aa48dec00a4667a25 | https://github.com/MAC-AutoML/XCompression/tree/9f76eb3ccfb3057110ecf12aa48dec00a4667a25 | from torch.nn import Module
import torch
import torch.nn as nn
import torch.utils.data
def patchify2d(w_in, w_out, k, *, bias=True):
"""Helper for building a patchify layer as used by ViT models."""
return nn.Conv2d(w_in, w_out, k, stride=k, padding=0, bias=bias)
def patchify2d_cx(cx, w_in, w_out, k, *, bias=True):
"""Accumulates complexity of patchify2d into cx = (h, w, flops, params, acts)."""
err_str = 'Only kernel sizes divisible by the input size are supported.'
assert cx['h'] % k == 0 and cx['w'] % k == 0, err_str
h, w, flops, params, acts = cx['h'], cx['w'], cx['flops'], cx['params'
], cx['acts']
h, w = h // k, w // k
flops += k * k * w_in * w_out * h * w + (w_out * h * w if bias else 0)
params += k * k * w_in * w_out + (w_out if bias else 0)
acts += w_out * h * w
return {'h': h, 'w': w, 'flops': flops, 'params': params, 'acts': acts}
class Model(Module):
"""The patchify vision transformer stem as per https://arxiv.org/abs/2010.11929."""
def __init__(self, w_in, w_out, k):
super().__init__()
self.patchify = patchify2d(w_in, w_out, k, bias=True)
def forward(self, x):
return self.patchify(x)
@staticmethod
def complexity(cx, w_in, w_out, k):
return patchify2d_cx(cx, w_in, w_out, k, bias=True)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
WDV52Linear | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/oo/coo46bs5zpai6ydehcambdwmxekrecdbbhcbxivotaebjhzp7llm.py
# Topologically Sorted Source Nodes: [weight_wd_1], Original ATen: [aten.clone, aten._unsafe_view]
# Source node to ATen node mapping:
# weight_wd_1 => clone, view
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute,), kwargs = {memory_format: torch.contiguous_format})
# %view : [num_users=2] = call_function[target=torch.ops.aten.reshape.default](args = (%clone, [16, 4]), kwargs = {})
triton_poi_fused__unsafe_view_clone_0 = async_compile.triton('triton_poi_fused__unsafe_view_clone_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__unsafe_view_clone_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__unsafe_view_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4)
x2 = xindex
tmp0 = tl.load(in_ptr0 + ((4*(x1 % 4)) + (16*x0) + (x1 // 4)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x2), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kk/ckkn7vlnz5gfvzigo3gxfhxwb6i7tj47p7plxoih22qwcvc7em26.py
# Topologically Sorted Source Nodes: [weight_wd_5], Original ATen: [aten.clone, aten._unsafe_view]
# Source node to ATen node mapping:
# weight_wd_5 => clone_2, view_4
# Graph fragment:
# %clone_2 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_2,), kwargs = {memory_format: torch.contiguous_format})
# %view_4 : [num_users=2] = call_function[target=torch.ops.aten.reshape.default](args = (%clone_2, [16, 4]), kwargs = {})
triton_poi_fused__unsafe_view_clone_1 = async_compile.triton('triton_poi_fused__unsafe_view_clone_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__unsafe_view_clone_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__unsafe_view_clone_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + ((4*x1) + (16*(y0 // 4)) + (y0 % 4)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x1 + (4*y0)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4p/c4p5vybvk5kczshqzudonrympy7hqc7dyeur4eomof353wgjgkm6.py
# Topologically Sorted Source Nodes: [tanh, mul, weight_wd_9], Original ATen: [aten.tanh, aten.mul, aten.add]
# Source node to ATen node mapping:
# mul => mul
# tanh => tanh
# weight_wd_9 => add
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%squeeze,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_6, %tanh), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_7), kwargs = {})
triton_poi_fused_add_mul_tanh_2 = async_compile.triton('triton_poi_fused_add_mul_tanh_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_tanh_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_tanh_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp4 = tl.load(in_ptr2 + (x0), xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + (x0), tmp5, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/er/certponyoudrebqs7ovkmzybyrkb26c4kqh3oaysaxf4ryytitj2.py
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias_wd_6], Original ATen: [aten.tanh, aten.mul, aten.add]
# Source node to ATen node mapping:
# bias_wd_6 => add_1
# mul_1 => mul_1
# tanh_1 => tanh_1
# Graph fragment:
# %tanh_1 : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%squeeze_1,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_12, %tanh_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_13), kwargs = {})
triton_poi_fused_add_mul_tanh_3 = async_compile.triton('triton_poi_fused_add_mul_tanh_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_tanh_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_tanh_3(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp4 = tl.load(in_ptr2 + (x0), xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + (x0), tmp5, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, 1), (1, 1))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4, 1), (1, 1))
assert_size_stride(primals_12, (4, ), (1, ))
assert_size_stride(primals_13, (4, ), (1, ))
assert_size_stride(primals_14, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_1], Original ATen: [aten.clone, aten._unsafe_view]
stream0 = get_raw_stream(0)
triton_poi_fused__unsafe_view_clone_0.run(primals_1, buf0, 64, grid=grid(64), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_1], Original ATen: [aten.mm]
extern_kernels.mm(buf0, primals_2, out=buf1)
del primals_2
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_3], Original ATen: [aten.clone, aten._unsafe_view]
triton_poi_fused__unsafe_view_clone_0.run(buf1, buf2, 64, grid=grid(64), stream=stream0)
buf3 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [weight_wd_3], Original ATen: [aten.mm]
extern_kernels.mm(buf2, primals_3, out=buf3)
buf4 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_5], Original ATen: [aten.clone, aten._unsafe_view]
triton_poi_fused__unsafe_view_clone_1.run(buf3, buf4, 16, 4, grid=grid(16, 4), stream=stream0)
buf5 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [weight_wd_5], Original ATen: [aten.mm]
extern_kernels.mm(buf4, primals_4, out=buf5)
buf6 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_7], Original ATen: [aten.clone, aten._unsafe_view]
triton_poi_fused__unsafe_view_clone_1.run(buf5, buf6, 16, 4, grid=grid(16, 4), stream=stream0)
del buf5
buf7 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight_wd_7], Original ATen: [aten.mm]
extern_kernels.mm(buf6, primals_5, out=buf7)
buf8 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), primals_9, out=buf8)
del primals_9
buf9 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_3], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf8, (4, 4), (1, 4), 0), primals_10, out=buf9)
buf10 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [bias_wd_4], Original ATen: [aten.mm]
extern_kernels.mm(buf9, primals_11, out=buf10)
buf11 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh, mul, weight_wd_9], Original ATen: [aten.tanh, aten.mul, aten.add]
triton_poi_fused_add_mul_tanh_2.run(primals_6, buf7, primals_7, buf11, 16, grid=grid(16), stream=stream0)
del primals_7
buf12 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias_wd_6], Original ATen: [aten.tanh, aten.mul, aten.add]
triton_poi_fused_add_mul_tanh_3.run(primals_12, buf10, primals_13, buf12, 4, grid=grid(4), stream=stream0)
del primals_13
buf13 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [tanh_1, mul_1, bias_wd_6, linear], Original ATen: [aten.tanh, aten.mul, aten.add, aten.addmm]
extern_kernels.addmm(buf12, reinterpret_tensor(primals_14, (64, 4), (4, 1), 0), reinterpret_tensor(buf11, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf13)
del buf11
del buf12
return (reinterpret_tensor(buf13, (4, 4, 4, 4), (64, 16, 4, 1), 0), primals_6, primals_10, primals_12, buf7, buf10, reinterpret_tensor(primals_14, (64, 4), (4, 1), 0), reinterpret_tensor(buf9, (4, 4), (1, 4), 0), reinterpret_tensor(primals_11, (1, 4), (1, 1), 0), buf8, primals_8, reinterpret_tensor(buf6, (4, 16), (1, 4), 0), reinterpret_tensor(primals_5, (1, 4), (1, 1), 0), reinterpret_tensor(buf4, (4, 16), (1, 4), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), reinterpret_tensor(buf2, (4, 16), (1, 4), 0), reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), reinterpret_tensor(buf0, (4, 16), (1, 4), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 1), (1, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, 1), (1, 1), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class WDV52Linear(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, ts, in_features, out_features, wd_in_features,
wd_out_features, wd_decoder_layers, wd_require_gradient=False, bias
=True):
super(WDV52Linear, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.ts = ts
self.wd_decoder_layers = wd_decoder_layers
self.out_wd_weight = Parameter(torch.Tensor(wd_out_features,
out_features))
self.in_wd_weight = Parameter(torch.Tensor(wd_in_features, in_features)
)
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_out_features,
wd_in_features, ts), requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.out_wd_bias = Parameter(torch.Tensor(wd_out_features,
out_features))
self.ly_wd_bias = Parameter(torch.Tensor(ts, ts))
self.wd_bias = Parameter(torch.Tensor(ts, 1))
self.bias = Parameter(torch.Tensor(wd_out_features, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.xavier_normal_(self.out_wd_weight)
torch.nn.init.xavier_normal_(self.in_wd_weight)
torch.nn.init.xavier_normal_(self.ly_wd_weight)
torch.nn.init.constant_(self.wd_weight, 1 / self.ts)
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.xavier_normal_(self.out_wd_bias)
torch.nn.init.xavier_normal_(self.ly_wd_bias)
torch.nn.init.constant_(self.wd_bias, 1 / self.ts)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight)
bound = 1 / math.sqrt(fan_in)
init.uniform_(self.bias, -bound, bound)
def forward(self, input):
weight = self.weight
weight_wd = torch.transpose(weight, 0, 2)
weight_wd = torch.matmul(weight_wd, self.out_wd_weight)
weight_wd = torch.transpose(weight_wd, 0, 2)
weight_wd = torch.matmul(weight_wd, self.ly_wd_weight)
weight_wd = torch.transpose(weight_wd, 1, 2)
weight_wd = torch.matmul(weight_wd, self.in_wd_weight)
weight_wd = torch.transpose(weight_wd, 1, 2)
weight_wd = torch.matmul(weight_wd, self.wd_weight)
weight_wd = weight_wd.squeeze(-1)
weight_wd = self.tanh_weight_weight * torch.tanh(weight_wd
) + self.tanh_bias_weight
bias = self.bias
bias_wd = torch.transpose(bias, 0, 1)
bias_wd = torch.matmul(bias_wd, self.out_wd_bias)
bias_wd = torch.transpose(bias_wd, 0, 1)
bias_wd = torch.matmul(bias_wd, self.ly_wd_bias)
bias_wd = torch.matmul(bias_wd, self.wd_bias)
bias_wd = bias_wd.squeeze(-1)
bias_wd = self.tanh_weight_bias * torch.tanh(bias_wd
) + self.tanh_bias_bias
return F.linear(input, weight_wd, bias_wd)
def extra_repr(self):
return 'in_features={}, out_features={}, bias={}'.format(self.
in_features, self.out_features, self.bias is not None)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'ts': 4, 'in_features': 4, 'out_features': 4,
'wd_in_features': 4, 'wd_out_features': 4, 'wd_decoder_layers': 1}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import math
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__unsafe_view_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (4 * (x1 % 4) + 16 * x0 + x1 // 4), xmask,
eviction_policy='evict_last')
tl.store(out_ptr0 + x2, tmp0, xmask)
@triton.jit
def triton_poi_fused__unsafe_view_clone_1(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + (4 * x1 + 16 * (y0 // 4) + y0 % 4), xmask &
ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x1 + 4 * y0), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_mul_tanh_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp4 = tl.load(in_ptr2 + x0, xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + x0, tmp5, xmask)
@triton.jit
def triton_poi_fused_add_mul_tanh_3(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp4 = tl.load(in_ptr2 + x0, xmask)
tmp2 = libdevice.tanh(tmp1)
tmp3 = tmp0 * tmp2
tmp5 = tmp3 + tmp4
tl.store(out_ptr0 + x0, tmp5, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, 1), (1, 1))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4, 1), (1, 1))
assert_size_stride(primals_12, (4,), (1,))
assert_size_stride(primals_13, (4,), (1,))
assert_size_stride(primals_14, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__unsafe_view_clone_0[grid(64)](primals_1, buf0, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(buf0, primals_2, out=buf1)
del primals_2
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
triton_poi_fused__unsafe_view_clone_0[grid(64)](buf1, buf2, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf3 = buf1
del buf1
extern_kernels.mm(buf2, primals_3, out=buf3)
buf4 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
triton_poi_fused__unsafe_view_clone_1[grid(16, 4)](buf3, buf4, 16,
4, XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
buf5 = buf3
del buf3
extern_kernels.mm(buf4, primals_4, out=buf5)
buf6 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
triton_poi_fused__unsafe_view_clone_1[grid(16, 4)](buf5, buf6, 16,
4, XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
del buf5
buf7 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
extern_kernels.mm(buf6, primals_5, out=buf7)
buf8 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_8, (4, 4), (1, 4), 0),
primals_9, out=buf8)
del primals_9
buf9 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf8, (4, 4), (1, 4), 0),
primals_10, out=buf9)
buf10 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.mm(buf9, primals_11, out=buf10)
buf11 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_add_mul_tanh_2[grid(16)](primals_6, buf7,
primals_7, buf11, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf12 = empty_strided_cuda((4,), (1,), torch.float32)
triton_poi_fused_add_mul_tanh_3[grid(4)](primals_12, buf10,
primals_13, buf12, 4, XBLOCK=4, num_warps=1, num_stages=1)
del primals_13
buf13 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(buf12, reinterpret_tensor(primals_14, (64, 4),
(4, 1), 0), reinterpret_tensor(buf11, (4, 4), (1, 4), 0), alpha
=1, beta=1, out=buf13)
del buf11
del buf12
return reinterpret_tensor(buf13, (4, 4, 4, 4), (64, 16, 4, 1), 0
), primals_6, primals_10, primals_12, buf7, buf10, reinterpret_tensor(
primals_14, (64, 4), (4, 1), 0), reinterpret_tensor(buf9, (4, 4), (
1, 4), 0), reinterpret_tensor(primals_11, (1, 4), (1, 1), 0
), buf8, primals_8, reinterpret_tensor(buf6, (4, 16), (1, 4), 0
), reinterpret_tensor(primals_5, (1, 4), (1, 1), 0
), reinterpret_tensor(buf4, (4, 16), (1, 4), 0), reinterpret_tensor(
primals_4, (4, 4), (1, 4), 0), reinterpret_tensor(buf2, (4, 16), (1,
4), 0), reinterpret_tensor(primals_3, (4, 4), (1, 4), 0
), reinterpret_tensor(buf0, (4, 16), (1, 4), 0)
class WDV52LinearNew(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, ts, in_features, out_features, wd_in_features,
wd_out_features, wd_decoder_layers, wd_require_gradient=False, bias
=True):
super(WDV52LinearNew, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.ts = ts
self.wd_decoder_layers = wd_decoder_layers
self.out_wd_weight = Parameter(torch.Tensor(wd_out_features,
out_features))
self.in_wd_weight = Parameter(torch.Tensor(wd_in_features, in_features)
)
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_out_features,
wd_in_features, ts), requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.out_wd_bias = Parameter(torch.Tensor(wd_out_features,
out_features))
self.ly_wd_bias = Parameter(torch.Tensor(ts, ts))
self.wd_bias = Parameter(torch.Tensor(ts, 1))
self.bias = Parameter(torch.Tensor(wd_out_features, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.xavier_normal_(self.out_wd_weight)
torch.nn.init.xavier_normal_(self.in_wd_weight)
torch.nn.init.xavier_normal_(self.ly_wd_weight)
torch.nn.init.constant_(self.wd_weight, 1 / self.ts)
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.xavier_normal_(self.out_wd_bias)
torch.nn.init.xavier_normal_(self.ly_wd_bias)
torch.nn.init.constant_(self.wd_bias, 1 / self.ts)
torch.nn.init.constant_(self.tanh_bias_weight, 0.0)
torch.nn.init.constant_(self.tanh_bias_bias, 0.0)
fan_in, _ = init._calculate_fan_in_and_fan_out(self.weight)
bound = 1 / math.sqrt(fan_in)
init.uniform_(self.bias, -bound, bound)
def extra_repr(self):
return 'in_features={}, out_features={}, bias={}'.format(self.
in_features, self.out_features, self.bias is not None)
def forward(self, input_0):
primals_2 = self.out_wd_weight
primals_3 = self.in_wd_weight
primals_4 = self.ly_wd_weight
primals_5 = self.wd_weight
primals_1 = self.weight
primals_6 = self.tanh_weight_weight
primals_7 = self.tanh_bias_weight
primals_8 = self.out_wd_bias
primals_9 = self.ly_wd_bias
primals_11 = self.wd_bias
primals_10 = self.bias
primals_12 = self.tanh_weight_bias
primals_13 = self.tanh_bias_bias
primals_14 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14])
return output[0]
| Lollipop321/weight-distillation | WDV52Linear | false | 5,568 | [
"BSD-3-Clause"
] | 1 | cfc76ec58e3e88094dde1825287b2968f9718431 | https://github.com/Lollipop321/weight-distillation/tree/cfc76ec58e3e88094dde1825287b2968f9718431 | import math
import torch
import torch.nn.functional as F
import torch.utils.data
from torch.nn import Parameter
import torch.onnx.operators
from torch.nn.parameter import Parameter
from torch.nn import init
import torch.optim
import torch.optim.lr_scheduler
class Model(torch.nn.Module):
"""Applies a linear transformation to the incoming data: :math:`y = xA^T + b`
Args:
in_features: size of each input sample
out_features: size of each output sample
bias: If set to ``False``, the layer will not learn an additive bias.
Default: ``True``
Shape:
- Input: :math:`(N, *, H_{in})` where :math:`*` means any number of
additional dimensions and :math:`H_{in} = \\text{in\\_features}`
- Output: :math:`(N, *, H_{out})` where all but the last dimension
are the same shape as the input and :math:`H_{out} = \\text{out\\_features}`.
Attributes:
weight: the learnable weights of the module of shape
:math:`(\\text{out\\_features}, \\text{in\\_features})`. The values are
initialized from :math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})`, where
:math:`k = \\frac{1}{\\text{in\\_features}}`
bias: the learnable bias of the module of shape :math:`(\\text{out\\_features})`.
If :attr:`bias` is ``True``, the values are initialized from
:math:`\\mathcal{U}(-\\sqrt{k}, \\sqrt{k})` where
:math:`k = \\frac{1}{\\text{in\\_features}}`
Examples::
>>> m = nn.Linear(20, 30)
>>> input = torch.randn(128, 20)
>>> output = m(input)
>>> print(output.size())
torch.Size([128, 30])
"""
__constants__ = ['bias', 'in_features', 'out_features']
def __init__(self, ts, in_features, out_features, wd_in_features,
wd_out_features, wd_decoder_layers, wd_require_gradient=False, bias
=True):
super().__init__()
self.in_features = in_features
self.out_features = out_features
self.ts = ts
self.wd_decoder_layers = wd_decoder_layers
self.out_wd_weight = Parameter(torch.Tensor(wd_out_features,
out_features))
self.in_wd_weight = Parameter(torch.Tensor(wd_in_features, in_features)
)
self.ly_wd_weight = Parameter(torch.Tensor(ts, ts))
self.wd_weight = Parameter(torch.Tensor(ts, 1))
self.weight = Parameter(torch.Tensor(wd_out_features,
wd_in_features, ts), requires_grad=wd_require_gradient)
self.tanh_weight_weight = Parameter(torch.Tensor(out_features,
in_features))
self.tanh_bias_weight = Parameter(torch.Tensor(out_features,
in_features))
if bias:
self.out_wd_bias = Parameter(torch.Tensor(wd_out_features,
out_features))
self.ly_wd_bias = Parameter(torch.Tensor(ts, ts))
self.wd_bias = Parameter(torch.Tensor(ts, 1))
self.bias = Parameter(torch.Tensor(wd_out_features, ts),
requires_grad=wd_require_gradient)
self.tanh_weight_bias = Parameter(torch.Tensor(out_features))
self.tanh_bias_bias = Parameter(torch.Tensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
torch.nn.init.xavier_normal_(self.out_wd_weight)
torch.nn.init.xavier_normal_(self.in_wd_weight)
torch.nn.init.xavier_normal_(self.ly_wd_weight)
torch.nn.init.constant_(self.wd_weight, 1 / self.ts)
torch.nn.init.constant_(self.tanh_weight_weight, 1.0)
torch.nn.init.constant_(self.tanh_weight_bias, 1.0)
init.kaiming_uniform_(self.weight, a=math.sqrt(5))
if self.bias is not None:
torch.nn.init.xavier_normal_(self.out_wd_bias)
torch.nn.init.xavier_normal_(self.ly_wd_bias)
torch.nn.init.constant_(self.wd_bias, 1 / self.ts)
# ... truncated (>4000 chars) for memory efficiency |
DsBlock | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/rl/crlzvualg2za225vkja5fjpnxss76chgi2glfm2hpmsgp2hm7zo3.py
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# out_1 => relu
# Graph fragment:
# %relu : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%squeeze,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%squeeze_5, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 64)
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(1, 1, 1), padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf5, 256, grid=grid(256), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [out_2], Original ATen: [aten.max_pool3d_with_indices]
buf2 = torch.ops.aten.max_pool3d_with_indices.default(buf1, [2, 2, 2], [2, 2, 2])
buf3 = buf2[0]
buf4 = buf2[1]
del buf2
return (buf3, buf1, primals_1, reinterpret_tensor(primals_3, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1), 0), buf1, buf4, buf5, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 3, 3, 3), (108, 27, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class DsBlock(nn.Module):
def __init__(self, in_channels, out_channels, pooling):
super(DsBlock, self).__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
self.pooling = pooling
if pooling:
self.mp = nn.MaxPool3d(kernel_size=2, stride=2, padding=0)
def forward(self, x):
out = self.conv(x)
out = self.relu(out)
before_pool = out
if self.pooling:
out = self.mp(out)
return out, before_pool
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'pooling': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 64
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3, 3), (108, 27, 9, 3, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_3, (1,
4, 4, 4, 4), (256, 64, 16, 4, 1), 0), primals_1, stride=(1, 1,
1), padding=(1, 1, 1), dilation=(1, 1, 1), transposed=False,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 4, 4, 4), (256, 64, 16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_2, buf5, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = torch.ops.aten.max_pool3d_with_indices.default(buf1, [2, 2,
2], [2, 2, 2])
buf3 = buf2[0]
buf4 = buf2[1]
del buf2
return buf3, buf1, primals_1, reinterpret_tensor(primals_3, (1, 4, 4, 4,
4), (256, 64, 16, 4, 1), 0), buf1, buf4, buf5
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class DsBlockNew(nn.Module):
def __init__(self, in_channels, out_channels, pooling):
super(DsBlockNew, self).__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
self.pooling = pooling
if pooling:
self.mp = nn.MaxPool3d(kernel_size=2, stride=2, padding=0)
def forward(self, input_0):
primals_1 = self.conv.weight
primals_2 = self.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0], output[1]
| MATHplus-Young-Academy/P2-Cardiac-Motion | DsBlock | false | 5,569 | [
"Apache-2.0"
] | 1 | 844995e8e5760f981c425d13c0bd7f2f3bb8baec | https://github.com/MATHplus-Young-Academy/P2-Cardiac-Motion/tree/844995e8e5760f981c425d13c0bd7f2f3bb8baec | import torch
import torch.nn as nn
def conv3x3(in_channels, out_channels, stride=1, padding=1, bias=True):
return nn.Conv3d(in_channels, out_channels, kernel_size=3, stride=
stride, padding=padding, bias=bias)
class Model(nn.Module):
def __init__(self, in_channels, out_channels, pooling):
super().__init__()
self.conv = conv3x3(in_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
self.pooling = pooling
if pooling:
self.mp = nn.MaxPool3d(kernel_size=2, stride=2, padding=0)
def forward(self, x):
out = self.conv(x)
out = self.relu(out)
before_pool = out
if self.pooling:
out = self.mp(out)
return out, before_pool
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
BinaryDiceLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/gd/cgdtd7ki7lurypoeyfwjebdfquygdeupjef4ltfbbbdk5u7owcpl.py
# Topologically Sorted Source Nodes: [intersection, sum_1, sum_2, sum_3], Original ATen: [aten.mul, aten.sum]
# Source node to ATen node mapping:
# intersection => mul
# sum_1 => sum_1
# sum_2 => sum_2
# sum_3 => sum_3
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view, %view_1), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [1]), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%view, [1]), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%view_1, [1]), kwargs = {})
triton_per_fused_mul_sum_0 = async_compile.triton('triton_per_fused_mul_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mul_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 3, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (64*x0)), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + (r1 + (64*x0)), xmask, other=0.0)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp10 = tl.where(xmask, tmp8, 0)
tmp11 = tl.sum(tmp10, 1)[:, None]
tmp12 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp14 = tl.where(xmask, tmp12, 0)
tmp15 = tl.sum(tmp14, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp7, xmask)
tl.store(out_ptr1 + (x0), tmp11, xmask)
tl.store(out_ptr2 + (x0), tmp15, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/e4/ce4meaau7uw5ohjmea5wahpx7ugnh7nquzqbbvbbeqmpgntgbyen.py
# Topologically Sorted Source Nodes: [add, mul_1, add_1, add_2, score, sum_4, truediv_1, score_1], Original ATen: [aten.add, aten.mul, aten.div, aten.sum, aten.rsub]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# add_2 => add_2
# mul_1 => mul_1
# score => div
# score_1 => sub
# sum_4 => sum_4
# truediv_1 => div_1
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_1, 1e-15), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_2, %sum_3), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, 1e-15), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_1, %add_2), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%div,), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_4, 4), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %div_1), kwargs = {})
triton_per_fused_add_div_mul_rsub_sum_1 = async_compile.triton('triton_per_fused_add_div_mul_rsub_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 4],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {4: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mul_rsub_sum_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_mul_rsub_sum_1(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp5 = tl.load(in_ptr1 + (r0), None)
tmp6 = tl.load(in_ptr2 + (r0), None)
tmp1 = 1e-15
tmp2 = tmp0 + tmp1
tmp3 = 2.0
tmp4 = tmp2 * tmp3
tmp7 = tmp5 + tmp6
tmp8 = tmp7 + tmp1
tmp9 = tmp4 / tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.sum(tmp10, 1)[:, None]
tmp13 = 0.25
tmp14 = tmp12 * tmp13
tmp15 = 1.0
tmp16 = tmp15 - tmp14
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp16, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, ), (1, ), torch.float32)
buf1 = empty_strided_cuda((4, ), (1, ), torch.float32)
buf2 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [intersection, sum_1, sum_2, sum_3], Original ATen: [aten.mul, aten.sum]
stream0 = get_raw_stream(0)
triton_per_fused_mul_sum_0.run(arg1_1, arg0_1, buf0, buf1, buf2, 4, 64, grid=grid(4), stream=stream0)
del arg0_1
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [add, mul_1, add_1, add_2, score, sum_4, truediv_1, score_1], Original ATen: [aten.add, aten.mul, aten.div, aten.sum, aten.rsub]
triton_per_fused_add_div_mul_rsub_sum_1.run(buf4, buf0, buf1, buf2, 1, 4, grid=grid(1), stream=stream0)
del buf0
del buf1
del buf2
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class BinaryDiceLoss(nn.Module):
"""SoftDice loss
"""
def __init__(self):
super(BinaryDiceLoss, self).__init__()
self.SM = nn.Sigmoid()
def forward(self, logits, labels):
num = labels.size(0)
m1 = self.SM(logits).view(num, -1)
m2 = labels.view(num, -1)
intersection = m1 * m2
score = 2.0 * (intersection.sum(1) + 1e-15) / (m1.sum(1) + m2.sum(1
) + 1e-15)
score = 1 - score.sum() / num
return score
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1,
out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 4
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 64 * x0), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + (r1 + 64 * x0), xmask, other=0.0)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tmp8 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp10 = tl.where(xmask, tmp8, 0)
tmp11 = tl.sum(tmp10, 1)[:, None]
tmp12 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp14 = tl.where(xmask, tmp12, 0)
tmp15 = tl.sum(tmp14, 1)[:, None]
tl.store(out_ptr0 + x0, tmp7, xmask)
tl.store(out_ptr1 + x0, tmp11, xmask)
tl.store(out_ptr2 + x0, tmp15, xmask)
@triton.jit
def triton_per_fused_add_div_mul_rsub_sum_1(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp5 = tl.load(in_ptr1 + r0, None)
tmp6 = tl.load(in_ptr2 + r0, None)
tmp1 = 1e-15
tmp2 = tmp0 + tmp1
tmp3 = 2.0
tmp4 = tmp2 * tmp3
tmp7 = tmp5 + tmp6
tmp8 = tmp7 + tmp1
tmp9 = tmp4 / tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.sum(tmp10, 1)[:, None]
tmp13 = 0.25
tmp14 = tmp12 * tmp13
tmp15 = 1.0
tmp16 = tmp15 - tmp14
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp16, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4,), (1,), torch.float32)
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
buf2 = empty_strided_cuda((4,), (1,), torch.float32)
get_raw_stream(0)
triton_per_fused_mul_sum_0[grid(4)](arg1_1, arg0_1, buf0, buf1,
buf2, 4, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
triton_per_fused_add_div_mul_rsub_sum_1[grid(1)](buf4, buf0, buf1,
buf2, 1, 4, XBLOCK=1, num_warps=2, num_stages=1)
del buf0
del buf1
del buf2
return buf4,
class BinaryDiceLossNew(nn.Module):
"""SoftDice loss
"""
def __init__(self):
super(BinaryDiceLossNew, self).__init__()
self.SM = nn.Sigmoid()
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MIPT-Oulu/3D-Histo-Grading | BinaryDiceLoss | false | 5,570 | [
"MIT"
] | 1 | b779a154d0e5b104fc152c8952124768fb7b1dc6 | https://github.com/MIPT-Oulu/3D-Histo-Grading/tree/b779a154d0e5b104fc152c8952124768fb7b1dc6 | import torch
import torch.nn as nn
class Model(nn.Module):
"""SoftDice loss
"""
def __init__(self):
super().__init__()
self.SM = nn.Sigmoid()
def forward(self, logits, labels):
num = labels.size(0)
m1 = self.SM(logits).view(num, -1)
m2 = labels.view(num, -1)
intersection = m1 * m2
score = 2.0 * (intersection.sum(1) + 1e-15) / (m1.sum(1) + m2.sum(1
) + 1e-15)
score = 1 - score.sum() / num
return score
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
MultiHeadAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/an/cancvgq5crrl53xjlus3mms3in3pgeh5vjfezvvfgpcqipjh7byq.py
# Topologically Sorted Source Nodes: [q_2, x], Original ATen: [aten.mul, aten.clone]
# Source node to ATen node mapping:
# q_2 => mul
# x => clone
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_3, 1.0), kwargs = {})
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_mul_0 = async_compile.triton('triton_poi_fused_clone_mul_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_mul_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_mul_0(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x2 + (4*y3)), tmp4, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/mh/cmhet4vfl4jlxtge4zzaaa2nugvxpr5f4ge7rs72qwe2aow7vxwy.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x => clone_1
# Graph fragment:
# %clone_1 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_1,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_1 = async_compile.triton('triton_poi_fused_clone_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_1(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + (4*y3)), tmp2, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/tt/cttmvktt3m2x2nl56afa7l3abaxt7wlehowakdzngkhgs35f3n7u.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_1 => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_11, [3], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_11, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_2 = async_compile.triton('triton_poi_fused__softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ry/cryn7ntc2gpkbfzbre3xh7lffx7zkbskw6oihbzsekkgajmdbki6.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_1 => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [3], True), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_3 = async_compile.triton('triton_poi_fused__softmax_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_3(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/6t/c6t5a5ere3lqjiu7zh3uu4oxmpdoujdaqqmeunxqapgzo4m74uav.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x_4 => clone_4
# Graph fragment:
# %clone_4 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_7,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_4 = async_compile.triton('triton_poi_fused_clone_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_4(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + (4*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, ), (1, ))
assert_size_stride(primals_9, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_6, (16, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf1)
del primals_4
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_9, (16, 4), (4, 1), 0), reinterpret_tensor(primals_7, (4, 4), (1, 4), 0), out=buf2)
del primals_7
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [q_2, x], Original ATen: [aten.mul, aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_mul_0.run(buf0, primals_3, buf3, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_3
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.clone]
triton_poi_fused_clone_1.run(buf1, primals_5, buf4, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_5
buf5 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf3, (16, 4, 1), (4, 1, 0), 0), reinterpret_tensor(buf4, (16, 1, 4), (4, 0, 1), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf5, buf6, 256, grid=grid(256), stream=stream0)
buf7 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_3.run(buf6, buf7, 256, grid=grid(256), stream=stream0)
del buf6
buf8 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.clone]
triton_poi_fused_clone_1.run(buf2, primals_8, buf8, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_8
buf9 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf7, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf8, (16, 4, 1), (4, 1, 0), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.clone]
triton_poi_fused_clone_4.run(buf9, buf10, 16, 4, grid=grid(16, 4), stream=stream0)
buf11 = reinterpret_tensor(buf9, (16, 4), (4, 1), 0); del buf9 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf10, (16, 4), (4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf11)
del primals_11
return (reinterpret_tensor(buf11, (4, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (16, 4), (4, 1), 0), reinterpret_tensor(primals_9, (16, 4), (4, 1), 0), buf7, reinterpret_tensor(buf10, (16, 4), (4, 1), 0), primals_10, reinterpret_tensor(buf8, (16, 1, 4), (4, 1, 1), 0), reinterpret_tensor(buf3, (16, 1, 4), (4, 1, 1), 0), reinterpret_tensor(buf4, (16, 4, 1), (4, 1, 4), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class MultiHeadAttention(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super(MultiHeadAttention, self).__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, q, k, v, attn_bias=None):
orig_q_size = q.size()
d_k = self.att_size
d_v = self.att_size
batch_size = q.size(0)
q = self.linear_q(q).view(batch_size, -1, self.num_heads, d_k)
k = self.linear_k(k).view(batch_size, -1, self.num_heads, d_k)
v = self.linear_v(v).view(batch_size, -1, self.num_heads, d_v)
q = q.transpose(1, 2)
v = v.transpose(1, 2)
k = k.transpose(1, 2).transpose(2, 3)
q = q * self.scale
x = torch.matmul(q, k)
if attn_bias is not None:
x = x + attn_bias
x = torch.softmax(x, dim=3)
x = self.att_dropout(x)
x = x.matmul(v)
x = x.transpose(1, 2).contiguous()
x = x.view(batch_size, -1, self.num_heads * d_v)
x = self.output_layer(x)
assert x.size() == orig_q_size
return x
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [[], {'hidden_size': 4, 'attention_dropout_rate': 0.5,
'num_heads': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_mul_0(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x2 + 4 * y3), tmp4, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_1(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 4 * y3), tmp2, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_clone_4(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4,), (1,))
assert_size_stride(primals_9, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_6, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf1)
del primals_4
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_9, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_7, (4, 4), (1, 4), 0), out=buf2)
del primals_7
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_mul_0[grid(16, 4)](buf0, primals_3, buf3, 16,
4, XBLOCK=4, YBLOCK=8, num_warps=1, num_stages=1)
del primals_3
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_clone_1[grid(16, 4)](buf1, primals_5, buf4, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_5
buf5 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf3, (16, 4, 1), (4, 1, 0),
0), reinterpret_tensor(buf4, (16, 1, 4), (4, 0, 1), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_2[grid(256)](buf5, buf6, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf7 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused__softmax_3[grid(256)](buf6, buf7, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del buf6
buf8 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf1
triton_poi_fused_clone_1[grid(16, 4)](buf2, primals_8, buf8, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_8
buf9 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf7, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf8, (16, 4, 1), (4, 1, 0), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
triton_poi_fused_clone_4[grid(16, 4)](buf9, buf10, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf11 = reinterpret_tensor(buf9, (16, 4), (4, 1), 0)
del buf9
extern_kernels.addmm(primals_11, reinterpret_tensor(buf10, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf11)
del primals_11
return reinterpret_tensor(buf11, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_6, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_9, (16, 4), (4, 1), 0
), buf7, reinterpret_tensor(buf10, (16, 4), (4, 1), 0
), primals_10, reinterpret_tensor(buf8, (16, 1, 4), (4, 1, 1), 0
), reinterpret_tensor(buf3, (16, 1, 4), (4, 1, 1), 0
), reinterpret_tensor(buf4, (16, 4, 1), (4, 1, 4), 0)
class MultiHeadAttentionNew(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super(MultiHeadAttentionNew, self).__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, input_0, input_1, input_2):
primals_2 = self.linear_q.weight
primals_3 = self.linear_q.bias
primals_4 = self.linear_k.weight
primals_5 = self.linear_k.bias
primals_7 = self.linear_v.weight
primals_8 = self.linear_v.bias
primals_10 = self.output_layer.weight
primals_11 = self.output_layer.bias
primals_1 = input_0
primals_6 = input_1
primals_9 = input_2
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11])
return output[0]
| Luo-Chang/Graphormer | MultiHeadAttention | false | 5,571 | [
"MIT"
] | 1 | b35b3ca6369e25cdae80e1617bfc3921feeb3158 | https://github.com/Luo-Chang/Graphormer/tree/b35b3ca6369e25cdae80e1617bfc3921feeb3158 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super().__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, q, k, v, attn_bias=None):
orig_q_size = q.size()
d_k = self.att_size
d_v = self.att_size
batch_size = q.size(0)
q = self.linear_q(q).view(batch_size, -1, self.num_heads, d_k)
k = self.linear_k(k).view(batch_size, -1, self.num_heads, d_k)
v = self.linear_v(v).view(batch_size, -1, self.num_heads, d_v)
q = q.transpose(1, 2)
v = v.transpose(1, 2)
k = k.transpose(1, 2).transpose(2, 3)
q = q * self.scale
x = torch.matmul(q, k)
if attn_bias is not None:
x = x + attn_bias
x = torch.softmax(x, dim=3)
x = self.att_dropout(x)
x = x.matmul(v)
x = x.transpose(1, 2).contiguous()
x = x.view(batch_size, -1, self.num_heads * d_v)
x = self.output_layer(x)
assert x.size() == orig_q_size
return x
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4, 4])
]
def get_init_inputs():
return [[], {'hidden_size': 4, 'attention_dropout_rate': 0.5,
'num_heads': 4}]
|
Discrete | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/3n/c3nyedrc56xoj6pmjzzgnpithkx2vti6qsjnj43ybcoj67zutjs4.py
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# softmax => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%arg0_1, [0], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (64 + x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (128 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (192 + x0), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/y3/cy3yhdklte2jljt3jlkxw4g7pzz4g3oiwcgjauhd3xpun5n7blb6.py
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# softmax => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [0], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (64 + x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (128 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (192 + x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
del buf0
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Discrete(nn.Module):
def __init__(self):
super(Discrete, self).__init__()
def forward(self, x):
return nn.functional.softmax(x, dim=0)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (64 + x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (128 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (192 + x0), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (64 + x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (128 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (192 + x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(256)](arg0_1, buf0, 256, XBLOCK=
128, num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_1[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf0
return buf1,
class DiscreteNew(nn.Module):
def __init__(self):
super(DiscreteNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MPGek/client | Discrete | false | 5,572 | [
"Apache-2.0"
] | 1 | 541d760c5cb8776b1ad5fcf1362d7382811cbc61 | https://github.com/MPGek/client/tree/541d760c5cb8776b1ad5fcf1362d7382811cbc61 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, x):
return nn.functional.softmax(x, dim=0)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
AttentionPool2d | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ov/covbryzjnff2kb26c5gkcqbvct6kdwzanlx3iu6ee24itsit76o3.py
# Topologically Sorted Source Nodes: [mean], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# mean => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view, [-1], True), kwargs = {})
triton_per_fused_mean_0 = async_compile.triton('triton_per_fused_mean_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[16, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mean_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mean_0(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 16
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (16*x0)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ct/cctgbe64jgxq3sxjjjqccvq653sunfecfcizp3jcofnl7uiib7wo.py
# Topologically Sorted Source Nodes: [x_1, x_2], Original ATen: [aten.cat, aten.add]
# Source node to ATen node mapping:
# x_1 => cat
# x_2 => add
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%mean, %view], -1), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%cat, %unsqueeze), kwargs = {})
triton_poi_fused_add_cat_1 = async_compile.triton('triton_poi_fused_add_cat_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_cat_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_cat_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 17
x3 = (xindex // 17)
x4 = xindex % 68
x5 = xindex
tmp15 = tl.load(in_ptr2 + (x4), xmask, eviction_policy='evict_last')
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x3), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = 16.0
tmp7 = tmp5 / tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tmp11 = tl.full([1], 17, tl.int64)
tmp12 = tmp0 < tmp11
tmp13 = tl.load(in_ptr1 + ((16*x3) + ((-1) + x0)), tmp10 & xmask, eviction_policy='evict_last', other=0.0)
tmp14 = tl.where(tmp4, tmp9, tmp13)
tmp16 = tmp14 + tmp15
tl.store(out_ptr0 + (x5), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rq/crqhbpxwwpvdowuqzuzvjdahxv45tx2y4dpxda2rurvr5kralgbn.py
# Topologically Sorted Source Nodes: [mul], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# mul => mul
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%getitem, 0.7071067811865475), kwargs = {})
triton_poi_fused_mul_2 = async_compile.triton('triton_poi_fused_mul_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 68)
x3 = xindex % 68
x1 = (xindex // 17) % 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x3 + (204*x2)), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.7071067811865475
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x4), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/cr/ccrjqh2olqyevqox4t3kpoubo2s2m44cuvtxmf2k3slvdcgmmlcm.py
# Topologically Sorted Source Nodes: [mul_1], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# mul_1 => mul_1
# Graph fragment:
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%getitem_1, 0.7071067811865475), kwargs = {})
triton_poi_fused_mul_3 = async_compile.triton('triton_poi_fused_mul_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_3(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 68)
x3 = xindex % 68
x1 = (xindex // 17) % 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + (68 + x3 + (204*x2)), xmask)
tmp1 = tl.load(in_ptr1 + (4 + x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.7071067811865475
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x4), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/6p/c6p34hneq7lp7a3tjiwk44lxqe2hzbpdjgbgmanamv35xvznwb2j.py
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# softmax => amax, div, exp, sub, sum_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_6, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_6, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_per_fused__softmax_4 = async_compile.triton('triton_per_fused__softmax_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[128, 32],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__softmax_4(in_ptr0, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 68
rnumel = 17
RBLOCK: tl.constexpr = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (17*x0)), rmask & xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(rmask & xmask, tmp1, float("-inf"))
tmp4 = triton_helpers.max2(tmp3, 1)[:, None]
tmp5 = tmp0 - tmp4
tmp6 = tl_math.exp(tmp5)
tmp7 = tl.broadcast_to(tmp6, [XBLOCK, RBLOCK])
tmp9 = tl.where(rmask & xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tmp6 / tmp10
tl.store(out_ptr2 + (r1 + (17*x0)), tmp11, rmask & xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/wf/cwf36kt6t5p6sv4fjknukcyy4vz6ejfamnreogfggthvquklzkf6.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_3 => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%add, %primals_3, %primals_4, [1], [0], [1], False, [0], 1), kwargs = {})
triton_poi_fused_convolution_5 = async_compile.triton('triton_poi_fused_convolution_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_5(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 816
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 17) % 12
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x3), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/oe/coet3g5rxo652nti4d4ogiwpej2mrpgh4cyidpv6aivegc4mda7s.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_5 => convolution_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%view_11, %primals_5, %primals_6, [1], [0], [1], False, [0], 1), kwargs = {})
triton_poi_fused_convolution_6 = async_compile.triton('triton_poi_fused_convolution_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 32], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_6(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 17
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (68*y1)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + (17*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/uk/cuk32wvqx6lzak6biu2zut26pzpqa3ell4xsxs3qv3cpuuon7kkw.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_5 => convolution_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%view_11, %primals_5, %primals_6, [1], [0], [1], False, [0], 1), kwargs = {})
triton_poi_fused_convolution_7 = async_compile.triton('triton_poi_fused_convolution_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_7', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_7(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 17) % 4
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x3), tmp2, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 17), (17, 1))
assert_size_stride(primals_3, (12, 4, 1), (4, 1, 1))
assert_size_stride(primals_4, (12, ), (1, ))
assert_size_stride(primals_5, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_6, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [mean], Original ATen: [aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_mean_0.run(primals_1, buf0, 16, 16, grid=grid(16), stream=stream0)
buf1 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1, x_2], Original ATen: [aten.cat, aten.add]
triton_poi_fused_add_cat_1.run(buf0, primals_1, primals_2, buf1, 272, grid=grid(272), stream=stream0)
del buf0
del primals_1
del primals_2
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf2, (4, 12, 17), (204, 17, 1))
buf3 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul], Original ATen: [aten.mul]
triton_poi_fused_mul_2.run(buf2, primals_4, buf3, 272, grid=grid(272), stream=stream0)
buf4 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul_1], Original ATen: [aten.mul]
triton_poi_fused_mul_3.run(buf2, primals_4, buf4, 272, grid=grid(272), stream=stream0)
buf5 = empty_strided_cuda((4, 17, 17), (289, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf3, (4, 17, 4), (68, 1, 17), 0), buf4, out=buf5)
buf8 = empty_strided_cuda((4, 17, 17), (289, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
triton_per_fused__softmax_4.run(buf5, buf8, 68, 17, grid=grid(68), stream=stream0)
del buf5
buf9 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.convolution]
triton_poi_fused_convolution_5.run(buf9, primals_4, 816, grid=grid(816), stream=stream0)
del primals_4
buf10 = empty_strided_cuda((4, 17, 4), (68, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [a], Original ATen: [aten.bmm]
extern_kernels.bmm(buf8, reinterpret_tensor(buf9, (4, 17, 4), (204, 1, 17), 136), out=buf10)
buf11 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
triton_poi_fused_convolution_6.run(buf10, buf11, 16, 17, grid=grid(16, 17), stream=stream0)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_5, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf12, (4, 4, 17), (68, 17, 1))
del buf11
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
triton_poi_fused_convolution_7.run(buf13, primals_6, 272, grid=grid(272), stream=stream0)
del primals_6
return (reinterpret_tensor(buf13, (4, 4), (68, 17), 0), primals_3, primals_5, buf1, buf8, reinterpret_tensor(buf10, (4, 4, 17), (68, 1, 4), 0), reinterpret_tensor(buf9, (4, 4, 17), (204, 17, 1), 136), buf3, reinterpret_tensor(buf4, (4, 17, 4), (68, 1, 17), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 17), (17, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((12, 4, 1), (4, 1, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((12, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 4, 1), (4, 1, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import numpy as np
import torch as th
import torch.nn as nn
def count_flops_attn(model, _x, y):
"""
A counter for the `thop` package to count the operations in an
attention operation.
Meant to be used like:
macs, params = thop.profile(
model,
inputs=(inputs, timestamps),
custom_ops={QKVAttention: QKVAttention.count_flops},
)
"""
b, c, *spatial = y[0].shape
num_spatial = int(np.prod(spatial))
matmul_ops = 2 * b * num_spatial ** 2 * c
model.total_ops += th.DoubleTensor([matmul_ops])
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class QKVAttention(nn.Module):
"""
A module which performs QKV attention and splits in a different order.
"""
def __init__(self, n_heads):
super().__init__()
self.n_heads = n_heads
def forward(self, qkv):
"""
Apply QKV attention.
:param qkv: an [N x (3 * H * C) x T] tensor of Qs, Ks, and Vs.
:return: an [N x (H * C) x T] tensor after attention.
"""
bs, width, length = qkv.shape
assert width % (3 * self.n_heads) == 0
ch = width // (3 * self.n_heads)
q, k, v = qkv.chunk(3, dim=1)
scale = 1 / math.sqrt(math.sqrt(ch))
weight = th.einsum('bct,bcs->bts', (q * scale).view(bs * self.
n_heads, ch, length), (k * scale).view(bs * self.n_heads, ch,
length))
weight = th.softmax(weight.float(), dim=-1).type(weight.dtype)
a = th.einsum('bts,bcs->bct', weight, v.reshape(bs * self.n_heads,
ch, length))
return a.reshape(bs, -1, length)
@staticmethod
def count_flops(model, _x, y):
return count_flops_attn(model, _x, y)
class AttentionPool2d(nn.Module):
"""
Adapted from CLIP: https://github.com/openai/CLIP/blob/main/clip/model.py
"""
def __init__(self, spacial_dim: 'int', embed_dim: 'int',
num_heads_channels: 'int', output_dim: 'int'=None):
super().__init__()
self.positional_embedding = nn.Parameter(th.randn(embed_dim,
spacial_dim ** 2 + 1) / embed_dim ** 0.5)
self.qkv_proj = conv_nd(1, embed_dim, 3 * embed_dim, 1)
self.c_proj = conv_nd(1, embed_dim, output_dim or embed_dim, 1)
self.num_heads = embed_dim // num_heads_channels
self.attention = QKVAttention(self.num_heads)
def forward(self, x):
b, c, *_spatial = x.shape
x = x.reshape(b, c, -1)
x = th.cat([x.mean(dim=-1, keepdim=True), x], dim=-1)
x = x + self.positional_embedding[None, :, :]
x = self.qkv_proj(x)
x = self.attention(x)
x = self.c_proj(x)
return x[:, :, 0]
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'spacial_dim': 4, 'embed_dim': 4, 'num_heads_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import math
import numpy as np
import torch as th
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_mean_0(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK: tl.
constexpr):
xnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 16 * x0), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused_add_cat_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 17
x3 = xindex // 17
x4 = xindex % 68
x5 = xindex
tmp15 = tl.load(in_ptr2 + x4, xmask, eviction_policy='evict_last')
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + x3, tmp4 & xmask, eviction_policy='evict_last',
other=0.0)
tmp6 = 16.0
tmp7 = tmp5 / tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tl.full([1], 17, tl.int64)
tmp13 = tl.load(in_ptr1 + (16 * x3 + (-1 + x0)), tmp10 & xmask,
eviction_policy='evict_last', other=0.0)
tmp14 = tl.where(tmp4, tmp9, tmp13)
tmp16 = tmp14 + tmp15
tl.store(out_ptr0 + x5, tmp16, xmask)
@triton.jit
def triton_poi_fused_mul_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 68
x3 = xindex % 68
x1 = xindex // 17 % 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x3 + 204 * x2), xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.7071067811865475
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x4, tmp4, xmask)
@triton.jit
def triton_poi_fused_mul_3(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 68
x3 = xindex % 68
x1 = xindex // 17 % 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + (68 + x3 + 204 * x2), xmask)
tmp1 = tl.load(in_ptr1 + (4 + x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.7071067811865475
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x4, tmp4, xmask)
@triton.jit
def triton_per_fused__softmax_4(in_ptr0, out_ptr2, xnumel, rnumel, XBLOCK:
tl.constexpr):
xnumel = 68
rnumel = 17
RBLOCK: tl.constexpr = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 17 * x0), rmask & xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(rmask & xmask, tmp1, float('-inf'))
tmp4 = triton_helpers.max2(tmp3, 1)[:, None]
tmp5 = tmp0 - tmp4
tmp6 = tl_math.exp(tmp5)
tmp7 = tl.broadcast_to(tmp6, [XBLOCK, RBLOCK])
tmp9 = tl.where(rmask & xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tmp6 / tmp10
tl.store(out_ptr2 + (r1 + 17 * x0), tmp11, rmask & xmask)
@triton.jit
def triton_poi_fused_convolution_5(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 816
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 17 % 12
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
@triton.jit
def triton_poi_fused_convolution_6(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 17
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 68 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 17 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_convolution_7(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 272
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 17 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 17), (17, 1))
assert_size_stride(primals_3, (12, 4, 1), (4, 1, 1))
assert_size_stride(primals_4, (12,), (1,))
assert_size_stride(primals_5, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_6, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
get_raw_stream(0)
triton_per_fused_mean_0[grid(16)](primals_1, buf0, 16, 16, XBLOCK=1,
num_warps=2, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
triton_poi_fused_add_cat_1[grid(272)](buf0, primals_1, primals_2,
buf1, 272, XBLOCK=256, num_warps=4, num_stages=1)
del buf0
del primals_1
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf2, (4, 12, 17), (204, 17, 1))
buf3 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
triton_poi_fused_mul_2[grid(272)](buf2, primals_4, buf3, 272,
XBLOCK=128, num_warps=4, num_stages=1)
buf4 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
triton_poi_fused_mul_3[grid(272)](buf2, primals_4, buf4, 272,
XBLOCK=256, num_warps=4, num_stages=1)
buf5 = empty_strided_cuda((4, 17, 17), (289, 17, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf3, (4, 17, 4), (68, 1, 17),
0), buf4, out=buf5)
buf8 = empty_strided_cuda((4, 17, 17), (289, 17, 1), torch.float32)
triton_per_fused__softmax_4[grid(68)](buf5, buf8, 68, 17, XBLOCK=8,
num_warps=2, num_stages=1)
del buf5
buf9 = buf2
del buf2
triton_poi_fused_convolution_5[grid(816)](buf9, primals_4, 816,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_4
buf10 = empty_strided_cuda((4, 17, 4), (68, 4, 1), torch.float32)
extern_kernels.bmm(buf8, reinterpret_tensor(buf9, (4, 17, 4), (204,
1, 17), 136), out=buf10)
buf11 = empty_strided_cuda((4, 4, 17), (68, 17, 1), torch.float32)
triton_poi_fused_convolution_6[grid(16, 17)](buf10, buf11, 16, 17,
XBLOCK=32, YBLOCK=16, num_warps=4, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_5, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf12, (4, 4, 17), (68, 17, 1))
del buf11
buf13 = buf12
del buf12
triton_poi_fused_convolution_7[grid(272)](buf13, primals_6, 272,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_6
return reinterpret_tensor(buf13, (4, 4), (68, 17), 0
), primals_3, primals_5, buf1, buf8, reinterpret_tensor(buf10, (4,
4, 17), (68, 1, 4), 0), reinterpret_tensor(buf9, (4, 4, 17), (204,
17, 1), 136), buf3, reinterpret_tensor(buf4, (4, 17, 4), (68, 1, 17), 0
)
def count_flops_attn(model, _x, y):
"""
A counter for the `thop` package to count the operations in an
attention operation.
Meant to be used like:
macs, params = thop.profile(
model,
inputs=(inputs, timestamps),
custom_ops={QKVAttention: QKVAttention.count_flops},
)
"""
b, c, *spatial = y[0].shape
num_spatial = int(np.prod(spatial))
matmul_ops = 2 * b * num_spatial ** 2 * c
model.total_ops += th.DoubleTensor([matmul_ops])
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class QKVAttention(nn.Module):
"""
A module which performs QKV attention and splits in a different order.
"""
def __init__(self, n_heads):
super().__init__()
self.n_heads = n_heads
def forward(self, qkv):
"""
Apply QKV attention.
:param qkv: an [N x (3 * H * C) x T] tensor of Qs, Ks, and Vs.
:return: an [N x (H * C) x T] tensor after attention.
"""
bs, width, length = qkv.shape
assert width % (3 * self.n_heads) == 0
ch = width // (3 * self.n_heads)
q, k, v = qkv.chunk(3, dim=1)
scale = 1 / math.sqrt(math.sqrt(ch))
weight = th.einsum('bct,bcs->bts', (q * scale).view(bs * self.
n_heads, ch, length), (k * scale).view(bs * self.n_heads, ch,
length))
weight = th.softmax(weight.float(), dim=-1).type(weight.dtype)
a = th.einsum('bts,bcs->bct', weight, v.reshape(bs * self.n_heads,
ch, length))
return a.reshape(bs, -1, length)
@staticmethod
def count_flops(model, _x, y):
return count_flops_attn(model, _x, y)
class AttentionPool2dNew(nn.Module):
"""
Adapted from CLIP: https://github.com/openai/CLIP/blob/main/clip/model.py
"""
def __init__(self, spacial_dim: 'int', embed_dim: 'int',
num_heads_channels: 'int', output_dim: 'int'=None):
super().__init__()
self.positional_embedding = nn.Parameter(th.randn(embed_dim,
spacial_dim ** 2 + 1) / embed_dim ** 0.5)
self.qkv_proj = conv_nd(1, embed_dim, 3 * embed_dim, 1)
self.c_proj = conv_nd(1, embed_dim, output_dim or embed_dim, 1)
self.num_heads = embed_dim // num_heads_channels
self.attention = QKVAttention(self.num_heads)
def forward(self, input_0):
primals_2 = self.positional_embedding
primals_3 = self.qkv_proj.weight
primals_4 = self.qkv_proj.bias
primals_5 = self.c_proj.weight
primals_6 = self.c_proj.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| Liujingxiu23/guided-diffusion | AttentionPool2d | false | 5,573 | [
"MIT"
] | 1 | 0ba878e517b276c45d1195eb29f6f5f72659a05b | https://github.com/Liujingxiu23/guided-diffusion/tree/0ba878e517b276c45d1195eb29f6f5f72659a05b | import math
import torch
import numpy as np
import torch as th
import torch.nn as nn
def count_flops_attn(model, _x, y):
"""
A counter for the `thop` package to count the operations in an
attention operation.
Meant to be used like:
macs, params = thop.profile(
model,
inputs=(inputs, timestamps),
custom_ops={QKVAttention: QKVAttention.count_flops},
)
"""
b, c, *spatial = y[0].shape
num_spatial = int(np.prod(spatial))
matmul_ops = 2 * b * num_spatial ** 2 * c
model.total_ops += th.DoubleTensor([matmul_ops])
def conv_nd(dims, *args, **kwargs):
"""
Create a 1D, 2D, or 3D convolution module.
"""
if dims == 1:
return nn.Conv1d(*args, **kwargs)
elif dims == 2:
return nn.Conv2d(*args, **kwargs)
elif dims == 3:
return nn.Conv3d(*args, **kwargs)
raise ValueError(f'unsupported dimensions: {dims}')
class QKVAttention(nn.Module):
"""
A module which performs QKV attention and splits in a different order.
"""
def __init__(self, n_heads):
super().__init__()
self.n_heads = n_heads
def forward(self, qkv):
"""
Apply QKV attention.
:param qkv: an [N x (3 * H * C) x T] tensor of Qs, Ks, and Vs.
:return: an [N x (H * C) x T] tensor after attention.
"""
bs, width, length = qkv.shape
assert width % (3 * self.n_heads) == 0
ch = width // (3 * self.n_heads)
q, k, v = qkv.chunk(3, dim=1)
scale = 1 / math.sqrt(math.sqrt(ch))
weight = th.einsum('bct,bcs->bts', (q * scale).view(bs * self.
n_heads, ch, length), (k * scale).view(bs * self.n_heads, ch,
length))
weight = th.softmax(weight.float(), dim=-1).type(weight.dtype)
a = th.einsum('bts,bcs->bct', weight, v.reshape(bs * self.n_heads,
ch, length))
return a.reshape(bs, -1, length)
@staticmethod
def count_flops(model, _x, y):
return count_flops_attn(model, _x, y)
class Model(nn.Module):
"""
Adapted from CLIP: https://github.com/openai/CLIP/blob/main/clip/model.py
"""
def __init__(self, spacial_dim: 'int', embed_dim: 'int',
num_heads_channels: 'int', output_dim: 'int'=None):
super().__init__()
self.positional_embedding = nn.Parameter(th.randn(embed_dim,
spacial_dim ** 2 + 1) / embed_dim ** 0.5)
self.qkv_proj = conv_nd(1, embed_dim, 3 * embed_dim, 1)
self.c_proj = conv_nd(1, embed_dim, output_dim or embed_dim, 1)
self.num_heads = embed_dim // num_heads_channels
self.attention = QKVAttention(self.num_heads)
def forward(self, x):
b, c, *_spatial = x.shape
x = x.reshape(b, c, -1)
x = th.cat([x.mean(dim=-1, keepdim=True), x], dim=-1)
x = x + self.positional_embedding[None, :, :]
x = self.qkv_proj(x)
x = self.attention(x)
x = self.c_proj(x)
return x[:, :, 0]
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
CrossEntropyDiceLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/34/c34rlteenetjierkjaoyb2s72hhjrg66q5se6frawcegdjakljxp.py
# Topologically Sorted Source Nodes: [tp, tp_1, tp_2, tp_3, mul_3, mul_4, sub, fn, fn_1, fn_2, mul_5, add, sub_1, fp, fp_1, fp_2, add_1, add_2, truediv, result, result_1], Original ATen: [aten.mul, aten.sum, aten.add, aten.rsub, aten.div, aten.mean]
# Source node to ATen node mapping:
# add => add_1
# add_1 => add_2
# add_2 => add_3
# fn => mul_1
# fn_1 => sum_2
# fn_2 => sum_5
# fp => mul_2
# fp_1 => sum_3
# fp_2 => sum_6
# mul_3 => mul_3
# mul_4 => mul_4
# mul_5 => mul_5
# result => sub_2
# result_1 => mean
# sub => sub
# sub_1 => sub_1
# tp => mul
# tp_1 => sum_1
# tp_2 => sum_4
# tp_3 => add
# truediv => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, %arg0_1), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [-1]), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_1, [-1]), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_4, 0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %arg0_1), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [-1]), kwargs = {})
# %sum_5 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_2, [-1]), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_5, 1.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %mul_5), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg0_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, %sub_1), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_2, [-1]), kwargs = {})
# %sum_6 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_3, [-1]), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, %sum_6), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_3, %add_3), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %div), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sub_2,), kwargs = {})
triton_per_fused_add_div_mean_mul_rsub_sum_0 = async_compile.triton('triton_per_fused_add_div_mean_mul_rsub_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 16],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mean_mul_rsub_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 32, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_sum_0(in_ptr0, in_ptr1, out_ptr3, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (16*r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (16*r0), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (16*r0)), None, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + (16*r0)), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (16*r0)), None, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + (16*r0)), None, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (16*r0)), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (3 + (16*r0)), None, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (4 + (16*r0)), None, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (4 + (16*r0)), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (5 + (16*r0)), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (5 + (16*r0)), None, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr0 + (6 + (16*r0)), None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + (6 + (16*r0)), None, eviction_policy='evict_last')
tmp26 = tl.load(in_ptr0 + (7 + (16*r0)), None, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr1 + (7 + (16*r0)), None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr0 + (8 + (16*r0)), None, eviction_policy='evict_last')
tmp32 = tl.load(in_ptr1 + (8 + (16*r0)), None, eviction_policy='evict_last')
tmp34 = tl.load(in_ptr0 + (9 + (16*r0)), None, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr1 + (9 + (16*r0)), None, eviction_policy='evict_last')
tmp38 = tl.load(in_ptr0 + (10 + (16*r0)), None, eviction_policy='evict_last')
tmp39 = tl.load(in_ptr1 + (10 + (16*r0)), None, eviction_policy='evict_last')
tmp42 = tl.load(in_ptr0 + (11 + (16*r0)), None, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr1 + (11 + (16*r0)), None, eviction_policy='evict_last')
tmp47 = tl.load(in_ptr0 + (12 + (16*r0)), None, eviction_policy='evict_last')
tmp48 = tl.load(in_ptr1 + (12 + (16*r0)), None, eviction_policy='evict_last')
tmp50 = tl.load(in_ptr0 + (13 + (16*r0)), None, eviction_policy='evict_last')
tmp51 = tl.load(in_ptr1 + (13 + (16*r0)), None, eviction_policy='evict_last')
tmp54 = tl.load(in_ptr0 + (14 + (16*r0)), None, eviction_policy='evict_last')
tmp55 = tl.load(in_ptr1 + (14 + (16*r0)), None, eviction_policy='evict_last')
tmp58 = tl.load(in_ptr0 + (15 + (16*r0)), None, eviction_policy='evict_last')
tmp59 = tl.load(in_ptr1 + (15 + (16*r0)), None, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tmp17 = tmp15 * tmp16
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tmp24 = tmp22 * tmp23
tmp25 = tmp21 + tmp24
tmp28 = tmp26 * tmp27
tmp29 = tmp25 + tmp28
tmp30 = tmp14 + tmp29
tmp33 = tmp31 * tmp32
tmp36 = tmp34 * tmp35
tmp37 = tmp33 + tmp36
tmp40 = tmp38 * tmp39
tmp41 = tmp37 + tmp40
tmp44 = tmp42 * tmp43
tmp45 = tmp41 + tmp44
tmp46 = tmp30 + tmp45
tmp49 = tmp47 * tmp48
tmp52 = tmp50 * tmp51
tmp53 = tmp49 + tmp52
tmp56 = tmp54 * tmp55
tmp57 = tmp53 + tmp56
tmp60 = tmp58 * tmp59
tmp61 = tmp57 + tmp60
tmp62 = tmp46 + tmp61
tmp63 = 1.0
tmp64 = tmp63 - tmp0
tmp65 = tmp64 * tmp1
tmp66 = tmp63 - tmp3
tmp67 = tmp66 * tmp4
tmp68 = tmp65 + tmp67
tmp69 = tmp63 - tmp7
tmp70 = tmp69 * tmp8
tmp71 = tmp68 + tmp70
tmp72 = tmp63 - tmp11
tmp73 = tmp72 * tmp12
tmp74 = tmp71 + tmp73
tmp75 = tmp63 - tmp15
tmp76 = tmp75 * tmp16
tmp77 = tmp63 - tmp18
tmp78 = tmp77 * tmp19
tmp79 = tmp76 + tmp78
tmp80 = tmp63 - tmp22
tmp81 = tmp80 * tmp23
tmp82 = tmp79 + tmp81
tmp83 = tmp63 - tmp26
tmp84 = tmp83 * tmp27
tmp85 = tmp82 + tmp84
tmp86 = tmp74 + tmp85
tmp87 = tmp63 - tmp31
tmp88 = tmp87 * tmp32
tmp89 = tmp63 - tmp34
tmp90 = tmp89 * tmp35
tmp91 = tmp88 + tmp90
tmp92 = tmp63 - tmp38
tmp93 = tmp92 * tmp39
tmp94 = tmp91 + tmp93
tmp95 = tmp63 - tmp42
tmp96 = tmp95 * tmp43
tmp97 = tmp94 + tmp96
tmp98 = tmp86 + tmp97
tmp99 = tmp63 - tmp47
tmp100 = tmp99 * tmp48
tmp101 = tmp63 - tmp50
tmp102 = tmp101 * tmp51
tmp103 = tmp100 + tmp102
tmp104 = tmp63 - tmp54
tmp105 = tmp104 * tmp55
tmp106 = tmp103 + tmp105
tmp107 = tmp63 - tmp58
tmp108 = tmp107 * tmp59
tmp109 = tmp106 + tmp108
tmp110 = tmp98 + tmp109
tmp111 = tmp63 - tmp1
tmp112 = tmp0 * tmp111
tmp113 = tmp63 - tmp4
tmp114 = tmp3 * tmp113
tmp115 = tmp112 + tmp114
tmp116 = tmp63 - tmp8
tmp117 = tmp7 * tmp116
tmp118 = tmp115 + tmp117
tmp119 = tmp63 - tmp12
tmp120 = tmp11 * tmp119
tmp121 = tmp118 + tmp120
tmp122 = tmp63 - tmp16
tmp123 = tmp15 * tmp122
tmp124 = tmp63 - tmp19
tmp125 = tmp18 * tmp124
tmp126 = tmp123 + tmp125
tmp127 = tmp63 - tmp23
tmp128 = tmp22 * tmp127
tmp129 = tmp126 + tmp128
tmp130 = tmp63 - tmp27
tmp131 = tmp26 * tmp130
tmp132 = tmp129 + tmp131
tmp133 = tmp121 + tmp132
tmp134 = tmp63 - tmp32
tmp135 = tmp31 * tmp134
tmp136 = tmp63 - tmp35
tmp137 = tmp34 * tmp136
tmp138 = tmp135 + tmp137
tmp139 = tmp63 - tmp39
tmp140 = tmp38 * tmp139
tmp141 = tmp138 + tmp140
tmp142 = tmp63 - tmp43
tmp143 = tmp42 * tmp142
tmp144 = tmp141 + tmp143
tmp145 = tmp133 + tmp144
tmp146 = tmp63 - tmp48
tmp147 = tmp47 * tmp146
tmp148 = tmp63 - tmp51
tmp149 = tmp50 * tmp148
tmp150 = tmp147 + tmp149
tmp151 = tmp63 - tmp55
tmp152 = tmp54 * tmp151
tmp153 = tmp150 + tmp152
tmp154 = tmp63 - tmp59
tmp155 = tmp58 * tmp154
tmp156 = tmp153 + tmp155
tmp157 = tmp145 + tmp156
tmp158 = 0.0
tmp159 = tmp62 + tmp158
tmp160 = 2.0
tmp161 = tmp159 * tmp160
tmp162 = tmp110 * tmp63
tmp163 = tmp161 + tmp162
tmp164 = tmp163 + tmp157
tmp165 = 1e-06
tmp166 = tmp164 + tmp165
tmp167 = tmp161 / tmp166
tmp168 = tmp63 - tmp167
tmp169 = tl.broadcast_to(tmp168, [XBLOCK, RBLOCK])
tmp171 = tl.sum(tmp169, 1)[:, None]
tl.store(out_ptr3 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp171, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ji/cjiatqqtn7v7exy34pj4bpfh6qxtkb2r465o4j2gx3br3ej5xjgr.py
# Topologically Sorted Source Nodes: [tp_3, mul_3, mul_4, mul_5, add, add_1, add_2, truediv, result, result_1, mul_6, target_1, ce, mul_7, add_4], Original ATen: [aten.add, aten.mul, aten.div, aten.rsub, aten.mean, aten.argmax, aten.nll_loss2d_forward]
# Source node to ATen node mapping:
# add => add_1
# add_1 => add_2
# add_2 => add_3
# add_4 => add_5
# ce => convert_element_type, div_1, full_default_1, ne_1, ne_2, neg, sum_7, sum_8, where_1
# mul_3 => mul_3
# mul_4 => mul_4
# mul_5 => mul_5
# mul_6 => mul_6
# mul_7 => mul_7
# result => sub_2
# result_1 => mean
# target_1 => argmax
# tp_3 => add
# truediv => div
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_4, 0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_5, 1.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %mul_5), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, %sum_6), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_3, %add_3), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %div), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sub_2,), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mean, 0.5), kwargs = {})
# %argmax : [num_users=4] = call_function[target=torch.ops.aten.argmax.default](args = (%arg0_1, 1), kwargs = {})
# %ne_1 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%argmax, -100), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%squeeze,), kwargs = {})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_1 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%ne_1, %neg, %full_default_1), kwargs = {})
# %sum_8 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%where_1,), kwargs = {})
# %ne_2 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%argmax, -100), kwargs = {})
# %sum_7 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%ne_2,), kwargs = {})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%sum_7, torch.float32), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_8, %convert_element_type), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_1, 0.5), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_6, %mul_7), kwargs = {})
triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1 = async_compile.triton('triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = (rindex // 16)
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0 + (64*r1)), None)
tmp1 = tl.load(in_ptr0 + (16 + r0 + (64*r1)), None)
tmp17 = tl.load(in_ptr0 + (32 + r0 + (64*r1)), None)
tmp32 = tl.load(in_ptr0 + (48 + r0 + (64*r1)), None)
tmp69 = tl.load(in_out_ptr0 + (0))
tmp70 = tl.broadcast_to(tmp69, [XBLOCK, 1])
tmp2 = tmp0 > tmp1
tmp3 = tmp0 == tmp1
tmp4 = tmp0 != tmp0
tmp5 = tmp1 != tmp1
tmp6 = tmp4 > tmp5
tmp7 = tmp2 | tmp6
tmp8 = tmp4 & tmp5
tmp9 = tmp3 | tmp8
tmp10 = tl.full([1, 1], 0, tl.int64)
tmp11 = tl.full([1, 1], 1, tl.int64)
tmp12 = tmp10 < tmp11
tmp13 = tmp9 & tmp12
tmp14 = tmp7 | tmp13
tmp15 = tl.where(tmp14, tmp0, tmp1)
tmp16 = tl.where(tmp14, tmp10, tmp11)
tmp18 = tmp15 > tmp17
tmp19 = tmp15 == tmp17
tmp20 = tmp15 != tmp15
tmp21 = tmp17 != tmp17
tmp22 = tmp20 > tmp21
tmp23 = tmp18 | tmp22
tmp24 = tmp20 & tmp21
tmp25 = tmp19 | tmp24
tmp26 = tl.full([1, 1], 2, tl.int64)
tmp27 = tmp16 < tmp26
tmp28 = tmp25 & tmp27
tmp29 = tmp23 | tmp28
tmp30 = tl.where(tmp29, tmp15, tmp17)
tmp31 = tl.where(tmp29, tmp16, tmp26)
tmp33 = tmp30 > tmp32
tmp34 = tmp30 == tmp32
tmp35 = tmp30 != tmp30
tmp36 = tmp32 != tmp32
tmp37 = tmp35 > tmp36
tmp38 = tmp33 | tmp37
tmp39 = tmp35 & tmp36
tmp40 = tmp34 | tmp39
tmp41 = tl.full([1, 1], 3, tl.int64)
tmp42 = tmp31 < tmp41
tmp43 = tmp40 & tmp42
tmp44 = tmp38 | tmp43
tmp45 = tl.where(tmp44, tmp30, tmp32)
tmp46 = tl.where(tmp44, tmp31, tmp41)
tmp47 = tl.full([1, 1], -100, tl.int64)
tmp48 = tmp46 != tmp47
tmp49 = tl.where(tmp48, tmp46, tmp10)
tmp50 = tl.full([XBLOCK, RBLOCK], 4, tl.int32)
tmp51 = tmp49 + tmp50
tmp52 = tmp49 < 0
tmp53 = tl.where(tmp52, tmp51, tmp49)
tl.device_assert((0 <= tmp53) & (tmp53 < 4), "index out of bounds: 0 <= tmp53 < 4")
tmp55 = tl.load(in_ptr1 + (r0 + (16*tmp53) + (64*r1)), None)
tmp56 = 1e-06
tmp57 = tmp55 + tmp56
tmp58 = tl_math.log(tmp57)
tmp59 = -tmp58
tmp60 = 0.0
tmp61 = tl.where(tmp48, tmp59, tmp60)
tmp62 = tl.broadcast_to(tmp61, [XBLOCK, RBLOCK])
tmp64 = tl.sum(tmp62, 1)[:, None]
tmp65 = tmp48.to(tl.int64)
tmp66 = tl.broadcast_to(tmp65, [XBLOCK, RBLOCK])
tmp68 = tl.sum(tmp66, 1)[:, None]
tmp71 = 16.0
tmp72 = tmp70 / tmp71
tmp73 = 0.5
tmp74 = tmp72 * tmp73
tmp75 = tmp68.to(tl.float32)
tmp76 = tmp64 / tmp75
tmp77 = tmp76 * tmp73
tmp78 = tmp74 + tmp77
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp78, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf3 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [tp, tp_1, tp_2, tp_3, mul_3, mul_4, sub, fn, fn_1, fn_2, mul_5, add, sub_1, fp, fp_1, fp_2, add_1, add_2, truediv, result, result_1], Original ATen: [aten.mul, aten.sum, aten.add, aten.rsub, aten.div, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_add_div_mean_mul_rsub_sum_0.run(arg1_1, arg0_1, buf3, 1, 16, grid=grid(1), stream=stream0)
buf7 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [tp_3, mul_3, mul_4, mul_5, add, add_1, add_2, truediv, result, result_1, mul_6, target_1, ce, mul_7, add_4], Original ATen: [aten.add, aten.mul, aten.div, aten.rsub, aten.mean, aten.argmax, aten.nll_loss2d_forward]
triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1.run(buf7, arg0_1, arg1_1, 1, 64, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf7, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
class FScoreLoss(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super(FScoreLoss, self).__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
assert not target.requires_grad
if self.make_onehot:
target = make_onehot_segmentation(target, range(input_.shape[1]
), newaxis=self.make_onehot_newaxis)
target = target.float()
tp = input_ * target
fn = (1 - input_) * target
fp = input_ * (1 - target)
while tp.dim() > 2:
tp = tp.sum(-1, keepdim=False)
fn = fn.sum(-1, keepdim=False)
fp = fp.sum(-1, keepdim=False)
tp += self.tp_bias
result = 1 - (1 + self.beta * self.beta) * tp / ((1 + self.beta *
self.beta) * tp + self.beta * self.beta * fn + fp + self.eps)
if self.reduction != 'none':
if self.weight is not None:
self.weight = self.weight.float()
self.weight = self.weight
weight = self.weight.expand_as(result)
result = result * weight
if self.ignore_index is not None:
for cls in sorted(self.ignore_index, reverse=True):
if cls == result.shape[1] - 1:
result = result[:, :-1]
elif cls == 0:
result = result[:, 1:]
else:
result = torch.cat([result[:, :cls], result[:, cls +
1:]], 1)
if self.reduction == 'mean':
result = torch.mean(result)
elif self.reduction == 'sum':
result = torch.sum(result)
else:
raise ValueError(
"reduction must be 'none', 'mean' or 'sum', but is {}".
format(self.reduction))
return result
class DiceLoss(FScoreLoss):
"""FScoreLoss with beta=1."""
def __init__(self, **kwargs):
kwargs['beta'] = 1.0
super(DiceLoss, self).__init__(**kwargs)
def forward(self, input_, target):
return super(DiceLoss, self).forward(input_, target)
class CrossEntropyDiceLoss(DiceLoss):
"""Weighted sum of CE and Dice losses.
Expects softmax inputs!
Args:
ce_weight: The weight for the CE loss. Weight for the Dice loss will be
1 - ce_weight.
"""
def __init__(self, ce_weight: 'float'=0.5, *args, **kwargs):
super().__init__(*args, **kwargs)
self.ce_weight = ce_weight
self.ce = nn.NLLLoss(weight=kwargs.get('weight', None))
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
dice = super().forward(input_, target)
if self.make_onehot:
if self.make_onehot_newaxis:
pass
else:
target = target.squeeze()
else:
target = torch.argmax(target, 1, keepdim=False)
target = target.long()
ce = self.ce(torch.log(input_ + self.eps), target)
return (1 - self.ce_weight) * dice + self.ce_weight * ce
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_sum_0(in_ptr0, in_ptr1, out_ptr3,
xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + 16 * r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 16 * r0, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 16 * r0), None, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 16 * r0), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 16 * r0), None, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + 16 * r0), None, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 16 * r0), None, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (3 + 16 * r0), None, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr0 + (4 + 16 * r0), None, eviction_policy='evict_last'
)
tmp16 = tl.load(in_ptr1 + (4 + 16 * r0), None, eviction_policy='evict_last'
)
tmp18 = tl.load(in_ptr0 + (5 + 16 * r0), None, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr1 + (5 + 16 * r0), None, eviction_policy='evict_last'
)
tmp22 = tl.load(in_ptr0 + (6 + 16 * r0), None, eviction_policy='evict_last'
)
tmp23 = tl.load(in_ptr1 + (6 + 16 * r0), None, eviction_policy='evict_last'
)
tmp26 = tl.load(in_ptr0 + (7 + 16 * r0), None, eviction_policy='evict_last'
)
tmp27 = tl.load(in_ptr1 + (7 + 16 * r0), None, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr0 + (8 + 16 * r0), None, eviction_policy='evict_last'
)
tmp32 = tl.load(in_ptr1 + (8 + 16 * r0), None, eviction_policy='evict_last'
)
tmp34 = tl.load(in_ptr0 + (9 + 16 * r0), None, eviction_policy='evict_last'
)
tmp35 = tl.load(in_ptr1 + (9 + 16 * r0), None, eviction_policy='evict_last'
)
tmp38 = tl.load(in_ptr0 + (10 + 16 * r0), None, eviction_policy=
'evict_last')
tmp39 = tl.load(in_ptr1 + (10 + 16 * r0), None, eviction_policy=
'evict_last')
tmp42 = tl.load(in_ptr0 + (11 + 16 * r0), None, eviction_policy=
'evict_last')
tmp43 = tl.load(in_ptr1 + (11 + 16 * r0), None, eviction_policy=
'evict_last')
tmp47 = tl.load(in_ptr0 + (12 + 16 * r0), None, eviction_policy=
'evict_last')
tmp48 = tl.load(in_ptr1 + (12 + 16 * r0), None, eviction_policy=
'evict_last')
tmp50 = tl.load(in_ptr0 + (13 + 16 * r0), None, eviction_policy=
'evict_last')
tmp51 = tl.load(in_ptr1 + (13 + 16 * r0), None, eviction_policy=
'evict_last')
tmp54 = tl.load(in_ptr0 + (14 + 16 * r0), None, eviction_policy=
'evict_last')
tmp55 = tl.load(in_ptr1 + (14 + 16 * r0), None, eviction_policy=
'evict_last')
tmp58 = tl.load(in_ptr0 + (15 + 16 * r0), None, eviction_policy=
'evict_last')
tmp59 = tl.load(in_ptr1 + (15 + 16 * r0), None, eviction_policy=
'evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tmp17 = tmp15 * tmp16
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tmp24 = tmp22 * tmp23
tmp25 = tmp21 + tmp24
tmp28 = tmp26 * tmp27
tmp29 = tmp25 + tmp28
tmp30 = tmp14 + tmp29
tmp33 = tmp31 * tmp32
tmp36 = tmp34 * tmp35
tmp37 = tmp33 + tmp36
tmp40 = tmp38 * tmp39
tmp41 = tmp37 + tmp40
tmp44 = tmp42 * tmp43
tmp45 = tmp41 + tmp44
tmp46 = tmp30 + tmp45
tmp49 = tmp47 * tmp48
tmp52 = tmp50 * tmp51
tmp53 = tmp49 + tmp52
tmp56 = tmp54 * tmp55
tmp57 = tmp53 + tmp56
tmp60 = tmp58 * tmp59
tmp61 = tmp57 + tmp60
tmp62 = tmp46 + tmp61
tmp63 = 1.0
tmp64 = tmp63 - tmp0
tmp65 = tmp64 * tmp1
tmp66 = tmp63 - tmp3
tmp67 = tmp66 * tmp4
tmp68 = tmp65 + tmp67
tmp69 = tmp63 - tmp7
tmp70 = tmp69 * tmp8
tmp71 = tmp68 + tmp70
tmp72 = tmp63 - tmp11
tmp73 = tmp72 * tmp12
tmp74 = tmp71 + tmp73
tmp75 = tmp63 - tmp15
tmp76 = tmp75 * tmp16
tmp77 = tmp63 - tmp18
tmp78 = tmp77 * tmp19
tmp79 = tmp76 + tmp78
tmp80 = tmp63 - tmp22
tmp81 = tmp80 * tmp23
tmp82 = tmp79 + tmp81
tmp83 = tmp63 - tmp26
tmp84 = tmp83 * tmp27
tmp85 = tmp82 + tmp84
tmp86 = tmp74 + tmp85
tmp87 = tmp63 - tmp31
tmp88 = tmp87 * tmp32
tmp89 = tmp63 - tmp34
tmp90 = tmp89 * tmp35
tmp91 = tmp88 + tmp90
tmp92 = tmp63 - tmp38
tmp93 = tmp92 * tmp39
tmp94 = tmp91 + tmp93
tmp95 = tmp63 - tmp42
tmp96 = tmp95 * tmp43
tmp97 = tmp94 + tmp96
tmp98 = tmp86 + tmp97
tmp99 = tmp63 - tmp47
tmp100 = tmp99 * tmp48
tmp101 = tmp63 - tmp50
tmp102 = tmp101 * tmp51
tmp103 = tmp100 + tmp102
tmp104 = tmp63 - tmp54
tmp105 = tmp104 * tmp55
tmp106 = tmp103 + tmp105
tmp107 = tmp63 - tmp58
tmp108 = tmp107 * tmp59
tmp109 = tmp106 + tmp108
tmp110 = tmp98 + tmp109
tmp111 = tmp63 - tmp1
tmp112 = tmp0 * tmp111
tmp113 = tmp63 - tmp4
tmp114 = tmp3 * tmp113
tmp115 = tmp112 + tmp114
tmp116 = tmp63 - tmp8
tmp117 = tmp7 * tmp116
tmp118 = tmp115 + tmp117
tmp119 = tmp63 - tmp12
tmp120 = tmp11 * tmp119
tmp121 = tmp118 + tmp120
tmp122 = tmp63 - tmp16
tmp123 = tmp15 * tmp122
tmp124 = tmp63 - tmp19
tmp125 = tmp18 * tmp124
tmp126 = tmp123 + tmp125
tmp127 = tmp63 - tmp23
tmp128 = tmp22 * tmp127
tmp129 = tmp126 + tmp128
tmp130 = tmp63 - tmp27
tmp131 = tmp26 * tmp130
tmp132 = tmp129 + tmp131
tmp133 = tmp121 + tmp132
tmp134 = tmp63 - tmp32
tmp135 = tmp31 * tmp134
tmp136 = tmp63 - tmp35
tmp137 = tmp34 * tmp136
tmp138 = tmp135 + tmp137
tmp139 = tmp63 - tmp39
tmp140 = tmp38 * tmp139
tmp141 = tmp138 + tmp140
tmp142 = tmp63 - tmp43
tmp143 = tmp42 * tmp142
tmp144 = tmp141 + tmp143
tmp145 = tmp133 + tmp144
tmp146 = tmp63 - tmp48
tmp147 = tmp47 * tmp146
tmp148 = tmp63 - tmp51
tmp149 = tmp50 * tmp148
tmp150 = tmp147 + tmp149
tmp151 = tmp63 - tmp55
tmp152 = tmp54 * tmp151
tmp153 = tmp150 + tmp152
tmp154 = tmp63 - tmp59
tmp155 = tmp58 * tmp154
tmp156 = tmp153 + tmp155
tmp157 = tmp145 + tmp156
tmp158 = 0.0
tmp159 = tmp62 + tmp158
tmp160 = 2.0
tmp161 = tmp159 * tmp160
tmp162 = tmp110 * tmp63
tmp163 = tmp161 + tmp162
tmp164 = tmp163 + tmp157
tmp165 = 1e-06
tmp166 = tmp164 + tmp165
tmp167 = tmp161 / tmp166
tmp168 = tmp63 - tmp167
tmp169 = tl.broadcast_to(tmp168, [XBLOCK, RBLOCK])
tmp171 = tl.sum(tmp169, 1)[:, None]
tl.store(out_ptr3 + tl.full([XBLOCK, 1], 0, tl.int32), tmp171, None)
@triton.jit
def triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1(
in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp1 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp17 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp32 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp69 = tl.load(in_out_ptr0 + 0)
tmp70 = tl.broadcast_to(tmp69, [XBLOCK, 1])
tmp2 = tmp0 > tmp1
tmp3 = tmp0 == tmp1
tmp4 = tmp0 != tmp0
tmp5 = tmp1 != tmp1
tmp6 = tmp4 > tmp5
tmp7 = tmp2 | tmp6
tmp8 = tmp4 & tmp5
tmp9 = tmp3 | tmp8
tmp10 = tl.full([1, 1], 0, tl.int64)
tmp11 = tl.full([1, 1], 1, tl.int64)
tmp12 = tmp10 < tmp11
tmp13 = tmp9 & tmp12
tmp14 = tmp7 | tmp13
tmp15 = tl.where(tmp14, tmp0, tmp1)
tmp16 = tl.where(tmp14, tmp10, tmp11)
tmp18 = tmp15 > tmp17
tmp19 = tmp15 == tmp17
tmp20 = tmp15 != tmp15
tmp21 = tmp17 != tmp17
tmp22 = tmp20 > tmp21
tmp23 = tmp18 | tmp22
tmp24 = tmp20 & tmp21
tmp25 = tmp19 | tmp24
tmp26 = tl.full([1, 1], 2, tl.int64)
tmp27 = tmp16 < tmp26
tmp28 = tmp25 & tmp27
tmp29 = tmp23 | tmp28
tmp30 = tl.where(tmp29, tmp15, tmp17)
tmp31 = tl.where(tmp29, tmp16, tmp26)
tmp33 = tmp30 > tmp32
tmp34 = tmp30 == tmp32
tmp35 = tmp30 != tmp30
tmp36 = tmp32 != tmp32
tmp37 = tmp35 > tmp36
tmp38 = tmp33 | tmp37
tmp39 = tmp35 & tmp36
tmp40 = tmp34 | tmp39
tmp41 = tl.full([1, 1], 3, tl.int64)
tmp42 = tmp31 < tmp41
tmp43 = tmp40 & tmp42
tmp44 = tmp38 | tmp43
tl.where(tmp44, tmp30, tmp32)
tmp46 = tl.where(tmp44, tmp31, tmp41)
tmp47 = tl.full([1, 1], -100, tl.int64)
tmp48 = tmp46 != tmp47
tmp49 = tl.where(tmp48, tmp46, tmp10)
tmp50 = tl.full([XBLOCK, RBLOCK], 4, tl.int32)
tmp51 = tmp49 + tmp50
tmp52 = tmp49 < 0
tmp53 = tl.where(tmp52, tmp51, tmp49)
tl.device_assert((0 <= tmp53) & (tmp53 < 4),
'index out of bounds: 0 <= tmp53 < 4')
tmp55 = tl.load(in_ptr1 + (r0 + 16 * tmp53 + 64 * r1), None)
tmp56 = 1e-06
tmp57 = tmp55 + tmp56
tmp58 = tl_math.log(tmp57)
tmp59 = -tmp58
tmp60 = 0.0
tmp61 = tl.where(tmp48, tmp59, tmp60)
tmp62 = tl.broadcast_to(tmp61, [XBLOCK, RBLOCK])
tmp64 = tl.sum(tmp62, 1)[:, None]
tmp65 = tmp48.to(tl.int64)
tmp66 = tl.broadcast_to(tmp65, [XBLOCK, RBLOCK])
tmp68 = tl.sum(tmp66, 1)[:, None]
tmp71 = 16.0
tmp72 = tmp70 / tmp71
tmp73 = 0.5
tmp74 = tmp72 * tmp73
tmp75 = tmp68.to(tl.float32)
tmp76 = tmp64 / tmp75
tmp77 = tmp76 * tmp73
tmp78 = tmp74 + tmp77
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp78, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf3 = empty_strided_cuda((), (), torch.float32)
get_raw_stream(0)
triton_per_fused_add_div_mean_mul_rsub_sum_0[grid(1)](arg1_1,
arg0_1, buf3, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
buf7 = buf3
del buf3
triton_per_fused_add_argmax_div_mean_mul_nll_loss2d_forward_rsub_1[grid
(1)](buf7, arg0_1, arg1_1, 1, 64, XBLOCK=1, num_warps=2,
num_stages=1)
del arg0_1
del arg1_1
return buf7,
class FScoreLoss(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super(FScoreLoss, self).__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
assert not target.requires_grad
if self.make_onehot:
target = make_onehot_segmentation(target, range(input_.shape[1]
), newaxis=self.make_onehot_newaxis)
target = target.float()
tp = input_ * target
fn = (1 - input_) * target
fp = input_ * (1 - target)
while tp.dim() > 2:
tp = tp.sum(-1, keepdim=False)
fn = fn.sum(-1, keepdim=False)
fp = fp.sum(-1, keepdim=False)
tp += self.tp_bias
result = 1 - (1 + self.beta * self.beta) * tp / ((1 + self.beta *
self.beta) * tp + self.beta * self.beta * fn + fp + self.eps)
if self.reduction != 'none':
if self.weight is not None:
self.weight = self.weight.float()
self.weight = self.weight
weight = self.weight.expand_as(result)
result = result * weight
if self.ignore_index is not None:
for cls in sorted(self.ignore_index, reverse=True):
if cls == result.shape[1] - 1:
result = result[:, :-1]
elif cls == 0:
result = result[:, 1:]
else:
result = torch.cat([result[:, :cls], result[:, cls +
1:]], 1)
if self.reduction == 'mean':
result = torch.mean(result)
elif self.reduction == 'sum':
result = torch.sum(result)
else:
raise ValueError(
"reduction must be 'none', 'mean' or 'sum', but is {}".
format(self.reduction))
return result
class DiceLoss(FScoreLoss):
"""FScoreLoss with beta=1."""
def __init__(self, **kwargs):
kwargs['beta'] = 1.0
super(DiceLoss, self).__init__(**kwargs)
def forward(self, input_, target):
return super(DiceLoss, self).forward(input_, target)
class CrossEntropyDiceLossNew(DiceLoss):
"""Weighted sum of CE and Dice losses.
Expects softmax inputs!
Args:
ce_weight: The weight for the CE loss. Weight for the Dice loss will be
1 - ce_weight.
"""
def __init__(self, ce_weight: 'float'=0.5, *args, **kwargs):
super().__init__(*args, **kwargs)
self.ce_weight = ce_weight
self.ce = nn.NLLLoss(weight=kwargs.get('weight', None))
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MIC-DKFZ/image-time-series | CrossEntropyDiceLoss | false | 5,574 | [
"MIT"
] | 1 | 0480d5cb6936c7d9e839b6741f18c10893d78d8a | https://github.com/MIC-DKFZ/image-time-series/tree/0480d5cb6936c7d9e839b6741f18c10893d78d8a | import torch
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
class FScoreLoss(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super().__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
assert not target.requires_grad
if self.make_onehot:
target = make_onehot_segmentation(target, range(input_.shape[1]
), newaxis=self.make_onehot_newaxis)
target = target.float()
tp = input_ * target
fn = (1 - input_) * target
fp = input_ * (1 - target)
while tp.dim() > 2:
tp = tp.sum(-1, keepdim=False)
fn = fn.sum(-1, keepdim=False)
fp = fp.sum(-1, keepdim=False)
tp += self.tp_bias
result = 1 - (1 + self.beta * self.beta) * tp / ((1 + self.beta *
self.beta) * tp + self.beta * self.beta * fn + fp + self.eps)
if self.reduction != 'none':
if self.weight is not None:
self.weight = self.weight.float()
self.weight = self.weight
weight = self.weight.expand_as(result)
result = result * weight
if self.ignore_index is not None:
for cls in sorted(self.ignore_index, reverse=True):
if cls == result.shape[1] - 1:
result = result[:, :-1]
elif cls == 0:
result = result[:, 1:]
else:
result = torch.cat([result[:, :cls], result[:, cls +
1:]], 1)
if self.reduction == 'mean':
result = torch.mean(result)
elif self.reduction == 'sum':
result = torch.sum(result)
else:
raise ValueError(
"reduction must be 'none', 'mean' or 'sum', but is {}".
format(self.reduction))
return result
class DiceLoss(FScoreLoss):
"""FScoreLoss with beta=1."""
def __init__(self, **
# ... truncated (>4000 chars) for memory efficiency |
AvgPoolPad | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/pr/cprzlfpjjqlj6tudvbc455jxno35xlnta4wgmkbc6uo5zmcxii4s.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.constant_pad_nd, aten.avg_pool2d]
# Source node to ATen node mapping:
# x => constant_pad_nd
# x_1 => avg_pool2d
# Graph fragment:
# %constant_pad_nd : [num_users=1] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%arg0_1, [1, 0, 1, 0], 0.0), kwargs = {})
# %avg_pool2d : [num_users=1] = call_function[target=torch.ops.aten.avg_pool2d.default](args = (%constant_pad_nd, [3, 3], [2, 2], [1, 1], False, False), kwargs = {})
triton_poi_fused_avg_pool2d_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_avg_pool2d_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_avg_pool2d_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 9, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_avg_pool2d_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 3) % 3
x0 = xindex % 3
x2 = (xindex // 9)
x4 = xindex
tmp0 = (-1) + (2*x1)
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 5, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = (-1) + (2*x0)
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = (-2) + (2*x1)
tmp12 = tmp11 >= tmp1
tmp13 = (-2) + (2*x0)
tmp14 = tmp13 >= tmp1
tmp15 = tmp12 & tmp14
tmp16 = tmp15 & tmp10
tmp17 = tl.load(in_ptr0 + ((-10) + (2*x0) + (8*x1) + (16*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.full(tmp17.shape, 0.0, tmp17.dtype)
tmp19 = tl.where(tmp10, tmp17, tmp18)
tmp20 = 2*x0
tmp21 = tmp20 >= tmp1
tmp22 = tmp20 < tmp3
tmp23 = tmp21 & tmp22
tmp24 = tmp5 & tmp23
tmp25 = tmp12 & tmp7
tmp26 = tmp25 & tmp24
tmp27 = tl.load(in_ptr0 + ((-9) + (2*x0) + (8*x1) + (16*x2)), tmp26 & xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tl.full(tmp27.shape, 0.0, tmp27.dtype)
tmp29 = tl.where(tmp24, tmp27, tmp28)
tmp30 = tmp29 + tmp19
tmp31 = 1 + (2*x0)
tmp32 = tmp31 >= tmp1
tmp33 = tmp31 < tmp3
tmp34 = tmp32 & tmp33
tmp35 = tmp5 & tmp34
tmp36 = tmp12 & tmp21
tmp37 = tmp36 & tmp35
tmp38 = tl.load(in_ptr0 + ((-8) + (2*x0) + (8*x1) + (16*x2)), tmp37 & xmask, eviction_policy='evict_last', other=0.0)
tmp39 = tl.full(tmp38.shape, 0.0, tmp38.dtype)
tmp40 = tl.where(tmp35, tmp38, tmp39)
tmp41 = tmp40 + tmp30
tmp42 = 2*x1
tmp43 = tmp42 >= tmp1
tmp44 = tmp42 < tmp3
tmp45 = tmp43 & tmp44
tmp46 = tmp45 & tmp9
tmp47 = tmp2 & tmp14
tmp48 = tmp47 & tmp46
tmp49 = tl.load(in_ptr0 + ((-6) + (2*x0) + (8*x1) + (16*x2)), tmp48 & xmask, eviction_policy='evict_last', other=0.0)
tmp50 = tl.full(tmp49.shape, 0.0, tmp49.dtype)
tmp51 = tl.where(tmp46, tmp49, tmp50)
tmp52 = tmp51 + tmp41
tmp53 = tmp45 & tmp23
tmp54 = tmp2 & tmp7
tmp55 = tmp54 & tmp53
tmp56 = tl.load(in_ptr0 + ((-5) + (2*x0) + (8*x1) + (16*x2)), tmp55 & xmask, eviction_policy='evict_last', other=0.0)
tmp57 = tl.full(tmp56.shape, 0.0, tmp56.dtype)
tmp58 = tl.where(tmp53, tmp56, tmp57)
tmp59 = tmp58 + tmp52
tmp60 = tmp45 & tmp34
tmp61 = tmp2 & tmp21
tmp62 = tmp61 & tmp60
tmp63 = tl.load(in_ptr0 + ((-4) + (2*x0) + (8*x1) + (16*x2)), tmp62 & xmask, eviction_policy='evict_last', other=0.0)
tmp64 = tl.full(tmp63.shape, 0.0, tmp63.dtype)
tmp65 = tl.where(tmp60, tmp63, tmp64)
tmp66 = tmp65 + tmp59
tmp67 = 1 + (2*x1)
tmp68 = tmp67 >= tmp1
tmp69 = tmp67 < tmp3
tmp70 = tmp68 & tmp69
tmp71 = tmp70 & tmp9
tmp72 = tmp43 & tmp14
tmp73 = tmp72 & tmp71
tmp74 = tl.load(in_ptr0 + ((-2) + (2*x0) + (8*x1) + (16*x2)), tmp73 & xmask, eviction_policy='evict_last', other=0.0)
tmp75 = tl.full(tmp74.shape, 0.0, tmp74.dtype)
tmp76 = tl.where(tmp71, tmp74, tmp75)
tmp77 = tmp76 + tmp66
tmp78 = tmp70 & tmp23
tmp79 = tmp43 & tmp7
tmp80 = tmp79 & tmp78
tmp81 = tl.load(in_ptr0 + ((-1) + (2*x0) + (8*x1) + (16*x2)), tmp80 & xmask, eviction_policy='evict_last', other=0.0)
tmp82 = tl.full(tmp81.shape, 0.0, tmp81.dtype)
tmp83 = tl.where(tmp78, tmp81, tmp82)
tmp84 = tmp83 + tmp77
tmp85 = tmp70 & tmp34
tmp86 = tmp43 & tmp21
tmp87 = tmp86 & tmp85
tmp88 = tl.load(in_ptr0 + ((2*x0) + (8*x1) + (16*x2)), tmp87 & xmask, eviction_policy='evict_last', other=0.0)
tmp89 = tl.full(tmp88.shape, 0.0, tmp88.dtype)
tmp90 = tl.where(tmp85, tmp88, tmp89)
tmp91 = tmp90 + tmp84
tmp92 = (((0) * ((0) >= ((-1) + (2*x0))) + ((-1) + (2*x0)) * (((-1) + (2*x0)) > (0)))*((0) * ((0) >= ((-1) + (2*x1))) + ((-1) + (2*x1)) * (((-1) + (2*x1)) > (0)))) + (((5) * ((5) <= (2 + (2*x0))) + (2 + (2*x0)) * ((2 + (2*x0)) < (5)))*((5) * ((5) <= (2 + (2*x1))) + (2 + (2*x1)) * ((2 + (2*x1)) < (5)))) + ((-1)*((0) * ((0) >= ((-1) + (2*x0))) + ((-1) + (2*x0)) * (((-1) + (2*x0)) > (0)))*((5) * ((5) <= (2 + (2*x1))) + (2 + (2*x1)) * ((2 + (2*x1)) < (5)))) + ((-1)*((0) * ((0) >= ((-1) + (2*x1))) + ((-1) + (2*x1)) * (((-1) + (2*x1)) > (0)))*((5) * ((5) <= (2 + (2*x0))) + (2 + (2*x0)) * ((2 + (2*x0)) < (5))))
tmp93 = tmp91 / tmp92
tl.store(out_ptr0 + (x4), tmp93, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.constant_pad_nd, aten.avg_pool2d]
stream0 = get_raw_stream(0)
triton_poi_fused_avg_pool2d_constant_pad_nd_0.run(arg0_1, buf0, 144, grid=grid(144), stream=stream0)
del arg0_1
return (reinterpret_tensor(buf0, (4, 4, 2, 2), (36, 9, 3, 1), 4), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
class AvgPoolPad(nn.Module):
def __init__(self, stride=2, padding=1):
super(AvgPoolPad, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.AvgPool2d(3, stride=stride, padding=padding,
count_include_pad=False)
def forward(self, x):
x = self.pad(x)
x = self.pool(x)
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_avg_pool2d_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 3 % 3
x0 = xindex % 3
x2 = xindex // 9
x4 = xindex
tmp0 = -1 + 2 * x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 5, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = -1 + 2 * x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = -2 + 2 * x1
tmp12 = tmp11 >= tmp1
tmp13 = -2 + 2 * x0
tmp14 = tmp13 >= tmp1
tmp15 = tmp12 & tmp14
tmp16 = tmp15 & tmp10
tmp17 = tl.load(in_ptr0 + (-10 + 2 * x0 + 8 * x1 + 16 * x2), tmp16 &
xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.full(tmp17.shape, 0.0, tmp17.dtype)
tmp19 = tl.where(tmp10, tmp17, tmp18)
tmp20 = 2 * x0
tmp21 = tmp20 >= tmp1
tmp22 = tmp20 < tmp3
tmp23 = tmp21 & tmp22
tmp24 = tmp5 & tmp23
tmp25 = tmp12 & tmp7
tmp26 = tmp25 & tmp24
tmp27 = tl.load(in_ptr0 + (-9 + 2 * x0 + 8 * x1 + 16 * x2), tmp26 &
xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tl.full(tmp27.shape, 0.0, tmp27.dtype)
tmp29 = tl.where(tmp24, tmp27, tmp28)
tmp30 = tmp29 + tmp19
tmp31 = 1 + 2 * x0
tmp32 = tmp31 >= tmp1
tmp33 = tmp31 < tmp3
tmp34 = tmp32 & tmp33
tmp35 = tmp5 & tmp34
tmp36 = tmp12 & tmp21
tmp37 = tmp36 & tmp35
tmp38 = tl.load(in_ptr0 + (-8 + 2 * x0 + 8 * x1 + 16 * x2), tmp37 &
xmask, eviction_policy='evict_last', other=0.0)
tmp39 = tl.full(tmp38.shape, 0.0, tmp38.dtype)
tmp40 = tl.where(tmp35, tmp38, tmp39)
tmp41 = tmp40 + tmp30
tmp42 = 2 * x1
tmp43 = tmp42 >= tmp1
tmp44 = tmp42 < tmp3
tmp45 = tmp43 & tmp44
tmp46 = tmp45 & tmp9
tmp47 = tmp2 & tmp14
tmp48 = tmp47 & tmp46
tmp49 = tl.load(in_ptr0 + (-6 + 2 * x0 + 8 * x1 + 16 * x2), tmp48 &
xmask, eviction_policy='evict_last', other=0.0)
tmp50 = tl.full(tmp49.shape, 0.0, tmp49.dtype)
tmp51 = tl.where(tmp46, tmp49, tmp50)
tmp52 = tmp51 + tmp41
tmp53 = tmp45 & tmp23
tmp54 = tmp2 & tmp7
tmp55 = tmp54 & tmp53
tmp56 = tl.load(in_ptr0 + (-5 + 2 * x0 + 8 * x1 + 16 * x2), tmp55 &
xmask, eviction_policy='evict_last', other=0.0)
tmp57 = tl.full(tmp56.shape, 0.0, tmp56.dtype)
tmp58 = tl.where(tmp53, tmp56, tmp57)
tmp59 = tmp58 + tmp52
tmp60 = tmp45 & tmp34
tmp61 = tmp2 & tmp21
tmp62 = tmp61 & tmp60
tmp63 = tl.load(in_ptr0 + (-4 + 2 * x0 + 8 * x1 + 16 * x2), tmp62 &
xmask, eviction_policy='evict_last', other=0.0)
tmp64 = tl.full(tmp63.shape, 0.0, tmp63.dtype)
tmp65 = tl.where(tmp60, tmp63, tmp64)
tmp66 = tmp65 + tmp59
tmp67 = 1 + 2 * x1
tmp68 = tmp67 >= tmp1
tmp69 = tmp67 < tmp3
tmp70 = tmp68 & tmp69
tmp71 = tmp70 & tmp9
tmp72 = tmp43 & tmp14
tmp73 = tmp72 & tmp71
tmp74 = tl.load(in_ptr0 + (-2 + 2 * x0 + 8 * x1 + 16 * x2), tmp73 &
xmask, eviction_policy='evict_last', other=0.0)
tmp75 = tl.full(tmp74.shape, 0.0, tmp74.dtype)
tmp76 = tl.where(tmp71, tmp74, tmp75)
tmp77 = tmp76 + tmp66
tmp78 = tmp70 & tmp23
tmp79 = tmp43 & tmp7
tmp80 = tmp79 & tmp78
tmp81 = tl.load(in_ptr0 + (-1 + 2 * x0 + 8 * x1 + 16 * x2), tmp80 &
xmask, eviction_policy='evict_last', other=0.0)
tmp82 = tl.full(tmp81.shape, 0.0, tmp81.dtype)
tmp83 = tl.where(tmp78, tmp81, tmp82)
tmp84 = tmp83 + tmp77
tmp85 = tmp70 & tmp34
tmp86 = tmp43 & tmp21
tmp87 = tmp86 & tmp85
tmp88 = tl.load(in_ptr0 + (2 * x0 + 8 * x1 + 16 * x2), tmp87 & xmask,
eviction_policy='evict_last', other=0.0)
tmp89 = tl.full(tmp88.shape, 0.0, tmp88.dtype)
tmp90 = tl.where(tmp85, tmp88, tmp89)
tmp91 = tmp90 + tmp84
tmp92 = (0 * (0 >= -1 + 2 * x0) + (-1 + 2 * x0) * (-1 + 2 * x0 > 0)) * (
0 * (0 >= -1 + 2 * x1) + (-1 + 2 * x1) * (-1 + 2 * x1 > 0)) + (5 *
(5 <= 2 + 2 * x0) + (2 + 2 * x0) * (2 + 2 * x0 < 5)) * (5 * (5 <= 2 +
2 * x1) + (2 + 2 * x1) * (2 + 2 * x1 < 5)) + -1 * (0 * (0 >= -1 + 2 *
x0) + (-1 + 2 * x0) * (-1 + 2 * x0 > 0)) * (5 * (5 <= 2 + 2 * x1) +
(2 + 2 * x1) * (2 + 2 * x1 < 5)) + -1 * (0 * (0 >= -1 + 2 * x1) + (
-1 + 2 * x1) * (-1 + 2 * x1 > 0)) * (5 * (5 <= 2 + 2 * x0) + (2 + 2 *
x0) * (2 + 2 * x0 < 5))
tmp93 = tmp91 / tmp92
tl.store(out_ptr0 + x4, tmp93, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_avg_pool2d_constant_pad_nd_0[grid(144)](arg0_1,
buf0, 144, XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return reinterpret_tensor(buf0, (4, 4, 2, 2), (36, 9, 3, 1), 4),
class AvgPoolPadNew(nn.Module):
def __init__(self, stride=2, padding=1):
super(AvgPoolPadNew, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.AvgPool2d(3, stride=stride, padding=padding,
count_include_pad=False)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MarioProjects/pytorchlib | AvgPoolPad | false | 5,575 | [
"MIT"
] | 1 | 81ea32304d899fbd10ae1efe1d124c0d7bc96f5c | https://github.com/MarioProjects/pytorchlib/tree/81ea32304d899fbd10ae1efe1d124c0d7bc96f5c | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, stride=2, padding=1):
super().__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.AvgPool2d(3, stride=stride, padding=padding,
count_include_pad=False)
def forward(self, x):
x = self.pad(x)
x = self.pool(x)
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
MaxPoolPad | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/jf/cjf7zenaxtvwhbfrvvghsyyrrhxyrlvtj5rotfw7n2nqtvscv3l7.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.constant_pad_nd, aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# x => constant_pad_nd
# x_1 => _low_memory_max_pool2d_with_offsets
# Graph fragment:
# %constant_pad_nd : [num_users=1] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%arg0_1, [1, 0, 1, 0], 0.0), kwargs = {})
# %_low_memory_max_pool2d_with_offsets : [num_users=1] = call_function[target=torch.ops.prims._low_memory_max_pool2d_with_offsets.default](args = (%constant_pad_nd, [3, 3], [2, 2], [1, 1], [1, 1], False), kwargs = {})
triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 9, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 3) % 3
x0 = xindex % 3
x2 = (xindex // 9)
x4 = xindex
tmp0 = (-1) + (2*x1)
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 5, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = (-1) + (2*x0)
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = (-2) + (2*x1)
tmp12 = tmp11 >= tmp1
tmp13 = (-2) + (2*x0)
tmp14 = tmp13 >= tmp1
tmp15 = tmp12 & tmp14
tmp16 = tmp15 & tmp10
tmp17 = tl.load(in_ptr0 + ((-10) + (2*x0) + (8*x1) + (16*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.full(tmp17.shape, float("-inf"), tmp17.dtype)
tmp19 = tl.where(tmp10, tmp17, tmp18)
tmp20 = 2*x0
tmp21 = tmp20 >= tmp1
tmp22 = tmp20 < tmp3
tmp23 = tmp21 & tmp22
tmp24 = tmp5 & tmp23
tmp25 = tmp12 & tmp7
tmp26 = tmp25 & tmp24
tmp27 = tl.load(in_ptr0 + ((-9) + (2*x0) + (8*x1) + (16*x2)), tmp26 & xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tl.full(tmp27.shape, float("-inf"), tmp27.dtype)
tmp29 = tl.where(tmp24, tmp27, tmp28)
tmp30 = triton_helpers.maximum(tmp29, tmp19)
tmp31 = 1 + (2*x0)
tmp32 = tmp31 >= tmp1
tmp33 = tmp31 < tmp3
tmp34 = tmp32 & tmp33
tmp35 = tmp5 & tmp34
tmp36 = tmp12 & tmp21
tmp37 = tmp36 & tmp35
tmp38 = tl.load(in_ptr0 + ((-8) + (2*x0) + (8*x1) + (16*x2)), tmp37 & xmask, eviction_policy='evict_last', other=0.0)
tmp39 = tl.full(tmp38.shape, float("-inf"), tmp38.dtype)
tmp40 = tl.where(tmp35, tmp38, tmp39)
tmp41 = triton_helpers.maximum(tmp40, tmp30)
tmp42 = 2*x1
tmp43 = tmp42 >= tmp1
tmp44 = tmp42 < tmp3
tmp45 = tmp43 & tmp44
tmp46 = tmp45 & tmp9
tmp47 = tmp2 & tmp14
tmp48 = tmp47 & tmp46
tmp49 = tl.load(in_ptr0 + ((-6) + (2*x0) + (8*x1) + (16*x2)), tmp48 & xmask, eviction_policy='evict_last', other=0.0)
tmp50 = tl.full(tmp49.shape, float("-inf"), tmp49.dtype)
tmp51 = tl.where(tmp46, tmp49, tmp50)
tmp52 = triton_helpers.maximum(tmp51, tmp41)
tmp53 = tmp45 & tmp23
tmp54 = tmp2 & tmp7
tmp55 = tmp54 & tmp53
tmp56 = tl.load(in_ptr0 + ((-5) + (2*x0) + (8*x1) + (16*x2)), tmp55 & xmask, eviction_policy='evict_last', other=0.0)
tmp57 = tl.full(tmp56.shape, float("-inf"), tmp56.dtype)
tmp58 = tl.where(tmp53, tmp56, tmp57)
tmp59 = triton_helpers.maximum(tmp58, tmp52)
tmp60 = tmp45 & tmp34
tmp61 = tmp2 & tmp21
tmp62 = tmp61 & tmp60
tmp63 = tl.load(in_ptr0 + ((-4) + (2*x0) + (8*x1) + (16*x2)), tmp62 & xmask, eviction_policy='evict_last', other=0.0)
tmp64 = tl.full(tmp63.shape, float("-inf"), tmp63.dtype)
tmp65 = tl.where(tmp60, tmp63, tmp64)
tmp66 = triton_helpers.maximum(tmp65, tmp59)
tmp67 = 1 + (2*x1)
tmp68 = tmp67 >= tmp1
tmp69 = tmp67 < tmp3
tmp70 = tmp68 & tmp69
tmp71 = tmp70 & tmp9
tmp72 = tmp43 & tmp14
tmp73 = tmp72 & tmp71
tmp74 = tl.load(in_ptr0 + ((-2) + (2*x0) + (8*x1) + (16*x2)), tmp73 & xmask, eviction_policy='evict_last', other=0.0)
tmp75 = tl.full(tmp74.shape, float("-inf"), tmp74.dtype)
tmp76 = tl.where(tmp71, tmp74, tmp75)
tmp77 = triton_helpers.maximum(tmp76, tmp66)
tmp78 = tmp70 & tmp23
tmp79 = tmp43 & tmp7
tmp80 = tmp79 & tmp78
tmp81 = tl.load(in_ptr0 + ((-1) + (2*x0) + (8*x1) + (16*x2)), tmp80 & xmask, eviction_policy='evict_last', other=0.0)
tmp82 = tl.full(tmp81.shape, float("-inf"), tmp81.dtype)
tmp83 = tl.where(tmp78, tmp81, tmp82)
tmp84 = triton_helpers.maximum(tmp83, tmp77)
tmp85 = tmp70 & tmp34
tmp86 = tmp43 & tmp21
tmp87 = tmp86 & tmp85
tmp88 = tl.load(in_ptr0 + ((2*x0) + (8*x1) + (16*x2)), tmp87 & xmask, eviction_policy='evict_last', other=0.0)
tmp89 = tl.full(tmp88.shape, float("-inf"), tmp88.dtype)
tmp90 = tl.where(tmp85, tmp88, tmp89)
tmp91 = triton_helpers.maximum(tmp90, tmp84)
tl.store(out_ptr0 + (x4), tmp91, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.constant_pad_nd, aten.max_pool2d_with_indices]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0.run(arg0_1, buf0, 144, grid=grid(144), stream=stream0)
del arg0_1
return (reinterpret_tensor(buf0, (4, 4, 2, 2), (36, 9, 3, 1), 4), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
class MaxPoolPad(nn.Module):
def __init__(self):
super(MaxPoolPad, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.MaxPool2d(3, stride=2, padding=1)
def forward(self, x):
x = self.pad(x)
x = self.pool(x)
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0(in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 3 % 3
x0 = xindex % 3
x2 = xindex // 9
x4 = xindex
tmp0 = -1 + 2 * x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 5, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = -1 + 2 * x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = -2 + 2 * x1
tmp12 = tmp11 >= tmp1
tmp13 = -2 + 2 * x0
tmp14 = tmp13 >= tmp1
tmp15 = tmp12 & tmp14
tmp16 = tmp15 & tmp10
tmp17 = tl.load(in_ptr0 + (-10 + 2 * x0 + 8 * x1 + 16 * x2), tmp16 &
xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.full(tmp17.shape, float('-inf'), tmp17.dtype)
tmp19 = tl.where(tmp10, tmp17, tmp18)
tmp20 = 2 * x0
tmp21 = tmp20 >= tmp1
tmp22 = tmp20 < tmp3
tmp23 = tmp21 & tmp22
tmp24 = tmp5 & tmp23
tmp25 = tmp12 & tmp7
tmp26 = tmp25 & tmp24
tmp27 = tl.load(in_ptr0 + (-9 + 2 * x0 + 8 * x1 + 16 * x2), tmp26 &
xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tl.full(tmp27.shape, float('-inf'), tmp27.dtype)
tmp29 = tl.where(tmp24, tmp27, tmp28)
tmp30 = triton_helpers.maximum(tmp29, tmp19)
tmp31 = 1 + 2 * x0
tmp32 = tmp31 >= tmp1
tmp33 = tmp31 < tmp3
tmp34 = tmp32 & tmp33
tmp35 = tmp5 & tmp34
tmp36 = tmp12 & tmp21
tmp37 = tmp36 & tmp35
tmp38 = tl.load(in_ptr0 + (-8 + 2 * x0 + 8 * x1 + 16 * x2), tmp37 &
xmask, eviction_policy='evict_last', other=0.0)
tmp39 = tl.full(tmp38.shape, float('-inf'), tmp38.dtype)
tmp40 = tl.where(tmp35, tmp38, tmp39)
tmp41 = triton_helpers.maximum(tmp40, tmp30)
tmp42 = 2 * x1
tmp43 = tmp42 >= tmp1
tmp44 = tmp42 < tmp3
tmp45 = tmp43 & tmp44
tmp46 = tmp45 & tmp9
tmp47 = tmp2 & tmp14
tmp48 = tmp47 & tmp46
tmp49 = tl.load(in_ptr0 + (-6 + 2 * x0 + 8 * x1 + 16 * x2), tmp48 &
xmask, eviction_policy='evict_last', other=0.0)
tmp50 = tl.full(tmp49.shape, float('-inf'), tmp49.dtype)
tmp51 = tl.where(tmp46, tmp49, tmp50)
tmp52 = triton_helpers.maximum(tmp51, tmp41)
tmp53 = tmp45 & tmp23
tmp54 = tmp2 & tmp7
tmp55 = tmp54 & tmp53
tmp56 = tl.load(in_ptr0 + (-5 + 2 * x0 + 8 * x1 + 16 * x2), tmp55 &
xmask, eviction_policy='evict_last', other=0.0)
tmp57 = tl.full(tmp56.shape, float('-inf'), tmp56.dtype)
tmp58 = tl.where(tmp53, tmp56, tmp57)
tmp59 = triton_helpers.maximum(tmp58, tmp52)
tmp60 = tmp45 & tmp34
tmp61 = tmp2 & tmp21
tmp62 = tmp61 & tmp60
tmp63 = tl.load(in_ptr0 + (-4 + 2 * x0 + 8 * x1 + 16 * x2), tmp62 &
xmask, eviction_policy='evict_last', other=0.0)
tmp64 = tl.full(tmp63.shape, float('-inf'), tmp63.dtype)
tmp65 = tl.where(tmp60, tmp63, tmp64)
tmp66 = triton_helpers.maximum(tmp65, tmp59)
tmp67 = 1 + 2 * x1
tmp68 = tmp67 >= tmp1
tmp69 = tmp67 < tmp3
tmp70 = tmp68 & tmp69
tmp71 = tmp70 & tmp9
tmp72 = tmp43 & tmp14
tmp73 = tmp72 & tmp71
tmp74 = tl.load(in_ptr0 + (-2 + 2 * x0 + 8 * x1 + 16 * x2), tmp73 &
xmask, eviction_policy='evict_last', other=0.0)
tmp75 = tl.full(tmp74.shape, float('-inf'), tmp74.dtype)
tmp76 = tl.where(tmp71, tmp74, tmp75)
tmp77 = triton_helpers.maximum(tmp76, tmp66)
tmp78 = tmp70 & tmp23
tmp79 = tmp43 & tmp7
tmp80 = tmp79 & tmp78
tmp81 = tl.load(in_ptr0 + (-1 + 2 * x0 + 8 * x1 + 16 * x2), tmp80 &
xmask, eviction_policy='evict_last', other=0.0)
tmp82 = tl.full(tmp81.shape, float('-inf'), tmp81.dtype)
tmp83 = tl.where(tmp78, tmp81, tmp82)
tmp84 = triton_helpers.maximum(tmp83, tmp77)
tmp85 = tmp70 & tmp34
tmp86 = tmp43 & tmp21
tmp87 = tmp86 & tmp85
tmp88 = tl.load(in_ptr0 + (2 * x0 + 8 * x1 + 16 * x2), tmp87 & xmask,
eviction_policy='evict_last', other=0.0)
tmp89 = tl.full(tmp88.shape, float('-inf'), tmp88.dtype)
tmp90 = tl.where(tmp85, tmp88, tmp89)
tmp91 = triton_helpers.maximum(tmp90, tmp84)
tl.store(out_ptr0 + x4, tmp91, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0[grid(144)](
arg0_1, buf0, 144, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return reinterpret_tensor(buf0, (4, 4, 2, 2), (36, 9, 3, 1), 4),
class MaxPoolPadNew(nn.Module):
def __init__(self):
super(MaxPoolPadNew, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.MaxPool2d(3, stride=2, padding=1)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MarioProjects/pytorchlib | MaxPoolPad | false | 5,576 | [
"MIT"
] | 1 | 81ea32304d899fbd10ae1efe1d124c0d7bc96f5c | https://github.com/MarioProjects/pytorchlib/tree/81ea32304d899fbd10ae1efe1d124c0d7bc96f5c | import torch
from torch import nn
class Model(nn.Module):
def __init__(self):
super().__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.MaxPool2d(3, stride=2, padding=1)
def forward(self, x):
x = self.pad(x)
x = self.pool(x)
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Conv2d | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/z3/cz3vliqlpgih6ihwoaxl6cmnicfmv2ygutcuphilcsragp3evc57.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x => convolution
# x_1 => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
buf1 = buf0; del buf0 # reuse
buf2 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 1, 1), torch.bool)
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_threshold_backward_0.run(buf1, primals_2, buf2, 16, grid=grid(16), stream=stream0)
del primals_2
return (buf1, primals_1, primals_3, buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from math import sqrt as sqrt
from itertools import product as product
class Conv2d(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1, NL
='relu', same_padding=False, bn=False):
super(Conv2d, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
if NL == 'relu':
self.relu = nn.ReLU(inplace=True)
elif NL == 'prelu':
self.relu = nn.PReLU()
else:
self.relu = None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
from math import sqrt as sqrt
from itertools import product as product
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_0(in_out_ptr0,
in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 1, 1), (4, 1, 1, 1))
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 1, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_convolution_relu_threshold_backward_0[grid(16)](buf1,
primals_2, buf2, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_2
return buf1, primals_1, primals_3, buf2
class Conv2dNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1, NL
='relu', same_padding=False, bn=False):
super(Conv2dNew, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
if NL == 'relu':
self.relu = nn.ReLU(inplace=True)
elif NL == 'prelu':
self.relu = nn.PReLU()
else:
self.relu = None
def forward(self, input_0):
primals_1 = self.conv.weight
primals_2 = self.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| MONICA-Project/sfn | Conv2d | false | 5,577 | [
"Apache-2.0"
] | 1 | 40509e520e83441068b5a2d151864fe3a5814d5e | https://github.com/MONICA-Project/sfn/tree/40509e520e83441068b5a2d151864fe3a5814d5e | import torch
import torch.nn as nn
from math import sqrt as sqrt
from itertools import product as product
class Model(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1, NL
='relu', same_padding=False, bn=False):
super().__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
if NL == 'relu':
self.relu = nn.ReLU(inplace=True)
elif NL == 'prelu':
self.relu = nn.PReLU()
else:
self.relu = None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
Policy | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/fj/cfjqcl5zyffb5ly6gkhpzdblek24xftrdhp34wnn56hi434h2szb.py
# Topologically Sorted Source Nodes: [alphas], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# alphas => amax, div, exp, sub, sum_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%primals_1, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 2)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (2*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (2*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp4 = tmp0 - tmp3
tmp5 = tl_math.exp(tmp4)
tmp6 = tmp1 - tmp3
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp2 - tmp3
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tmp5 / tmp10
tl.store(out_ptr0 + (x2), tmp11, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, = args
args.clear()
assert_size_stride(primals_1, (6, 2), (2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((6, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [alphas], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(primals_1, buf0, 12, grid=grid(12), stream=stream0)
del primals_1
return (buf0, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((6, 2), (2, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from copy import deepcopy
import torch.nn as nn
class Policy(nn.Module):
def __init__(self, max_nodes, search_space):
super(Policy, self).__init__()
self.max_nodes = max_nodes
self.search_space = deepcopy(search_space)
self.edge2index = {}
for i in range(1, max_nodes):
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
self.edge2index[node_str] = len(self.edge2index)
self.arch_parameters = nn.Parameter(0.001 * torch.randn(len(self.
edge2index), len(search_space)))
def generate_arch(self, actions):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
op_name = self.search_space[actions[self.edge2index[node_str]]]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def genotype(self):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
with torch.no_grad():
weights = self.arch_parameters[self.edge2index[node_str]]
op_name = self.search_space[weights.argmax().item()]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def forward(self):
alphas = nn.functional.softmax(self.arch_parameters, dim=-1)
return alphas
def get_inputs():
return []
def get_init_inputs():
return [[], {'max_nodes': 4, 'search_space': [4, 4]}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from copy import deepcopy
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 2
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 2 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 2 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp4 = tmp0 - tmp3
tmp5 = tl_math.exp(tmp4)
tmp6 = tmp1 - tmp3
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp2 - tmp3
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tmp5 / tmp10
tl.store(out_ptr0 + x2, tmp11, xmask)
def call(args):
primals_1, = args
args.clear()
assert_size_stride(primals_1, (6, 2), (2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((6, 2), (2, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(12)](primals_1, buf0, 12, XBLOCK=
16, num_warps=1, num_stages=1)
del primals_1
return buf0, buf0
class PolicyNew(nn.Module):
def __init__(self, max_nodes, search_space):
super(PolicyNew, self).__init__()
self.max_nodes = max_nodes
self.search_space = deepcopy(search_space)
self.edge2index = {}
for i in range(1, max_nodes):
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
self.edge2index[node_str] = len(self.edge2index)
self.arch_parameters = nn.Parameter(0.001 * torch.randn(len(self.
edge2index), len(search_space)))
def generate_arch(self, actions):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
op_name = self.search_space[actions[self.edge2index[node_str]]]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def genotype(self):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
with torch.no_grad():
weights = self.arch_parameters[self.edge2index[node_str]]
op_name = self.search_space[weights.argmax().item()]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def forward(self):
primals_1 = self.arch_parameters
output = call([primals_1])
return output[0]
| MUST-AI-Lab/NAS-Projects | Policy | false | 5,578 | [
"MIT"
] | 1 | fcb2aae34a2b3c02877fbdb41cda45e1e73327a6 | https://github.com/MUST-AI-Lab/NAS-Projects/tree/fcb2aae34a2b3c02877fbdb41cda45e1e73327a6 | import torch
from copy import deepcopy
import torch.nn as nn
class Model(nn.Module):
def __init__(self, max_nodes, search_space):
super().__init__()
self.max_nodes = max_nodes
self.search_space = deepcopy(search_space)
self.edge2index = {}
for i in range(1, max_nodes):
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
self.edge2index[node_str] = len(self.edge2index)
self.arch_parameters = nn.Parameter(0.001 * torch.randn(len(self.
edge2index), len(search_space)))
def generate_arch(self, actions):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
op_name = self.search_space[actions[self.edge2index[node_str]]]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def genotype(self):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
with torch.no_grad():
weights = self.arch_parameters[self.edge2index[node_str]]
op_name = self.search_space[weights.argmax().item()]
xlist.append((op_name, j))
genotypes.append(tuple(xlist))
return CellStructure(genotypes)
def forward(self):
alphas = nn.functional.softmax(self.arch_parameters, dim=-1)
return alphas
def get_inputs():
return []
def get_init_inputs():
return [4]
|
FScoreLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/fy/cfyns74s37xab276pc6dbqyls6hbctvmvaztno7qbvcax2iqsmkp.py
# Topologically Sorted Source Nodes: [tp, tp_1, tp_2, tp_3, mul_3, mul_4, sub, fn, fn_1, fn_2, mul_5, add, sub_1, fp, fp_1, fp_2, add_1, add_2, truediv, result, result_1], Original ATen: [aten.mul, aten.sum, aten.add, aten.rsub, aten.div, aten.mean]
# Source node to ATen node mapping:
# add => add_1
# add_1 => add_2
# add_2 => add_3
# fn => mul_1
# fn_1 => sum_2
# fn_2 => sum_5
# fp => mul_2
# fp_1 => sum_3
# fp_2 => sum_6
# mul_3 => mul_3
# mul_4 => mul_4
# mul_5 => mul_5
# result => sub_2
# result_1 => mean
# sub => sub
# sub_1 => sub_1
# tp => mul
# tp_1 => sum_1
# tp_2 => sum_4
# tp_3 => add
# truediv => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, %arg0_1), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [-1]), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_1, [-1]), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_4, 0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 2.0), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %arg0_1), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [-1]), kwargs = {})
# %sum_5 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_2, [-1]), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_5, 1.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %mul_5), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg0_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, %sub_1), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_2, [-1]), kwargs = {})
# %sum_6 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sum_3, [-1]), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, %sum_6), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_3, %add_3), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %div), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sub_2,), kwargs = {})
triton_per_fused_add_div_mean_mul_rsub_sum_0 = async_compile.triton('triton_per_fused_add_div_mean_mul_rsub_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 16],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mean_mul_rsub_sum_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 32, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_sum_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 16
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (16*r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (16*r0), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (16*r0)), None, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + (16*r0)), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (16*r0)), None, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + (16*r0)), None, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (16*r0)), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (3 + (16*r0)), None, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (4 + (16*r0)), None, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr1 + (4 + (16*r0)), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (5 + (16*r0)), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (5 + (16*r0)), None, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr0 + (6 + (16*r0)), None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + (6 + (16*r0)), None, eviction_policy='evict_last')
tmp26 = tl.load(in_ptr0 + (7 + (16*r0)), None, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr1 + (7 + (16*r0)), None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr0 + (8 + (16*r0)), None, eviction_policy='evict_last')
tmp32 = tl.load(in_ptr1 + (8 + (16*r0)), None, eviction_policy='evict_last')
tmp34 = tl.load(in_ptr0 + (9 + (16*r0)), None, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr1 + (9 + (16*r0)), None, eviction_policy='evict_last')
tmp38 = tl.load(in_ptr0 + (10 + (16*r0)), None, eviction_policy='evict_last')
tmp39 = tl.load(in_ptr1 + (10 + (16*r0)), None, eviction_policy='evict_last')
tmp42 = tl.load(in_ptr0 + (11 + (16*r0)), None, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr1 + (11 + (16*r0)), None, eviction_policy='evict_last')
tmp47 = tl.load(in_ptr0 + (12 + (16*r0)), None, eviction_policy='evict_last')
tmp48 = tl.load(in_ptr1 + (12 + (16*r0)), None, eviction_policy='evict_last')
tmp50 = tl.load(in_ptr0 + (13 + (16*r0)), None, eviction_policy='evict_last')
tmp51 = tl.load(in_ptr1 + (13 + (16*r0)), None, eviction_policy='evict_last')
tmp54 = tl.load(in_ptr0 + (14 + (16*r0)), None, eviction_policy='evict_last')
tmp55 = tl.load(in_ptr1 + (14 + (16*r0)), None, eviction_policy='evict_last')
tmp58 = tl.load(in_ptr0 + (15 + (16*r0)), None, eviction_policy='evict_last')
tmp59 = tl.load(in_ptr1 + (15 + (16*r0)), None, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tmp17 = tmp15 * tmp16
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tmp24 = tmp22 * tmp23
tmp25 = tmp21 + tmp24
tmp28 = tmp26 * tmp27
tmp29 = tmp25 + tmp28
tmp30 = tmp14 + tmp29
tmp33 = tmp31 * tmp32
tmp36 = tmp34 * tmp35
tmp37 = tmp33 + tmp36
tmp40 = tmp38 * tmp39
tmp41 = tmp37 + tmp40
tmp44 = tmp42 * tmp43
tmp45 = tmp41 + tmp44
tmp46 = tmp30 + tmp45
tmp49 = tmp47 * tmp48
tmp52 = tmp50 * tmp51
tmp53 = tmp49 + tmp52
tmp56 = tmp54 * tmp55
tmp57 = tmp53 + tmp56
tmp60 = tmp58 * tmp59
tmp61 = tmp57 + tmp60
tmp62 = tmp46 + tmp61
tmp63 = 1.0
tmp64 = tmp63 - tmp0
tmp65 = tmp64 * tmp1
tmp66 = tmp63 - tmp3
tmp67 = tmp66 * tmp4
tmp68 = tmp65 + tmp67
tmp69 = tmp63 - tmp7
tmp70 = tmp69 * tmp8
tmp71 = tmp68 + tmp70
tmp72 = tmp63 - tmp11
tmp73 = tmp72 * tmp12
tmp74 = tmp71 + tmp73
tmp75 = tmp63 - tmp15
tmp76 = tmp75 * tmp16
tmp77 = tmp63 - tmp18
tmp78 = tmp77 * tmp19
tmp79 = tmp76 + tmp78
tmp80 = tmp63 - tmp22
tmp81 = tmp80 * tmp23
tmp82 = tmp79 + tmp81
tmp83 = tmp63 - tmp26
tmp84 = tmp83 * tmp27
tmp85 = tmp82 + tmp84
tmp86 = tmp74 + tmp85
tmp87 = tmp63 - tmp31
tmp88 = tmp87 * tmp32
tmp89 = tmp63 - tmp34
tmp90 = tmp89 * tmp35
tmp91 = tmp88 + tmp90
tmp92 = tmp63 - tmp38
tmp93 = tmp92 * tmp39
tmp94 = tmp91 + tmp93
tmp95 = tmp63 - tmp42
tmp96 = tmp95 * tmp43
tmp97 = tmp94 + tmp96
tmp98 = tmp86 + tmp97
tmp99 = tmp63 - tmp47
tmp100 = tmp99 * tmp48
tmp101 = tmp63 - tmp50
tmp102 = tmp101 * tmp51
tmp103 = tmp100 + tmp102
tmp104 = tmp63 - tmp54
tmp105 = tmp104 * tmp55
tmp106 = tmp103 + tmp105
tmp107 = tmp63 - tmp58
tmp108 = tmp107 * tmp59
tmp109 = tmp106 + tmp108
tmp110 = tmp98 + tmp109
tmp111 = tmp63 - tmp1
tmp112 = tmp0 * tmp111
tmp113 = tmp63 - tmp4
tmp114 = tmp3 * tmp113
tmp115 = tmp112 + tmp114
tmp116 = tmp63 - tmp8
tmp117 = tmp7 * tmp116
tmp118 = tmp115 + tmp117
tmp119 = tmp63 - tmp12
tmp120 = tmp11 * tmp119
tmp121 = tmp118 + tmp120
tmp122 = tmp63 - tmp16
tmp123 = tmp15 * tmp122
tmp124 = tmp63 - tmp19
tmp125 = tmp18 * tmp124
tmp126 = tmp123 + tmp125
tmp127 = tmp63 - tmp23
tmp128 = tmp22 * tmp127
tmp129 = tmp126 + tmp128
tmp130 = tmp63 - tmp27
tmp131 = tmp26 * tmp130
tmp132 = tmp129 + tmp131
tmp133 = tmp121 + tmp132
tmp134 = tmp63 - tmp32
tmp135 = tmp31 * tmp134
tmp136 = tmp63 - tmp35
tmp137 = tmp34 * tmp136
tmp138 = tmp135 + tmp137
tmp139 = tmp63 - tmp39
tmp140 = tmp38 * tmp139
tmp141 = tmp138 + tmp140
tmp142 = tmp63 - tmp43
tmp143 = tmp42 * tmp142
tmp144 = tmp141 + tmp143
tmp145 = tmp133 + tmp144
tmp146 = tmp63 - tmp48
tmp147 = tmp47 * tmp146
tmp148 = tmp63 - tmp51
tmp149 = tmp50 * tmp148
tmp150 = tmp147 + tmp149
tmp151 = tmp63 - tmp55
tmp152 = tmp54 * tmp151
tmp153 = tmp150 + tmp152
tmp154 = tmp63 - tmp59
tmp155 = tmp58 * tmp154
tmp156 = tmp153 + tmp155
tmp157 = tmp145 + tmp156
tmp158 = 0.0
tmp159 = tmp62 + tmp158
tmp160 = 2.0
tmp161 = tmp159 * tmp160
tmp162 = tmp110 * tmp63
tmp163 = tmp161 + tmp162
tmp164 = tmp163 + tmp157
tmp165 = 1e-06
tmp166 = tmp164 + tmp165
tmp167 = tmp161 / tmp166
tmp168 = tmp63 - tmp167
tmp169 = tl.broadcast_to(tmp168, [XBLOCK, RBLOCK])
tmp171 = tl.sum(tmp169, 1)[:, None]
tmp172 = 16.0
tmp173 = tmp171 / tmp172
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp173, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [tp, tp_1, tp_2, tp_3, mul_3, mul_4, sub, fn, fn_1, fn_2, mul_5, add, sub_1, fp, fp_1, fp_2, add_1, add_2, truediv, result, result_1], Original ATen: [aten.mul, aten.sum, aten.add, aten.rsub, aten.div, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_add_div_mean_mul_rsub_sum_0.run(buf4, arg1_1, arg0_1, 1, 16, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
class FScoreLoss(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super(FScoreLoss, self).__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
assert not target.requires_grad
if self.make_onehot:
target = make_onehot_segmentation(target, range(input_.shape[1]
), newaxis=self.make_onehot_newaxis)
target = target.float()
tp = input_ * target
fn = (1 - input_) * target
fp = input_ * (1 - target)
while tp.dim() > 2:
tp = tp.sum(-1, keepdim=False)
fn = fn.sum(-1, keepdim=False)
fp = fp.sum(-1, keepdim=False)
tp += self.tp_bias
result = 1 - (1 + self.beta * self.beta) * tp / ((1 + self.beta *
self.beta) * tp + self.beta * self.beta * fn + fp + self.eps)
if self.reduction != 'none':
if self.weight is not None:
self.weight = self.weight.float()
self.weight = self.weight
weight = self.weight.expand_as(result)
result = result * weight
if self.ignore_index is not None:
for cls in sorted(self.ignore_index, reverse=True):
if cls == result.shape[1] - 1:
result = result[:, :-1]
elif cls == 0:
result = result[:, 1:]
else:
result = torch.cat([result[:, :cls], result[:, cls +
1:]], 1)
if self.reduction == 'mean':
result = torch.mean(result)
elif self.reduction == 'sum':
result = torch.sum(result)
else:
raise ValueError(
"reduction must be 'none', 'mean' or 'sum', but is {}".
format(self.reduction))
return result
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_sum_0(in_out_ptr0, in_ptr0,
in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + 16 * r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 16 * r0, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 16 * r0), None, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 16 * r0), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 16 * r0), None, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + 16 * r0), None, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 16 * r0), None, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (3 + 16 * r0), None, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr0 + (4 + 16 * r0), None, eviction_policy='evict_last'
)
tmp16 = tl.load(in_ptr1 + (4 + 16 * r0), None, eviction_policy='evict_last'
)
tmp18 = tl.load(in_ptr0 + (5 + 16 * r0), None, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr1 + (5 + 16 * r0), None, eviction_policy='evict_last'
)
tmp22 = tl.load(in_ptr0 + (6 + 16 * r0), None, eviction_policy='evict_last'
)
tmp23 = tl.load(in_ptr1 + (6 + 16 * r0), None, eviction_policy='evict_last'
)
tmp26 = tl.load(in_ptr0 + (7 + 16 * r0), None, eviction_policy='evict_last'
)
tmp27 = tl.load(in_ptr1 + (7 + 16 * r0), None, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr0 + (8 + 16 * r0), None, eviction_policy='evict_last'
)
tmp32 = tl.load(in_ptr1 + (8 + 16 * r0), None, eviction_policy='evict_last'
)
tmp34 = tl.load(in_ptr0 + (9 + 16 * r0), None, eviction_policy='evict_last'
)
tmp35 = tl.load(in_ptr1 + (9 + 16 * r0), None, eviction_policy='evict_last'
)
tmp38 = tl.load(in_ptr0 + (10 + 16 * r0), None, eviction_policy=
'evict_last')
tmp39 = tl.load(in_ptr1 + (10 + 16 * r0), None, eviction_policy=
'evict_last')
tmp42 = tl.load(in_ptr0 + (11 + 16 * r0), None, eviction_policy=
'evict_last')
tmp43 = tl.load(in_ptr1 + (11 + 16 * r0), None, eviction_policy=
'evict_last')
tmp47 = tl.load(in_ptr0 + (12 + 16 * r0), None, eviction_policy=
'evict_last')
tmp48 = tl.load(in_ptr1 + (12 + 16 * r0), None, eviction_policy=
'evict_last')
tmp50 = tl.load(in_ptr0 + (13 + 16 * r0), None, eviction_policy=
'evict_last')
tmp51 = tl.load(in_ptr1 + (13 + 16 * r0), None, eviction_policy=
'evict_last')
tmp54 = tl.load(in_ptr0 + (14 + 16 * r0), None, eviction_policy=
'evict_last')
tmp55 = tl.load(in_ptr1 + (14 + 16 * r0), None, eviction_policy=
'evict_last')
tmp58 = tl.load(in_ptr0 + (15 + 16 * r0), None, eviction_policy=
'evict_last')
tmp59 = tl.load(in_ptr1 + (15 + 16 * r0), None, eviction_policy=
'evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tmp17 = tmp15 * tmp16
tmp20 = tmp18 * tmp19
tmp21 = tmp17 + tmp20
tmp24 = tmp22 * tmp23
tmp25 = tmp21 + tmp24
tmp28 = tmp26 * tmp27
tmp29 = tmp25 + tmp28
tmp30 = tmp14 + tmp29
tmp33 = tmp31 * tmp32
tmp36 = tmp34 * tmp35
tmp37 = tmp33 + tmp36
tmp40 = tmp38 * tmp39
tmp41 = tmp37 + tmp40
tmp44 = tmp42 * tmp43
tmp45 = tmp41 + tmp44
tmp46 = tmp30 + tmp45
tmp49 = tmp47 * tmp48
tmp52 = tmp50 * tmp51
tmp53 = tmp49 + tmp52
tmp56 = tmp54 * tmp55
tmp57 = tmp53 + tmp56
tmp60 = tmp58 * tmp59
tmp61 = tmp57 + tmp60
tmp62 = tmp46 + tmp61
tmp63 = 1.0
tmp64 = tmp63 - tmp0
tmp65 = tmp64 * tmp1
tmp66 = tmp63 - tmp3
tmp67 = tmp66 * tmp4
tmp68 = tmp65 + tmp67
tmp69 = tmp63 - tmp7
tmp70 = tmp69 * tmp8
tmp71 = tmp68 + tmp70
tmp72 = tmp63 - tmp11
tmp73 = tmp72 * tmp12
tmp74 = tmp71 + tmp73
tmp75 = tmp63 - tmp15
tmp76 = tmp75 * tmp16
tmp77 = tmp63 - tmp18
tmp78 = tmp77 * tmp19
tmp79 = tmp76 + tmp78
tmp80 = tmp63 - tmp22
tmp81 = tmp80 * tmp23
tmp82 = tmp79 + tmp81
tmp83 = tmp63 - tmp26
tmp84 = tmp83 * tmp27
tmp85 = tmp82 + tmp84
tmp86 = tmp74 + tmp85
tmp87 = tmp63 - tmp31
tmp88 = tmp87 * tmp32
tmp89 = tmp63 - tmp34
tmp90 = tmp89 * tmp35
tmp91 = tmp88 + tmp90
tmp92 = tmp63 - tmp38
tmp93 = tmp92 * tmp39
tmp94 = tmp91 + tmp93
tmp95 = tmp63 - tmp42
tmp96 = tmp95 * tmp43
tmp97 = tmp94 + tmp96
tmp98 = tmp86 + tmp97
tmp99 = tmp63 - tmp47
tmp100 = tmp99 * tmp48
tmp101 = tmp63 - tmp50
tmp102 = tmp101 * tmp51
tmp103 = tmp100 + tmp102
tmp104 = tmp63 - tmp54
tmp105 = tmp104 * tmp55
tmp106 = tmp103 + tmp105
tmp107 = tmp63 - tmp58
tmp108 = tmp107 * tmp59
tmp109 = tmp106 + tmp108
tmp110 = tmp98 + tmp109
tmp111 = tmp63 - tmp1
tmp112 = tmp0 * tmp111
tmp113 = tmp63 - tmp4
tmp114 = tmp3 * tmp113
tmp115 = tmp112 + tmp114
tmp116 = tmp63 - tmp8
tmp117 = tmp7 * tmp116
tmp118 = tmp115 + tmp117
tmp119 = tmp63 - tmp12
tmp120 = tmp11 * tmp119
tmp121 = tmp118 + tmp120
tmp122 = tmp63 - tmp16
tmp123 = tmp15 * tmp122
tmp124 = tmp63 - tmp19
tmp125 = tmp18 * tmp124
tmp126 = tmp123 + tmp125
tmp127 = tmp63 - tmp23
tmp128 = tmp22 * tmp127
tmp129 = tmp126 + tmp128
tmp130 = tmp63 - tmp27
tmp131 = tmp26 * tmp130
tmp132 = tmp129 + tmp131
tmp133 = tmp121 + tmp132
tmp134 = tmp63 - tmp32
tmp135 = tmp31 * tmp134
tmp136 = tmp63 - tmp35
tmp137 = tmp34 * tmp136
tmp138 = tmp135 + tmp137
tmp139 = tmp63 - tmp39
tmp140 = tmp38 * tmp139
tmp141 = tmp138 + tmp140
tmp142 = tmp63 - tmp43
tmp143 = tmp42 * tmp142
tmp144 = tmp141 + tmp143
tmp145 = tmp133 + tmp144
tmp146 = tmp63 - tmp48
tmp147 = tmp47 * tmp146
tmp148 = tmp63 - tmp51
tmp149 = tmp50 * tmp148
tmp150 = tmp147 + tmp149
tmp151 = tmp63 - tmp55
tmp152 = tmp54 * tmp151
tmp153 = tmp150 + tmp152
tmp154 = tmp63 - tmp59
tmp155 = tmp58 * tmp154
tmp156 = tmp153 + tmp155
tmp157 = tmp145 + tmp156
tmp158 = 0.0
tmp159 = tmp62 + tmp158
tmp160 = 2.0
tmp161 = tmp159 * tmp160
tmp162 = tmp110 * tmp63
tmp163 = tmp161 + tmp162
tmp164 = tmp163 + tmp157
tmp165 = 1e-06
tmp166 = tmp164 + tmp165
tmp167 = tmp161 / tmp166
tmp168 = tmp63 - tmp167
tmp169 = tl.broadcast_to(tmp168, [XBLOCK, RBLOCK])
tmp171 = tl.sum(tmp169, 1)[:, None]
tmp172 = 16.0
tmp173 = tmp171 / tmp172
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp173, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
get_raw_stream(0)
triton_per_fused_add_div_mean_mul_rsub_sum_0[grid(1)](buf4, arg1_1,
arg0_1, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf4,
class FScoreLossNew(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super(FScoreLossNew, self).__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MIC-DKFZ/image-time-series | FScoreLoss | false | 5,579 | [
"MIT"
] | 1 | 0480d5cb6936c7d9e839b6741f18c10893d78d8a | https://github.com/MIC-DKFZ/image-time-series/tree/0480d5cb6936c7d9e839b6741f18c10893d78d8a | import torch
from typing import Union
from typing import Optional
from typing import Iterable
from torch import nn
class Model(nn.modules.loss._WeightedLoss):
"""Uses the 1 - F-score as a loss.
.. math::
F = rac{ (1 + eta^2) TP }{ (1 + eta^2) TP + eta^2 FN + FP }
Args:
beta: The beta in the above formula.
eps: Epsilon for numerical stability.
tp_bias: This is added to the TP count. Should add a little stability for
very small structures.
make_onehot: Convert the target segmentation to onehot internally. Turn this
off if the target is already onehot.
make_onehot_newaxis: 'newaxis' argument for the onehot conversion.
ignore_index: These classes will not contribute to the loss. Has no effect if
reduction is "none".
weight: Weights for the different classes. Has no effect if reduction is "none".
reduction: "mean", "sum", or "none".
"""
def __init__(self, beta: 'float'=1.0, eps: 'float'=1e-06, tp_bias:
'Union[int, float]'=0, make_onehot: 'bool'=False,
make_onehot_newaxis: 'bool'=False, ignore_index:
'Optional[Union[int, Iterable[int]]]'=None, weight:
'Optional[torch.Tensor]'=None, reduction='mean', **kwargs):
super().__init__(weight=weight, reduction=reduction)
self.beta = beta
self.eps = eps
self.tp_bias = tp_bias
self.make_onehot = make_onehot
self.make_onehot_newaxis = make_onehot_newaxis
self.ignore_index = ignore_index
if isinstance(ignore_index, int):
self.ignore_index = [ignore_index]
def forward(self, input_: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
"""Forward pass through the loss.
Args:
input_: Input with expected shape (B, C, ...) with C the number of classes.
target: Target, either onehot with shape (B, C, ...), or not with shape
either (B, ...) or (B, 1, ...). Make sure to set the make_onehot and
make_onehot_newaxis arguments appropriately.
Returns:
The loss.
"""
assert not target.requires_grad
if self.make_onehot:
target = make_onehot_segmentation(target, range(input_.shape[1]
), newaxis=self.make_onehot_newaxis)
target = target.float()
tp = input_ * target
fn = (1 - input_) * target
fp = input_ * (1 - target)
while tp.dim() > 2:
tp = tp.sum(-1, keepdim=False)
fn = fn.sum(-1, keepdim=False)
fp = fp.sum(-1, keepdim=False)
tp += self.tp_bias
result = 1 - (1 + self.beta * self.beta) * tp / ((1 + self.beta *
self.beta) * tp + self.beta * self.beta * fn + fp + self.eps)
if self.reduction != 'none':
if self.weight is not None:
self.weight = self.weight.float()
self.weight = self.weight
weight = self.weight.expand_as(result)
result = result * weight
if self.ignore_index is not None:
for cls in sorted(self.ignore_index, reverse=True):
if cls == result.shape[1] - 1:
result = result[:, :-1]
elif cls == 0:
result = result[:, 1:]
else:
result = torch.cat([result[:, :cls], result[:, cls +
1:]], 1)
if self.reduction == 'mean':
result = torch.mean(result)
elif self.reduction == 'sum':
result = torch.sum(result)
else:
raise ValueError(
"reduction must be 'none', 'mean' or 'sum', but is {}".
format(self.reduction))
return result
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_i
# ... truncated (>4000 chars) for memory efficiency |
Conv_Block_gn | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/aq/caqnmb3rs4apw4atzgbt3bm7z3qshygxt3ddekkomaxcoiuso5oy.py
# Topologically Sorted Source Nodes: [x, x_2], Original ATen: [aten.convolution, aten.native_group_norm]
# Source node to ATen node mapping:
# x => convolution
# x_2 => add, add_1, mul_2, rsqrt, var_mean
# Graph fragment:
# %convolution : [num_users=4] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view, [2, 3]), kwargs = {correction: 0, keepdim: True})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, %unsqueeze_5), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %unsqueeze_2), kwargs = {})
triton_per_fused_convolution_native_group_norm_0 = async_compile.triton('triton_per_fused_convolution_native_group_norm_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[4, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32', 8: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_convolution_native_group_norm_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 4, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_convolution_native_group_norm_0(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr2, out_ptr3, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = rindex < rnumel
r3 = rindex
x0 = xindex
r2 = (rindex // 9)
tmp0 = tl.load(in_out_ptr0 + (r3 + (36*x0)), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (r2), rmask, eviction_policy='evict_last', other=0.0)
tmp31 = tl.load(in_ptr1 + (r2), rmask, eviction_policy='evict_last', other=0.0)
tmp33 = tl.load(in_ptr2 + (r2), rmask, eviction_policy='evict_last', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tmp8 = tl.broadcast_to(tmp7, [XBLOCK, RBLOCK])
tmp10 = tl.where(rmask & xmask, tmp8, 0)
tmp11 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp13 = tl.where(rmask & xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tl.full([XBLOCK, 1], 36, tl.int32)
tmp16 = tmp15.to(tl.float32)
tmp17 = tmp14 / tmp16
tmp18 = tmp8 - tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask & xmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = tmp7 - tmp17
tmp25 = 36.0
tmp26 = tmp23 / tmp25
tmp27 = 1e-05
tmp28 = tmp26 + tmp27
tmp29 = libdevice.rsqrt(tmp28)
tmp30 = tmp24 * tmp29
tmp32 = tmp30 * tmp31
tmp34 = tmp32 + tmp33
tl.store(in_out_ptr0 + (r3 + (36*x0)), tmp2, rmask & xmask)
tl.store(out_ptr2 + (r3 + (36*x0)), tmp34, rmask & xmask)
tl.store(out_ptr3 + (x0), tmp29, xmask)
tl.store(out_ptr0 + (x0), tmp17, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 3, 3), (36, 9, 3, 1))
buf1 = buf0; del buf0 # reuse
buf2 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
buf5 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
buf6 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
# Topologically Sorted Source Nodes: [x, x_2], Original ATen: [aten.convolution, aten.native_group_norm]
stream0 = get_raw_stream(0)
triton_per_fused_convolution_native_group_norm_0.run(buf1, primals_2, primals_4, primals_5, buf2, buf5, buf6, 4, 36, grid=grid(4), stream=stream0)
del primals_2
del primals_5
return (buf5, primals_1, primals_3, primals_4, buf1, reinterpret_tensor(buf2, (4, 1), (1, 1), 0), reinterpret_tensor(buf6, (4, 1), (1, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
from torch.autograd.variable import *
class Conv_Block_gn(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, groups, stride=1
):
super(Conv_Block_gn, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=
kernel_size, stride=stride, padding=1)
self.relu = torch.nn.LeakyReLU()
self.bn = nn.GroupNorm(groups, out_channels)
def forward(self, x):
x = self.conv(x)
x = self.relu(x)
x = self.bn(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4,
'groups': 1}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.autograd.variable import *
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_convolution_native_group_norm_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, out_ptr0, out_ptr2, out_ptr3, xnumel, rnumel, XBLOCK:
tl.constexpr):
xnumel = 4
rnumel = 36
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r3 = rindex
x0 = xindex
r2 = rindex // 9
tmp0 = tl.load(in_out_ptr0 + (r3 + 36 * x0), rmask & xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + r2, rmask, eviction_policy='evict_last', other=0.0
)
tmp31 = tl.load(in_ptr1 + r2, rmask, eviction_policy='evict_last',
other=0.0)
tmp33 = tl.load(in_ptr2 + r2, rmask, eviction_policy='evict_last',
other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tmp8 = tl.broadcast_to(tmp7, [XBLOCK, RBLOCK])
tl.where(rmask & xmask, tmp8, 0)
tmp11 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp13 = tl.where(rmask & xmask, tmp11, 0)
tmp14 = tl.sum(tmp13, 1)[:, None]
tmp15 = tl.full([XBLOCK, 1], 36, tl.int32)
tmp16 = tmp15.to(tl.float32)
tmp17 = tmp14 / tmp16
tmp18 = tmp8 - tmp17
tmp19 = tmp18 * tmp18
tmp20 = tl.broadcast_to(tmp19, [XBLOCK, RBLOCK])
tmp22 = tl.where(rmask & xmask, tmp20, 0)
tmp23 = tl.sum(tmp22, 1)[:, None]
tmp24 = tmp7 - tmp17
tmp25 = 36.0
tmp26 = tmp23 / tmp25
tmp27 = 1e-05
tmp28 = tmp26 + tmp27
tmp29 = libdevice.rsqrt(tmp28)
tmp30 = tmp24 * tmp29
tmp32 = tmp30 * tmp31
tmp34 = tmp32 + tmp33
tl.store(in_out_ptr0 + (r3 + 36 * x0), tmp2, rmask & xmask)
tl.store(out_ptr2 + (r3 + 36 * x0), tmp34, rmask & xmask)
tl.store(out_ptr3 + x0, tmp29, xmask)
tl.store(out_ptr0 + x0, tmp17, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 3, 3), (36, 9, 3, 1))
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
buf5 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
buf6 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
get_raw_stream(0)
triton_per_fused_convolution_native_group_norm_0[grid(4)](buf1,
primals_2, primals_4, primals_5, buf2, buf5, buf6, 4, 36,
XBLOCK=1, num_warps=2, num_stages=1)
del primals_2
del primals_5
return buf5, primals_1, primals_3, primals_4, buf1, reinterpret_tensor(buf2
, (4, 1), (1, 1), 0), reinterpret_tensor(buf6, (4, 1), (1, 1), 0)
class Conv_Block_gnNew(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, groups, stride=1
):
super(Conv_Block_gnNew, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=
kernel_size, stride=stride, padding=1)
self.relu = torch.nn.LeakyReLU()
self.bn = nn.GroupNorm(groups, out_channels)
def forward(self, input_0):
primals_1 = self.conv.weight
primals_2 = self.conv.bias
primals_4 = self.bn.weight
primals_5 = self.bn.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| MRLoghmani/Separate_to_Adapt | Conv_Block_gn | false | 5,580 | [
"MIT"
] | 1 | 09c734448aa22b3879186f59952d9fd596d4a1f8 | https://github.com/MRLoghmani/Separate_to_Adapt/tree/09c734448aa22b3879186f59952d9fd596d4a1f8 | import torch
import torch.nn as nn
from torch.autograd.variable import *
class Model(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, groups, stride=1
):
super().__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=
kernel_size, stride=stride, padding=1)
self.relu = torch.nn.LeakyReLU()
self.bn = nn.GroupNorm(groups, out_channels)
def forward(self, x):
x = self.conv(x)
x = self.relu(x)
x = self.bn(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4,
'groups': 1}]
|
FCNet | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/bm/cbmd63mrouqmm2pha5x6evse3dkbpy5o4xnk5v7quflfkqfdvwck.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# output_1 => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[512],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 320
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 5
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (5, 4), (4, 1))
assert_size_stride(primals_2, (5, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 5), (5, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 5), (5, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 5), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 5), (80, 20, 5, 1), 0); del buf0 # reuse
buf3 = empty_strided_cuda((4, 4, 4, 5), (80, 20, 5, 1), torch.bool)
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf3, 320, grid=grid(320), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 5), (5, 1), 0), reinterpret_tensor(primals_4, (5, 4), (1, 5), 0), alpha=1, beta=1, out=buf2)
del primals_5
return (reinterpret_tensor(buf2, (256, ), (1, ), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 5), (5, 1), 0), primals_4, buf3, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((5, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((5, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 5), (5, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
class FCNet(nn.Module):
def __init__(self, input_size, output_size):
super().__init__()
self.l1 = nn.Linear(input_size, 5)
self.relu = nn.ReLU()
self.l2 = nn.Linear(5, output_size)
def forward(self, x):
output = self.l1(x)
output = self.relu(output)
output = self.l2(output)
return output.view(-1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_size': 4, 'output_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 320
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 5
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (5, 4), (4, 1))
assert_size_stride(primals_2, (5,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 5), (5, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 5), (5, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 5), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 5), (80, 20, 5, 1), 0)
del buf0
buf3 = empty_strided_cuda((4, 4, 4, 5), (80, 20, 5, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(320)](buf1,
primals_2, buf3, 320, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 5), (
5, 1), 0), reinterpret_tensor(primals_4, (5, 4), (1, 5), 0),
alpha=1, beta=1, out=buf2)
del primals_5
return reinterpret_tensor(buf2, (256,), (1,), 0), reinterpret_tensor(
primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 5), (
5, 1), 0), primals_4, buf3
class FCNetNew(nn.Module):
def __init__(self, input_size, output_size):
super().__init__()
self.l1 = nn.Linear(input_size, 5)
self.relu = nn.ReLU()
self.l2 = nn.Linear(5, output_size)
def forward(self, input_0):
primals_1 = self.l1.weight
primals_2 = self.l1.bias
primals_4 = self.l2.weight
primals_5 = self.l2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| Markus92/nni | FCNet | false | 5,581 | [
"MIT"
] | 1 | 2641c7343f4b411b002bea4f5648941268194ed7 | https://github.com/Markus92/nni/tree/2641c7343f4b411b002bea4f5648941268194ed7 | import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
class Model(nn.Module):
def __init__(self, input_size, output_size):
super().__init__()
self.l1 = nn.Linear(input_size, 5)
self.relu = nn.ReLU()
self.l2 = nn.Linear(5, output_size)
def forward(self, x):
output = self.l1(x)
output = self.relu(output)
output = self.l2(output)
return output.view(-1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
PFLDLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/nx/cnxzajv6akmfdny3r4ud7m5las64tc7zew2ydgeuizr3yk7g7cfn.py
# Topologically Sorted Source Nodes: [sub, cos, sub_1, weight_angle, sub_2, pow_1, l2_distant, mul, mean, mean_1], Original ATen: [aten.sub, aten.cos, aten.rsub, aten.sum, aten.pow, aten.mul, aten.mean]
# Source node to ATen node mapping:
# cos => cos
# l2_distant => sum_2
# mean => mean
# mean_1 => mean_1
# mul => mul
# pow_1 => pow_1
# sub => sub
# sub_1 => sub_1
# sub_2 => sub_2
# weight_angle => sum_1
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %cos : [num_users=1] = call_function[target=torch.ops.aten.cos.default](args = (%sub,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %cos), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%sub_1, [1]), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg2_1, %arg3_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub_2, 2), kwargs = {})
# %sum_2 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [1]), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_1, %sum_2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%mul,), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sum_2,), kwargs = {})
triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0 = async_compile.triton('triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {6: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 7), equal_to_1=(6,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1'], 'no_x_dim': False, 'num_load': 16, 'num_reduction': 2, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0(in_out_ptr0, in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = (rindex // 16)
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0 + (64*r1)), None)
tmp1 = tl.load(in_ptr1 + (r0 + (64*r1)), None)
tmp4 = tl.load(in_ptr0 + (16 + r0 + (64*r1)), None)
tmp5 = tl.load(in_ptr1 + (16 + r0 + (64*r1)), None)
tmp9 = tl.load(in_ptr0 + (32 + r0 + (64*r1)), None)
tmp10 = tl.load(in_ptr1 + (32 + r0 + (64*r1)), None)
tmp14 = tl.load(in_ptr0 + (48 + r0 + (64*r1)), None)
tmp15 = tl.load(in_ptr1 + (48 + r0 + (64*r1)), None)
tmp19 = tl.load(in_ptr2 + (r0 + (64*r1)), None)
tmp20 = tl.load(in_ptr3 + (r0 + (64*r1)), None)
tmp25 = tl.load(in_ptr2 + (16 + r0 + (64*r1)), None)
tmp26 = tl.load(in_ptr3 + (16 + r0 + (64*r1)), None)
tmp31 = tl.load(in_ptr2 + (32 + r0 + (64*r1)), None)
tmp32 = tl.load(in_ptr3 + (32 + r0 + (64*r1)), None)
tmp37 = tl.load(in_ptr2 + (48 + r0 + (64*r1)), None)
tmp38 = tl.load(in_ptr3 + (48 + r0 + (64*r1)), None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp6 = tmp4 - tmp5
tmp7 = tmp6 * tmp6
tmp8 = tmp3 + tmp7
tmp11 = tmp9 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tmp8 + tmp12
tmp16 = tmp14 - tmp15
tmp17 = tmp16 * tmp16
tmp18 = tmp13 + tmp17
tmp21 = tmp19 - tmp20
tmp22 = tl_math.cos(tmp21)
tmp23 = 1.0
tmp24 = tmp23 - tmp22
tmp27 = tmp25 - tmp26
tmp28 = tl_math.cos(tmp27)
tmp29 = tmp23 - tmp28
tmp30 = tmp24 + tmp29
tmp33 = tmp31 - tmp32
tmp34 = tl_math.cos(tmp33)
tmp35 = tmp23 - tmp34
tmp36 = tmp30 + tmp35
tmp39 = tmp37 - tmp38
tmp40 = tl_math.cos(tmp39)
tmp41 = tmp23 - tmp40
tmp42 = tmp36 + tmp41
tmp43 = tmp42 * tmp18
tmp44 = tl.broadcast_to(tmp43, [XBLOCK, RBLOCK])
tmp46 = tl.sum(tmp44, 1)[:, None]
tmp47 = tl.broadcast_to(tmp18, [XBLOCK, RBLOCK])
tmp49 = tl.sum(tmp47, 1)[:, None]
tmp50 = 64.0
tmp51 = tmp46 / tmp50
tmp52 = tmp49 / tmp50
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp51, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp52, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1, arg2_1, arg3_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg3_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf2; del buf2 # reuse
buf5 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [sub, cos, sub_1, weight_angle, sub_2, pow_1, l2_distant, mul, mean, mean_1], Original ATen: [aten.sub, aten.cos, aten.rsub, aten.sum, aten.pow, aten.mul, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0.run(buf4, buf5, arg2_1, arg3_1, arg0_1, arg1_1, 1, 64, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
del arg2_1
del arg3_1
return (buf4, buf5, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg2_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg3_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1, arg2_1, arg3_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
class PFLDLoss(nn.Module):
"""Weighted loss of L2 distance with the pose angle for PFLD."""
def __init__(self):
super(PFLDLoss, self).__init__()
def forward(self, landmark_gt, euler_angle_gt, angle, landmarks):
"""
Calculate weighted L2 loss for PFLD.
Parameters
----------
landmark_gt : tensor
the ground truth of landmarks
euler_angle_gt : tensor
the ground truth of pose angle
angle : tensor
the predicted pose angle
landmarks : float32
the predicted landmarks
Returns
-------
output: tensor
the weighted L2 loss
output: tensor
the normal L2 loss
"""
weight_angle = torch.sum(1 - torch.cos(angle - euler_angle_gt), axis=1)
l2_distant = torch.sum((landmark_gt - landmarks) ** 2, axis=1)
return torch.mean(weight_angle * l2_distant), torch.mean(l2_distant)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0(in_out_ptr0,
in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel, XBLOCK:
tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp1 = tl.load(in_ptr1 + (r0 + 64 * r1), None)
tmp4 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr1 + (16 + r0 + 64 * r1), None)
tmp9 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp10 = tl.load(in_ptr1 + (32 + r0 + 64 * r1), None)
tmp14 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp15 = tl.load(in_ptr1 + (48 + r0 + 64 * r1), None)
tmp19 = tl.load(in_ptr2 + (r0 + 64 * r1), None)
tmp20 = tl.load(in_ptr3 + (r0 + 64 * r1), None)
tmp25 = tl.load(in_ptr2 + (16 + r0 + 64 * r1), None)
tmp26 = tl.load(in_ptr3 + (16 + r0 + 64 * r1), None)
tmp31 = tl.load(in_ptr2 + (32 + r0 + 64 * r1), None)
tmp32 = tl.load(in_ptr3 + (32 + r0 + 64 * r1), None)
tmp37 = tl.load(in_ptr2 + (48 + r0 + 64 * r1), None)
tmp38 = tl.load(in_ptr3 + (48 + r0 + 64 * r1), None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp6 = tmp4 - tmp5
tmp7 = tmp6 * tmp6
tmp8 = tmp3 + tmp7
tmp11 = tmp9 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tmp8 + tmp12
tmp16 = tmp14 - tmp15
tmp17 = tmp16 * tmp16
tmp18 = tmp13 + tmp17
tmp21 = tmp19 - tmp20
tmp22 = tl_math.cos(tmp21)
tmp23 = 1.0
tmp24 = tmp23 - tmp22
tmp27 = tmp25 - tmp26
tmp28 = tl_math.cos(tmp27)
tmp29 = tmp23 - tmp28
tmp30 = tmp24 + tmp29
tmp33 = tmp31 - tmp32
tmp34 = tl_math.cos(tmp33)
tmp35 = tmp23 - tmp34
tmp36 = tmp30 + tmp35
tmp39 = tmp37 - tmp38
tmp40 = tl_math.cos(tmp39)
tmp41 = tmp23 - tmp40
tmp42 = tmp36 + tmp41
tmp43 = tmp42 * tmp18
tmp44 = tl.broadcast_to(tmp43, [XBLOCK, RBLOCK])
tmp46 = tl.sum(tmp44, 1)[:, None]
tmp47 = tl.broadcast_to(tmp18, [XBLOCK, RBLOCK])
tmp49 = tl.sum(tmp47, 1)[:, None]
tmp50 = 64.0
tmp51 = tmp46 / tmp50
tmp52 = tmp49 / tmp50
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp51, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + tl.full([XBLOCK, 1], 0, tl.int32), tmp52, None)
def call(args):
arg0_1, arg1_1, arg2_1, arg3_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg3_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf2
del buf2
buf5 = buf3
del buf3
get_raw_stream(0)
triton_per_fused_cos_mean_mul_pow_rsub_sub_sum_0[grid(1)](buf4,
buf5, arg2_1, arg3_1, arg0_1, arg1_1, 1, 64, XBLOCK=1,
num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
del arg3_1
return buf4, buf5
class PFLDLossNew(nn.Module):
"""Weighted loss of L2 distance with the pose angle for PFLD."""
def __init__(self):
super(PFLDLossNew, self).__init__()
def forward(self, input_0, input_1, input_2, input_3):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
arg3_1 = input_3
output = call([arg0_1, arg1_1, arg2_1, arg3_1])
return output[0], output[1]
| Markus92/nni | PFLDLoss | false | 5,582 | [
"MIT"
] | 1 | 2641c7343f4b411b002bea4f5648941268194ed7 | https://github.com/Markus92/nni/tree/2641c7343f4b411b002bea4f5648941268194ed7 | import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from typing import *
class Model(nn.Module):
"""Weighted loss of L2 distance with the pose angle for PFLD."""
def __init__(self):
super().__init__()
def forward(self, landmark_gt, euler_angle_gt, angle, landmarks):
"""
Calculate weighted L2 loss for PFLD.
Parameters
----------
landmark_gt : tensor
the ground truth of landmarks
euler_angle_gt : tensor
the ground truth of pose angle
angle : tensor
the predicted pose angle
landmarks : float32
the predicted landmarks
Returns
-------
output: tensor
the weighted L2 loss
output: tensor
the normal L2 loss
"""
weight_angle = torch.sum(1 - torch.cos(angle - euler_angle_gt), axis=1)
l2_distant = torch.sum((landmark_gt - landmarks) ** 2, axis=1)
return torch.mean(weight_angle * l2_distant), torch.mean(l2_distant)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
VarifocalLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ax/caxbx6i4bwtsbtfhik5fb25fzuvlc7qeuptufx227j5xebtapxg3.py
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits, gt, float_1, mul, pred_sigmoid, sub, abs_1, pow_1, mul_1, le, float_2, mul_2, focal_weight, loss, loss_1, loss_cls], Original ATen: [aten.binary_cross_entropy_with_logits, aten.gt, aten._to_copy, aten.mul, aten.sigmoid, aten.sub, aten.abs, aten.pow, aten.le, aten.add, aten.mean]
# Source node to ATen node mapping:
# abs_1 => abs_1
# binary_cross_entropy_with_logits => abs_2, exp, full_default, log1p, minimum, mul_3, neg, sub_1, sub_2, sub_3
# float_1 => convert_element_type
# float_2 => convert_element_type_1
# focal_weight => add
# gt => gt
# le => le
# loss => mul_4
# loss_1 => mean
# loss_cls => mul_5
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# pow_1 => pow_1
# pred_sigmoid => sigmoid
# sub => sub
# Graph fragment:
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg1_1), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %arg0_1), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %minimum : [num_users=1] = call_function[target=torch.ops.aten.minimum.default](args = (%full_default, %arg0_1), kwargs = {})
# %abs_2 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%arg0_1,), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%abs_2,), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%neg,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%minimum, %log1p), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_3, %sub_2), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%arg1_1, 0.0), kwargs = {})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%gt, torch.float32), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg1_1, %convert_element_type), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%arg0_1,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%sigmoid, %arg1_1), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%sub,), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%abs_1, 2.0), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_1, 0.75), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%arg1_1, 0.0), kwargs = {})
# %convert_element_type_1 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%le, torch.float32), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_1, %convert_element_type_1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %mul_2), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_3, %add), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%mul_4,), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mean, 1.0), kwargs = {})
triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0 = async_compile.triton('triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp3 = tl.load(in_ptr1 + (r0), None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = tmp0 > tmp5
tmp14 = tmp13.to(tl.float32)
tmp15 = tmp0 * tmp14
tmp16 = tl.sigmoid(tmp3)
tmp17 = tmp16 - tmp0
tmp18 = tl_math.abs(tmp17)
tmp19 = tmp18 * tmp18
tmp20 = 0.75
tmp21 = tmp19 * tmp20
tmp22 = tmp0 <= tmp5
tmp23 = tmp22.to(tl.float32)
tmp24 = tmp21 * tmp23
tmp25 = tmp15 + tmp24
tmp26 = tmp12 * tmp25
tmp27 = tl.broadcast_to(tmp26, [RBLOCK])
tmp29 = triton_helpers.promote_to_tensor(tl.sum(tmp27, 0))
tmp30 = 256.0
tmp31 = tmp29 / tmp30
tmp32 = tmp31 * tmp1
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp32, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [binary_cross_entropy_with_logits, gt, float_1, mul, pred_sigmoid, sub, abs_1, pow_1, mul_1, le, float_2, mul_2, focal_weight, loss, loss_1, loss_cls], Original ATen: [aten.binary_cross_entropy_with_logits, aten.gt, aten._to_copy, aten.mul, aten.sigmoid, aten.sub, aten.abs, aten.pow, aten.le, aten.add, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0.run(buf1, arg1_1, arg0_1, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn.functional as F
import torch.nn as nn
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def varifocal_loss(pred, target, weight=None, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', avg_factor=None):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
pred (torch.Tensor): The prediction with shape (N, C), C is the
number of classes
target (torch.Tensor): The learning target of the iou-aware
classification score with shape (N, C), C is the number of classes.
weight (torch.Tensor, optional): The weight of loss for each
prediction. Defaults to None.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal Loss.
Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive example with the iou target. Defaults to True.
reduction (str, optional): The method used to reduce the loss into
a scalar. Defaults to 'mean'. Options are "none", "mean" and
"sum".
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
"""
assert pred.size() == target.size()
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
if iou_weighted:
focal_weight = target * (target > 0.0).float() + alpha * (pred_sigmoid
- target).abs().pow(gamma) * (target <= 0.0).float()
else:
focal_weight = (target > 0.0).float() + alpha * (pred_sigmoid - target
).abs().pow(gamma) * (target <= 0.0).float()
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class VarifocalLoss(nn.Module):
def __init__(self, use_sigmoid=True, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', loss_weight=1.0):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
use_sigmoid (bool, optional): Whether the prediction is
used for sigmoid or softmax. Defaults to True.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal
Loss. Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive examples with the iou target. Defaults to True.
reduction (str, optional): The method used to reduce the loss into
a scalar. Defaults to 'mean'. Options are "none", "mean" and
"sum".
loss_weight (float, optional): Weight of loss. Defaults to 1.0.
"""
super(VarifocalLoss, self).__init__()
assert use_sigmoid is True, 'Only sigmoid varifocal loss supported now.'
assert alpha >= 0.0
self.use_sigmoid = use_sigmoid
self.alpha = alpha
self.gamma = gamma
self.iou_weighted = iou_weighted
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, pred, target, weight=None, avg_factor=None,
reduction_override=None):
"""Forward function.
Args:
pred (torch.Tensor): The prediction.
target (torch.Tensor): The learning target of the prediction.
weight (torch.Tensor, optional): The weight of loss for each
prediction. Defaults to None.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
reduction_override (str, optional): The reduction method used to
override the original reduction method of the loss.
Options are "none", "mean" and "sum".
Returns:
torch.Tensor: The calculated loss
"""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.use_sigmoid:
loss_cls = self.loss_weight * varifocal_loss(pred, target,
weight, alpha=self.alpha, gamma=self.gamma, iou_weighted=
self.iou_weighted, reduction=reduction, avg_factor=avg_factor)
else:
raise NotImplementedError
return loss_cls
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn.functional as F
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0(
in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp3 = tl.load(in_ptr1 + r0, None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = tmp0 > tmp5
tmp14 = tmp13.to(tl.float32)
tmp15 = tmp0 * tmp14
tmp16 = tl.sigmoid(tmp3)
tmp17 = tmp16 - tmp0
tmp18 = tl_math.abs(tmp17)
tmp19 = tmp18 * tmp18
tmp20 = 0.75
tmp21 = tmp19 * tmp20
tmp22 = tmp0 <= tmp5
tmp23 = tmp22.to(tl.float32)
tmp24 = tmp21 * tmp23
tmp25 = tmp15 + tmp24
tmp26 = tmp12 * tmp25
tmp27 = tl.broadcast_to(tmp26, [RBLOCK])
tmp29 = triton_helpers.promote_to_tensor(tl.sum(tmp27, 0))
tmp30 = 256.0
tmp31 = tmp29 / tmp30
tmp32 = tmp31 * tmp1
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp32, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused__to_copy_abs_add_binary_cross_entropy_with_logits_gt_le_mean_mul_pow_sigmoid_sub_0[
grid(1)](buf1, arg1_1, arg0_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def varifocal_loss(pred, target, weight=None, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', avg_factor=None):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
pred (torch.Tensor): The prediction with shape (N, C), C is the
number of classes
target (torch.Tensor): The learning target of the iou-aware
classification score with shape (N, C), C is the number of classes.
weight (torch.Tensor, optional): The weight of loss for each
prediction. Defaults to None.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal Loss.
Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive example with the iou target. Defaults to True.
reduction (str, optional): The method used to reduce the loss into
a scalar. Defaults to 'mean'. Options are "none", "mean" and
"sum".
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
"""
assert pred.size() == target.size()
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
if iou_weighted:
focal_weight = target * (target > 0.0).float() + alpha * (pred_sigmoid
- target).abs().pow(gamma) * (target <= 0.0).float()
else:
focal_weight = (target > 0.0).float() + alpha * (pred_sigmoid - target
).abs().pow(gamma) * (target <= 0.0).float()
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class VarifocalLossNew(nn.Module):
def __init__(self, use_sigmoid=True, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', loss_weight=1.0):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
use_sigmoid (bool, optional): Whether the prediction is
used for sigmoid or softmax. Defaults to True.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal
Loss. Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive examples with the iou target. Defaults to True.
reduction (str, optional): The method used to reduce the loss into
a scalar. Defaults to 'mean'. Options are "none", "mean" and
"sum".
loss_weight (float, optional): Weight of loss. Defaults to 1.0.
"""
super(VarifocalLossNew, self).__init__()
assert use_sigmoid is True, 'Only sigmoid varifocal loss supported now.'
assert alpha >= 0.0
self.use_sigmoid = use_sigmoid
self.alpha = alpha
self.gamma = gamma
self.iou_weighted = iou_weighted
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MatthewInkawhich/object_localization_network | VarifocalLoss | false | 5,583 | [
"Apache-2.0"
] | 1 | 3fddaacfcef33f03af48b746e95ebd7d74dbb27f | https://github.com/MatthewInkawhich/object_localization_network/tree/3fddaacfcef33f03af48b746e95ebd7d74dbb27f | import torch
import torch.nn.functional as F
import torch.nn as nn
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def varifocal_loss(pred, target, weight=None, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', avg_factor=None):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
pred (torch.Tensor): The prediction with shape (N, C), C is the
number of classes
target (torch.Tensor): The learning target of the iou-aware
classification score with shape (N, C), C is the number of classes.
weight (torch.Tensor, optional): The weight of loss for each
prediction. Defaults to None.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal Loss.
Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive example with the iou target. Defaults to True.
reduction (str, optional): The method used to reduce the loss into
a scalar. Defaults to 'mean'. Options are "none", "mean" and
"sum".
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
"""
assert pred.size() == target.size()
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
if iou_weighted:
focal_weight = target * (target > 0.0).float() + alpha * (pred_sigmoid
- target).abs().pow(gamma) * (target <= 0.0).float()
else:
focal_weight = (target > 0.0).float() + alpha * (pred_sigmoid - target
).abs().pow(gamma) * (target <= 0.0).float()
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class Model(nn.Module):
def __init__(self, use_sigmoid=True, alpha=0.75, gamma=2.0,
iou_weighted=True, reduction='mean', loss_weight=1.0):
"""`Varifocal Loss <https://arxiv.org/abs/2008.13367>`_
Args:
use_sigmoid (bool, optional): Whether the prediction is
used for sigmoid or softmax. Defaults to True.
alpha (float, optional): A balance factor for the negative part of
Varifocal Loss, which is different from the alpha of Focal
Loss. Defaults to 0.75.
gamma (float, optional): The gamma for calculating the modulating
factor. Defaults to 2.0.
iou_weighted (bool, optional): Whether to weight the loss of the
positive examples with the iou target. Defaults to True.
# ... truncated (>4000 chars) for memory efficiency |
EncoderLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/wd/cwdz7kqs3uwyg53zsyekt77eye7yjl6v7vulow2q6ni534mkf6zw.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# y => add, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_3, [2]), kwargs = {correction: 0, keepdim: True})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
triton_poi_fused_native_layer_norm_0 = async_compile.triton('triton_poi_fused_native_layer_norm_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_native_layer_norm_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + (x0), tmp8, xmask)
tl.store(out_ptr1 + (x0), tmp23, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/vs/cvsfvbs4wlaqvwxm3svg65dnhcq336ptudvn6xetnbnrtzj7xssn.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# y => add, add_1, mul, mul_1, rsqrt, sub, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_3, [2]), kwargs = {correction: 0, keepdim: True})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_3, %getitem_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %rsqrt), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %primals_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_2), kwargs = {})
triton_poi_fused_native_layer_norm_1 = async_compile.triton('triton_poi_fused_native_layer_norm_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_native_layer_norm_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_native_layer_norm_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kd/ckd3lhdfl7shyukzlxpvlilao7e5s7ib63y5ilra2gqzt6ijpwwo.py
# Topologically Sorted Source Nodes: [q_2, x], Original ATen: [aten.mul, aten.clone]
# Source node to ATen node mapping:
# q_2 => mul_2
# x => clone
# Graph fragment:
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_3, 1.0), kwargs = {})
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_mul_2 = async_compile.triton('triton_poi_fused_clone_mul_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_mul_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_mul_2(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x2 + (4*y3)), tmp4, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/3r/c3rsks6vi53ggj2qfjmhu7vc3vqskqtyr7gc4fdp74wzt6pdrjx4.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x => clone_1
# Graph fragment:
# %clone_1 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_1,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_3 = async_compile.triton('triton_poi_fused_clone_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + (4*y3)), tmp2, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/iv/civdgwzpphyda4rs4fr3g6w25bprv7bn4anqgivrgzavi7xr5pdl.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_1 => amax, exp, sub_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_11, [3], True), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_11, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_1,), kwargs = {})
triton_poi_fused__softmax_4 = async_compile.triton('triton_poi_fused__softmax_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_4(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/a4/ca4u6hbohfqkgchihihlu5hrf3vuqm27r2ncsg7xb6g4ikttl2at.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_1 => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [3], True), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_5 = async_compile.triton('triton_poi_fused__softmax_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_5(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4k/c4kxxzyxk45cygbwnqkt5vb2udxol67wuotkh6zmuwsinb63uprn.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x_4 => clone_4
# Graph fragment:
# %clone_4 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_7,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_6 = async_compile.triton('triton_poi_fused_clone_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 4], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_6(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (16*y1)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + (4*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/pw/cpw6jeu3vr4erot3bahfr3gqwa2vqailcc5xbtuiztd5l6w2zwjm.py
# Topologically Sorted Source Nodes: [x_7, y_2], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# x_7 => add_2
# y_2 => var_mean_1
# Graph fragment:
# %add_2 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %view_17), kwargs = {})
# %var_mean_1 : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add_2, [2]), kwargs = {correction: 0, keepdim: True})
triton_poi_fused_add_native_layer_norm_7 = async_compile.triton('triton_poi_fused_add_native_layer_norm_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_native_layer_norm_7', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_native_layer_norm_7(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = tmp27 / tmp15
tl.store(out_ptr0 + (x0), tmp16, xmask)
tl.store(out_ptr1 + (x0), tmp28, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/hy/chy77uudjf33jqcobrq24clniixoyn2vucadggy6c6n33mq4favb.py
# Topologically Sorted Source Nodes: [x_7, y_2], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# x_7 => add_2
# y_2 => add_3, add_4, mul_3, mul_4, rsqrt_1, sub_2
# Graph fragment:
# %add_2 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %view_17), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_2, 1e-05), kwargs = {})
# %rsqrt_1 : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_3,), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_2, %getitem_3), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_2, %rsqrt_1), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_3, %primals_12), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %primals_13), kwargs = {})
triton_poi_fused_add_native_layer_norm_8 = async_compile.triton('triton_poi_fused_add_native_layer_norm_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_native_layer_norm_8', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_native_layer_norm_8(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp3 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x1), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp6 = 1e-05
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp4 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + (x2), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dp/cdplbbjhtn7wjfs5zbdr7dqzrhv6sxravwmmbhqyrtfejnoccqhe.py
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.gelu]
# Source node to ATen node mapping:
# x_9 => add_5, erf, mul_5, mul_6, mul_7
# Graph fragment:
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_19, 0.5), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_19, 0.7071067811865476), kwargs = {})
# %erf : [num_users=1] = call_function[target=torch.ops.aten.erf.default](args = (%mul_6,), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%erf, 1), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_5, %add_5), kwargs = {})
triton_poi_fused_gelu_9 = async_compile.triton('triton_poi_fused_gelu_9', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_gelu_9', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_gelu_9(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = 0.7071067811865476
tmp4 = tmp0 * tmp3
tmp5 = libdevice.erf(tmp4)
tmp6 = 1.0
tmp7 = tmp5 + tmp6
tmp8 = tmp2 * tmp7
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5o/c5oq2dd4gv3qfn7dbi4gn37m4q5cqp3syup6457zs5qf2qor5xrv.py
# Topologically Sorted Source Nodes: [x_7, x_11], Original ATen: [aten.add]
# Source node to ATen node mapping:
# x_11 => add_6
# x_7 => add_2
# Graph fragment:
# %add_2 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %view_17), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, %view_21), kwargs = {})
triton_poi_fused_add_10 = async_compile.triton('triton_poi_fused_add_10', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_10', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_10(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp3 = tl.load(in_out_ptr0 + (x2), xmask)
tmp4 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17 = args
args.clear()
assert_size_stride(primals_1, (4, ), (1, ))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, ), (1, ))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4, ), (1, ))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4, ), (1, ))
assert_size_stride(primals_12, (4, ), (1, ))
assert_size_stride(primals_13, (4, ), (1, ))
assert_size_stride(primals_14, (4, 4), (4, 1))
assert_size_stride(primals_15, (4, ), (1, ))
assert_size_stride(primals_16, (4, 4), (4, 1))
assert_size_stride(primals_17, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf1 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_poi_fused_native_layer_norm_0.run(primals_3, buf0, buf1, 16, grid=grid(16), stream=stream0)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_1.run(primals_3, buf0, buf1, primals_1, primals_2, buf2, 64, grid=grid(64), stream=stream0)
del primals_1
del primals_2
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [q_2, x], Original ATen: [aten.mul, aten.clone]
triton_poi_fused_clone_mul_2.run(buf3, primals_5, buf6, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_5
buf7 = reinterpret_tensor(buf3, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.clone]
triton_poi_fused_clone_3.run(buf4, primals_7, buf7, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_7
buf8 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf6, (16, 4, 1), (4, 1, 0), 0), reinterpret_tensor(buf7, (16, 1, 4), (4, 0, 1), 0), out=buf8)
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_4.run(buf8, buf9, 256, grid=grid(256), stream=stream0)
buf10 = reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf8 # reuse
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_5.run(buf9, buf10, 256, grid=grid(256), stream=stream0)
del buf9
buf11 = reinterpret_tensor(buf4, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.clone]
triton_poi_fused_clone_3.run(buf5, primals_9, buf11, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_9
buf12 = reinterpret_tensor(buf5, (16, 4, 1), (4, 1, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf10, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf11, (16, 4, 1), (4, 1, 0), 0), out=buf12)
buf13 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.clone]
triton_poi_fused_clone_6.run(buf12, buf13, 16, 4, grid=grid(16, 4), stream=stream0)
buf14 = reinterpret_tensor(buf12, (16, 4), (4, 1), 0); del buf12 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf13, (16, 4), (4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf14)
del primals_11
buf15 = buf1; del buf1 # reuse
buf16 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x_7, y_2], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_7.run(primals_3, buf14, buf15, buf16, 16, grid=grid(16), stream=stream0)
buf17 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_7, y_2], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_8.run(primals_3, buf14, buf15, buf16, primals_12, primals_13, buf17, 64, grid=grid(64), stream=stream0)
del buf15
del buf16
del primals_13
buf18 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_15, reinterpret_tensor(buf17, (16, 4), (4, 1), 0), reinterpret_tensor(primals_14, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf18)
del primals_15
buf19 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_9], Original ATen: [aten.gelu]
triton_poi_fused_gelu_9.run(buf18, buf19, 64, grid=grid(64), stream=stream0)
buf20 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf19, (16, 4), (4, 1), 0), reinterpret_tensor(primals_16, (4, 4), (1, 4), 0), out=buf20)
buf21 = reinterpret_tensor(buf20, (4, 4, 4), (16, 4, 1), 0); del buf20 # reuse
# Topologically Sorted Source Nodes: [x_7, x_11], Original ATen: [aten.add]
triton_poi_fused_add_10.run(buf21, primals_3, buf14, primals_17, 64, grid=grid(64), stream=stream0)
del primals_17
return (buf21, primals_3, primals_12, reinterpret_tensor(buf2, (16, 4), (4, 1), 0), buf10, reinterpret_tensor(buf13, (16, 4), (4, 1), 0), buf14, reinterpret_tensor(buf17, (16, 4), (4, 1), 0), buf18, reinterpret_tensor(buf19, (16, 4), (4, 1), 0), primals_16, primals_14, primals_10, reinterpret_tensor(buf11, (16, 1, 4), (4, 1, 1), 0), reinterpret_tensor(buf6, (16, 1, 4), (4, 1, 1), 0), reinterpret_tensor(buf7, (16, 4, 1), (4, 1, 4), 0), primals_8, primals_6, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class FeedForwardNetwork(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate):
super(FeedForwardNetwork, self).__init__()
self.layer1 = nn.Linear(hidden_size, ffn_size)
self.gelu = nn.GELU()
self.layer2 = nn.Linear(ffn_size, hidden_size)
def forward(self, x):
x = self.layer1(x)
x = self.gelu(x)
x = self.layer2(x)
return x
class MultiHeadAttention(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super(MultiHeadAttention, self).__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, q, k, v, attn_bias=None):
orig_q_size = q.size()
d_k = self.att_size
d_v = self.att_size
batch_size = q.size(0)
q = self.linear_q(q).view(batch_size, -1, self.num_heads, d_k)
k = self.linear_k(k).view(batch_size, -1, self.num_heads, d_k)
v = self.linear_v(v).view(batch_size, -1, self.num_heads, d_v)
q = q.transpose(1, 2)
v = v.transpose(1, 2)
k = k.transpose(1, 2).transpose(2, 3)
q = q * self.scale
x = torch.matmul(q, k)
if attn_bias is not None:
x = x + attn_bias
x = torch.softmax(x, dim=3)
x = self.att_dropout(x)
x = x.matmul(v)
x = x.transpose(1, 2).contiguous()
x = x.view(batch_size, -1, self.num_heads * d_v)
x = self.output_layer(x)
assert x.size() == orig_q_size
return x
class EncoderLayer(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate,
attention_dropout_rate, num_heads):
super(EncoderLayer, self).__init__()
self.self_attention_norm = nn.LayerNorm(hidden_size)
self.self_attention = MultiHeadAttention(hidden_size,
attention_dropout_rate, num_heads)
self.self_attention_dropout = nn.Dropout(dropout_rate)
self.ffn_norm = nn.LayerNorm(hidden_size)
self.ffn = FeedForwardNetwork(hidden_size, ffn_size, dropout_rate)
self.ffn_dropout = nn.Dropout(dropout_rate)
def forward(self, x, attn_bias=None):
y = self.self_attention_norm(x)
y = self.self_attention(y, y, y, attn_bias)
y = self.self_attention_dropout(y)
x = x + y
y = self.ffn_norm(x)
y = self.ffn(y)
y = self.ffn_dropout(y)
x = x + y
return x
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'hidden_size': 4, 'ffn_size': 4, 'dropout_rate': 0.5,
'attention_dropout_rate': 0.5, 'num_heads': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-05
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + x0, tmp8, xmask)
tl.store(out_ptr1 + x0, tmp23, xmask)
@triton.jit
def triton_poi_fused_native_layer_norm_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3,
in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_clone_mul_2(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x2 + 4 * y3), tmp4, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 4 * y3), tmp2, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_5(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_clone_6(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_7(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = tmp27 / tmp15
tl.store(out_ptr0 + x0, tmp16, xmask)
tl.store(out_ptr1 + x0, tmp28, xmask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_8(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp6 = 1e-05
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp4 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + x2, tmp13, xmask)
@triton.jit
def triton_poi_fused_gelu_9(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = 0.7071067811865476
tmp4 = tmp0 * tmp3
tmp5 = libdevice.erf(tmp4)
tmp6 = 1.0
tmp7 = tmp5 + tmp6
tmp8 = tmp2 * tmp7
tl.store(out_ptr0 + x0, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_10(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp3 = tl.load(in_out_ptr0 + x2, xmask)
tmp4 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tl.store(in_out_ptr0 + x2, tmp6, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17) = args
args.clear()
assert_size_stride(primals_1, (4,), (1,))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4,), (1,))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (4,), (1,))
assert_size_stride(primals_13, (4,), (1,))
assert_size_stride(primals_14, (4, 4), (4, 1))
assert_size_stride(primals_15, (4,), (1,))
assert_size_stride(primals_16, (4, 4), (4, 1))
assert_size_stride(primals_17, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf1 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(16)](primals_3, buf0,
buf1, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_native_layer_norm_1[grid(64)](primals_3, buf0,
buf1, primals_1, primals_2, buf2, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del primals_1
del primals_2
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
triton_poi_fused_clone_mul_2[grid(16, 4)](buf3, primals_5, buf6, 16,
4, XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
del primals_5
buf7 = reinterpret_tensor(buf3, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf3
triton_poi_fused_clone_3[grid(16, 4)](buf4, primals_7, buf7, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf8 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf6, (16, 4, 1), (4, 1, 0),
0), reinterpret_tensor(buf7, (16, 1, 4), (4, 0, 1), 0), out=buf8)
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_4[grid(256)](buf8, buf9, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf10 = reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf8
triton_poi_fused__softmax_5[grid(256)](buf9, buf10, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf9
buf11 = reinterpret_tensor(buf4, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf4
triton_poi_fused_clone_3[grid(16, 4)](buf5, primals_9, buf11, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_9
buf12 = reinterpret_tensor(buf5, (16, 4, 1), (4, 1, 1), 0)
del buf5
extern_kernels.bmm(reinterpret_tensor(buf10, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf11, (16, 4, 1), (4, 1, 0), 0), out=buf12)
buf13 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
triton_poi_fused_clone_6[grid(16, 4)](buf12, buf13, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf14 = reinterpret_tensor(buf12, (16, 4), (4, 1), 0)
del buf12
extern_kernels.addmm(primals_11, reinterpret_tensor(buf13, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf14)
del primals_11
buf15 = buf1
del buf1
buf16 = buf0
del buf0
triton_poi_fused_add_native_layer_norm_7[grid(16)](primals_3, buf14,
buf15, buf16, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf17 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_native_layer_norm_8[grid(64)](primals_3, buf14,
buf15, buf16, primals_12, primals_13, buf17, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf15
del buf16
del primals_13
buf18 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_15, reinterpret_tensor(buf17, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_14, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf18)
del primals_15
buf19 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_gelu_9[grid(64)](buf18, buf19, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf20 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf19, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_16, (4, 4), (1, 4), 0), out=buf20)
buf21 = reinterpret_tensor(buf20, (4, 4, 4), (16, 4, 1), 0)
del buf20
triton_poi_fused_add_10[grid(64)](buf21, primals_3, buf14,
primals_17, 64, XBLOCK=64, num_warps=1, num_stages=1)
del primals_17
return buf21, primals_3, primals_12, reinterpret_tensor(buf2, (16, 4),
(4, 1), 0), buf10, reinterpret_tensor(buf13, (16, 4), (4, 1), 0
), buf14, reinterpret_tensor(buf17, (16, 4), (4, 1), 0
), buf18, reinterpret_tensor(buf19, (16, 4), (4, 1), 0
), primals_16, primals_14, primals_10, reinterpret_tensor(buf11, (
16, 1, 4), (4, 1, 1), 0), reinterpret_tensor(buf6, (16, 1, 4), (4,
1, 1), 0), reinterpret_tensor(buf7, (16, 4, 1), (4, 1, 4), 0
), primals_8, primals_6, primals_4
class FeedForwardNetwork(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate):
super(FeedForwardNetwork, self).__init__()
self.layer1 = nn.Linear(hidden_size, ffn_size)
self.gelu = nn.GELU()
self.layer2 = nn.Linear(ffn_size, hidden_size)
def forward(self, x):
x = self.layer1(x)
x = self.gelu(x)
x = self.layer2(x)
return x
class MultiHeadAttention(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super(MultiHeadAttention, self).__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, q, k, v, attn_bias=None):
orig_q_size = q.size()
d_k = self.att_size
d_v = self.att_size
batch_size = q.size(0)
q = self.linear_q(q).view(batch_size, -1, self.num_heads, d_k)
k = self.linear_k(k).view(batch_size, -1, self.num_heads, d_k)
v = self.linear_v(v).view(batch_size, -1, self.num_heads, d_v)
q = q.transpose(1, 2)
v = v.transpose(1, 2)
k = k.transpose(1, 2).transpose(2, 3)
q = q * self.scale
x = torch.matmul(q, k)
if attn_bias is not None:
x = x + attn_bias
x = torch.softmax(x, dim=3)
x = self.att_dropout(x)
x = x.matmul(v)
x = x.transpose(1, 2).contiguous()
x = x.view(batch_size, -1, self.num_heads * d_v)
x = self.output_layer(x)
assert x.size() == orig_q_size
return x
class EncoderLayerNew(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate,
attention_dropout_rate, num_heads):
super(EncoderLayerNew, self).__init__()
self.self_attention_norm = nn.LayerNorm(hidden_size)
self.self_attention = MultiHeadAttention(hidden_size,
attention_dropout_rate, num_heads)
self.self_attention_dropout = nn.Dropout(dropout_rate)
self.ffn_norm = nn.LayerNorm(hidden_size)
self.ffn = FeedForwardNetwork(hidden_size, ffn_size, dropout_rate)
self.ffn_dropout = nn.Dropout(dropout_rate)
def forward(self, input_0):
primals_1 = self.self_attention_norm.weight
primals_2 = self.self_attention_norm.bias
primals_4 = self.self_attention.linear_q.weight
primals_5 = self.self_attention.linear_q.bias
primals_6 = self.self_attention.linear_k.weight
primals_7 = self.self_attention.linear_k.bias
primals_8 = self.self_attention.linear_v.weight
primals_9 = self.self_attention.linear_v.bias
primals_10 = self.self_attention.output_layer.weight
primals_11 = self.self_attention.output_layer.bias
primals_12 = self.ffn_norm.weight
primals_13 = self.ffn_norm.bias
primals_14 = self.ffn.layer1.weight
primals_15 = self.ffn.layer1.bias
primals_16 = self.ffn.layer2.weight
primals_17 = self.ffn.layer2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17])
return output[0]
| Luo-Chang/Graphormer | EncoderLayer | false | 5,584 | [
"MIT"
] | 1 | b35b3ca6369e25cdae80e1617bfc3921feeb3158 | https://github.com/Luo-Chang/Graphormer/tree/b35b3ca6369e25cdae80e1617bfc3921feeb3158 | import torch
import torch.nn as nn
class FeedForwardNetwork(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate):
super().__init__()
self.layer1 = nn.Linear(hidden_size, ffn_size)
self.gelu = nn.GELU()
self.layer2 = nn.Linear(ffn_size, hidden_size)
def forward(self, x):
x = self.layer1(x)
x = self.gelu(x)
x = self.layer2(x)
return x
class MultiHeadAttention(nn.Module):
def __init__(self, hidden_size, attention_dropout_rate, num_heads):
super().__init__()
self.num_heads = num_heads
self.att_size = att_size = hidden_size // num_heads
self.scale = att_size ** -0.5
self.linear_q = nn.Linear(hidden_size, num_heads * att_size)
self.linear_k = nn.Linear(hidden_size, num_heads * att_size)
self.linear_v = nn.Linear(hidden_size, num_heads * att_size)
self.att_dropout = nn.Dropout(attention_dropout_rate)
self.output_layer = nn.Linear(num_heads * att_size, hidden_size)
def forward(self, q, k, v, attn_bias=None):
orig_q_size = q.size()
d_k = self.att_size
d_v = self.att_size
batch_size = q.size(0)
q = self.linear_q(q).view(batch_size, -1, self.num_heads, d_k)
k = self.linear_k(k).view(batch_size, -1, self.num_heads, d_k)
v = self.linear_v(v).view(batch_size, -1, self.num_heads, d_v)
q = q.transpose(1, 2)
v = v.transpose(1, 2)
k = k.transpose(1, 2).transpose(2, 3)
q = q * self.scale
x = torch.matmul(q, k)
if attn_bias is not None:
x = x + attn_bias
x = torch.softmax(x, dim=3)
x = self.att_dropout(x)
x = x.matmul(v)
x = x.transpose(1, 2).contiguous()
x = x.view(batch_size, -1, self.num_heads * d_v)
x = self.output_layer(x)
assert x.size() == orig_q_size
return x
class Model(nn.Module):
def __init__(self, hidden_size, ffn_size, dropout_rate,
attention_dropout_rate, num_heads):
super().__init__()
self.self_attention_norm = nn.LayerNorm(hidden_size)
self.self_attention = MultiHeadAttention(hidden_size,
attention_dropout_rate, num_heads)
self.self_attention_dropout = nn.Dropout(dropout_rate)
self.ffn_norm = nn.LayerNorm(hidden_size)
self.ffn = FeedForwardNetwork(hidden_size, ffn_size, dropout_rate)
self.ffn_dropout = nn.Dropout(dropout_rate)
def forward(self, x, attn_bias=None):
y = self.self_attention_norm(x)
y = self.self_attention(y, y, y, attn_bias)
y = self.self_attention_dropout(y)
x = x + y
y = self.ffn_norm(x)
y = self.ffn(y)
y = self.ffn_dropout(y)
x = x + y
return x
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'hidden_size': 4, 'ffn_size': 4, 'dropout_rate': 0.5,
'attention_dropout_rate': 0.5, 'num_heads': 4}]
|
Net | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/4d/c4d7os35bf4bckecmik4nlyqqsirmteh4sh3yxnab5lmuntnmwk2.py
# Unsorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
triton_poi_fused_0 = async_compile.triton('triton_poi_fused_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128, 16], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 128
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = (yindex // 4)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (4*x2) + (36*y1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/j5/cj5nf2owtsdm2zwcezqxpyn63iwddjyadpotkhm2ua52inoqxdcl.py
# Unsorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
triton_poi_fused_1 = async_compile.triton('triton_poi_fused_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 16], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = (yindex // 4)
tmp0 = tl.load(in_ptr0 + (x2 + (16*y3)), xmask & ymask)
tl.store(out_ptr0 + (y0 + (4*x2) + (64*y1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/wv/cwvtp6qflpb42kxrujmda5zselv7wvkz3fgp2tryo2ftsisaildr.py
# Unsorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
triton_poi_fused_2 = async_compile.triton('triton_poi_fused_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048, 16], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 2048
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 32
y1 = (yindex // 32)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (32*x2) + (288*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/nw/cnwm6ljuusoqjcwr2jdx6p2ue7ldghxjdr3oe62stiuqhsboiczy.py
# Unsorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
triton_poi_fused_3 = async_compile.triton('triton_poi_fused_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[8192, 16], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 8192
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 64
y1 = (yindex // 64)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (64*x2) + (576*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/tl/ctlxctn7eg6nwvpdhdhyqadp63cm2ogdwxsotfynexn2zw62nfbb.py
# Topologically Sorted Source Nodes: [conv2d, a1], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# a1 => relu
# conv2d => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_convolution_relu_4 = async_compile.triton('triton_poi_fused_convolution_relu_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_4', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2048
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ku/ckuw5gg26ddjp4n4da74yttcx6jxcy2y4vb2npxdoq42pzni2oot.py
# Topologically Sorted Source Nodes: [conv2d_1, a2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# a2 => relu_1
# conv2d_1 => convolution_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu, %primals_4, %primals_5, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_1,), kwargs = {})
triton_poi_fused_convolution_relu_5 = async_compile.triton('triton_poi_fused_convolution_relu_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_5(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/m3/cm3haovccm7lav2s6wgp3wthu7in42r335z2o7yva4d7olh5begj.py
# Topologically Sorted Source Nodes: [conv2d_2, a3], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# a3 => relu_2
# conv2d_2 => convolution_2
# Graph fragment:
# %convolution_2 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_1, %primals_6, %primals_7, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_2 : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_2,), kwargs = {})
triton_poi_fused_convolution_relu_6 = async_compile.triton('triton_poi_fused_convolution_relu_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[8192],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_6', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_6(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ir/cirx6nbkabstacj3yb3umtzb7ustxzn5ha5etdpsewqc2v53x42u.py
# Topologically Sorted Source Nodes: [conv2d_3, p1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_3 => convolution_3
# p1 => relu_3
# Graph fragment:
# %convolution_3 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_2, %primals_8, %primals_9, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_3 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_3,), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_3, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_7 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 16], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i1', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_7', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_7(in_ptr0, in_ptr1, out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (64*y1)), xmask & ymask)
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(out_ptr0 + (x2 + (16*y3)), tmp4, xmask & ymask)
tl.store(out_ptr1 + (y0 + (4*x2) + (64*y1)), tmp6, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/2i/c2ijyqlcmcsnugr7aonjxredzbtnlbckjkwduhz675ye4ugjebkw.py
# Topologically Sorted Source Nodes: [p_out], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# p_out => amax, div, exp, sub, sum_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%addmm, [0], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%addmm, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [0], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_8 = async_compile.triton('triton_poi_fused__softmax_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_8', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_8(in_out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), xmask)
tmp1 = tmp0 - tmp0
tmp2 = tl_math.exp(tmp1)
tmp3 = tmp2 / tmp2
tl.store(in_out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/er/cerrhd6dfklfkghffr4w4v6k4tkknpp6pjf2fpylkzd3qma7oygl.py
# Topologically Sorted Source Nodes: [conv2d_4, v1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_4 => convolution_4
# v1 => relu_4
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_2, %primals_12, %primals_13, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_4 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_4,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_4, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_9 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_9', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[8, 16], tile_hint=TileHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i1', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_9', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_9(in_ptr0, in_ptr1, out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 8
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 2
y1 = (yindex // 2)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (2*x2) + (32*y1)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(out_ptr0 + (x2 + (16*y3)), tmp4, xmask & ymask)
tl.store(out_ptr1 + (y0 + (2*x2) + (32*y1)), tmp6, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ac/cacwmday2hjmsf4z7tj45ooyukpcmpul6vx42cwcasvjda6tvfr5.py
# Topologically Sorted Source Nodes: [v_out], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# v_out => tanh
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_15), kwargs = {})
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%add_tensor,), kwargs = {})
triton_poi_fused_tanh_10 = async_compile.triton('triton_poi_fused_tanh_10', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_tanh_10', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_tanh_10(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
tmp0 = tl.load(in_out_ptr0 + (0))
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr0 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
tmp4 = tmp1 + tmp3
tmp5 = libdevice.tanh(tmp4)
tl.store(in_out_ptr0 + (tl.full([XBLOCK], 0, tl.int32)), tmp5, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15 = args
args.clear()
assert_size_stride(primals_1, (32, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (32, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (64, 32, 3, 3), (288, 9, 3, 1))
assert_size_stride(primals_5, (64, ), (1, ))
assert_size_stride(primals_6, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (128, ), (1, ))
assert_size_stride(primals_8, (4, 128, 1, 1), (128, 1, 1, 1))
assert_size_stride(primals_9, (4, ), (1, ))
assert_size_stride(primals_10, (64, 256), (256, 1))
assert_size_stride(primals_11, (64, ), (1, ))
assert_size_stride(primals_12, (2, 128, 1, 1), (128, 1, 1, 1))
assert_size_stride(primals_13, (2, ), (1, ))
assert_size_stride(primals_14, (1, 128), (128, 1))
assert_size_stride(primals_15, (1, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((32, 4, 3, 3), (36, 1, 12, 4), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(primals_1, buf0, 128, 9, grid=grid(128, 9), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(primals_3, buf1, 16, 16, grid=grid(16, 16), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((64, 32, 3, 3), (288, 1, 96, 32), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(primals_4, buf2, 2048, 9, grid=grid(2048, 9), stream=stream0)
del primals_4
buf3 = empty_strided_cuda((128, 64, 3, 3), (576, 1, 192, 64), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(primals_6, buf3, 8192, 9, grid=grid(8192, 9), stream=stream0)
del primals_6
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf1, buf0, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 32, 4, 4), (512, 1, 128, 32))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [conv2d, a1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_4.run(buf5, primals_2, 2048, grid=grid(2048), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, buf2, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 4, 4), (1024, 1, 256, 64))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, a2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_5.run(buf7, primals_5, 4096, grid=grid(4096), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, buf3, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 128, 4, 4), (2048, 1, 512, 128))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [conv2d_2, a3], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_6.run(buf9, primals_7, 8192, grid=grid(8192), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_8, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 4, 4, 4), (64, 1, 16, 4))
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf19 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_3, p1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_7.run(buf10, primals_9, buf11, buf19, 16, 16, grid=grid(16, 16), stream=stream0)
del buf10
del primals_9
buf12 = empty_strided_cuda((1, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf11, (1, 256), (0, 1), 0), reinterpret_tensor(primals_10, (256, 64), (1, 256), 0), alpha=1, beta=1, out=buf12)
del primals_11
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [p_out], Original ATen: [aten._softmax]
triton_poi_fused__softmax_8.run(buf13, 64, grid=grid(64), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf9, primals_12, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 2, 4, 4), (32, 1, 8, 2))
buf15 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
buf18 = empty_strided_cuda((4, 2, 4, 4), (32, 1, 8, 2), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_4, v1], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_9.run(buf14, primals_13, buf15, buf18, 8, 16, grid=grid(8, 16), stream=stream0)
del buf14
del primals_13
buf16 = empty_strided_cuda((1, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf15, (1, 128), (0, 1), 0), reinterpret_tensor(primals_14, (128, 1), (1, 128), 0), out=buf16)
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [v_out], Original ATen: [aten.tanh]
triton_poi_fused_tanh_10.run(buf17, primals_15, 1, grid=grid(1), stream=stream0)
del primals_15
return (buf13, buf17, buf0, buf1, buf2, buf3, primals_8, primals_12, buf5, buf7, buf9, reinterpret_tensor(buf11, (1, 256), (256, 1), 0), buf13, reinterpret_tensor(buf15, (1, 128), (128, 1), 0), buf17, primals_14, buf18, primals_10, buf19, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((32, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((64, 32, 3, 3), (288, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((128, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, 128, 1, 1), (128, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((64, 256), (256, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((2, 128, 1, 1), (128, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((1, 128), (128, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((1, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class Net(nn.Module):
def __init__(self):
super(Net, self).__init__()
self.conv1 = nn.Conv2d(4, 32, 3, padding=1)
self.conv2 = nn.Conv2d(32, 64, 3, padding=1)
self.conv3 = nn.Conv2d(64, 128, 3, padding=1)
self.p_fc1 = nn.Conv2d(128, 4, 1)
self.p_fc2 = nn.Linear(4 * 8 * 8, 64)
self.v_fc1 = nn.Conv2d(128, 2, 1)
self.v_fc2 = nn.Linear(2 * 8 * 8, 1)
def forward(self, x):
a1 = F.relu(self.conv1(x))
a2 = F.relu(self.conv2(a1))
a3 = F.relu(self.conv3(a2))
p1 = F.relu(self.p_fc1(a3))
p_act = p1.view(-1, 4 * 8 * 8)
p_out = F.softmax(self.p_fc2(p_act), dim=0)
v1 = F.relu(self.v_fc1(a3))
v_act = v1.view(-1, 2 * 8 * 8)
v_out = torch.tanh(self.v_fc2(v_act))
return p_out, v_out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 128
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask & ymask, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 4 * x2 + 36 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 16 * y3), xmask & ymask)
tl.store(out_ptr0 + (y0 + 4 * x2 + 64 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 32
y1 = yindex // 32
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 32 * x2 + 288 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 64
y1 = yindex // 64
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 64 * x2 + 576 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_5(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_6(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_7(in_ptr0, in_ptr1,
out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.
constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 64 * y1), xmask & ymask)
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(out_ptr0 + (x2 + 16 * y3), tmp4, xmask & ymask)
tl.store(out_ptr1 + (y0 + 4 * x2 + 64 * y1), tmp6, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_8(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tmp0 - tmp0
tmp2 = tl_math.exp(tmp1)
tmp3 = tmp2 / tmp2
tl.store(in_out_ptr0 + x0, tmp3, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_9(in_ptr0, in_ptr1,
out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.
constexpr):
ynumel = 8
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 2
y1 = yindex // 2
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 2 * x2 + 32 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(out_ptr0 + (x2 + 16 * y3), tmp4, xmask & ymask)
tl.store(out_ptr1 + (y0 + 2 * x2 + 32 * y1), tmp6, xmask & ymask)
@triton.jit
def triton_poi_fused_tanh_10(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
tmp0 = tl.load(in_out_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr0 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
tmp4 = tmp1 + tmp3
tmp5 = libdevice.tanh(tmp4)
tl.store(in_out_ptr0 + tl.full([XBLOCK], 0, tl.int32), tmp5, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15) = args
args.clear()
assert_size_stride(primals_1, (32, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (64, 32, 3, 3), (288, 9, 3, 1))
assert_size_stride(primals_5, (64,), (1,))
assert_size_stride(primals_6, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (128,), (1,))
assert_size_stride(primals_8, (4, 128, 1, 1), (128, 1, 1, 1))
assert_size_stride(primals_9, (4,), (1,))
assert_size_stride(primals_10, (64, 256), (256, 1))
assert_size_stride(primals_11, (64,), (1,))
assert_size_stride(primals_12, (2, 128, 1, 1), (128, 1, 1, 1))
assert_size_stride(primals_13, (2,), (1,))
assert_size_stride(primals_14, (1, 128), (128, 1))
assert_size_stride(primals_15, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((32, 4, 3, 3), (36, 1, 12, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(128, 9)](primals_1, buf0, 128, 9, XBLOCK=16,
YBLOCK=64, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.float32)
triton_poi_fused_1[grid(16, 16)](primals_3, buf1, 16, 16, XBLOCK=16,
YBLOCK=16, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((64, 32, 3, 3), (288, 1, 96, 32), torch.
float32)
triton_poi_fused_2[grid(2048, 9)](primals_4, buf2, 2048, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((128, 64, 3, 3), (576, 1, 192, 64), torch
.float32)
triton_poi_fused_3[grid(8192, 9)](primals_6, buf3, 8192, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_6
buf4 = extern_kernels.convolution(buf1, buf0, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 32, 4, 4), (512, 1, 128, 32))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_4[grid(2048)](buf5, primals_2,
2048, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf6 = extern_kernels.convolution(buf5, buf2, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 4, 4), (1024, 1, 256, 64))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_5[grid(4096)](buf7, primals_5,
4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf8 = extern_kernels.convolution(buf7, buf3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 128, 4, 4), (2048, 1, 512, 128))
buf9 = buf8
del buf8
triton_poi_fused_convolution_relu_6[grid(8192)](buf9, primals_7,
8192, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf10 = extern_kernels.convolution(buf9, primals_8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 4, 4, 4), (64, 1, 16, 4))
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf19 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_7[grid(16, 16)](
buf10, primals_9, buf11, buf19, 16, 16, XBLOCK=16, YBLOCK=16,
num_warps=4, num_stages=1)
del buf10
del primals_9
buf12 = empty_strided_cuda((1, 64), (64, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf11, (1, 256),
(0, 1), 0), reinterpret_tensor(primals_10, (256, 64), (1, 256),
0), alpha=1, beta=1, out=buf12)
del primals_11
buf13 = buf12
del buf12
triton_poi_fused__softmax_8[grid(64)](buf13, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf14 = extern_kernels.convolution(buf9, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 2, 4, 4), (32, 1, 8, 2))
buf15 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
buf18 = empty_strided_cuda((4, 2, 4, 4), (32, 1, 8, 2), torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_9[grid(8, 16)](
buf14, primals_13, buf15, buf18, 8, 16, XBLOCK=16, YBLOCK=8,
num_warps=4, num_stages=1)
del buf14
del primals_13
buf16 = empty_strided_cuda((1, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf15, (1, 128), (0, 1), 0),
reinterpret_tensor(primals_14, (128, 1), (1, 128), 0), out=buf16)
buf17 = buf16
del buf16
triton_poi_fused_tanh_10[grid(1)](buf17, primals_15, 1, XBLOCK=1,
num_warps=1, num_stages=1)
del primals_15
return (buf13, buf17, buf0, buf1, buf2, buf3, primals_8, primals_12,
buf5, buf7, buf9, reinterpret_tensor(buf11, (1, 256), (256, 1), 0),
buf13, reinterpret_tensor(buf15, (1, 128), (128, 1), 0), buf17,
primals_14, buf18, primals_10, buf19)
class NetNew(nn.Module):
def __init__(self):
super(NetNew, self).__init__()
self.conv1 = nn.Conv2d(4, 32, 3, padding=1)
self.conv2 = nn.Conv2d(32, 64, 3, padding=1)
self.conv3 = nn.Conv2d(64, 128, 3, padding=1)
self.p_fc1 = nn.Conv2d(128, 4, 1)
self.p_fc2 = nn.Linear(4 * 8 * 8, 64)
self.v_fc1 = nn.Conv2d(128, 2, 1)
self.v_fc2 = nn.Linear(2 * 8 * 8, 1)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_6 = self.conv3.weight
primals_7 = self.conv3.bias
primals_8 = self.p_fc1.weight
primals_9 = self.p_fc1.bias
primals_10 = self.p_fc2.weight
primals_11 = self.p_fc2.bias
primals_12 = self.v_fc1.weight
primals_13 = self.v_fc1.bias
primals_14 = self.v_fc2.weight
primals_15 = self.v_fc2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15])
return output[0], output[1]
| LouisCaixuran/gomoku | Net | false | 5,585 | [
"Apache-2.0"
] | 1 | c1b6d508522d9e8c78be827f326bbee54c4dfd8b | https://github.com/LouisCaixuran/gomoku/tree/c1b6d508522d9e8c78be827f326bbee54c4dfd8b | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self):
super().__init__()
self.conv1 = nn.Conv2d(4, 32, 3, padding=1)
self.conv2 = nn.Conv2d(32, 64, 3, padding=1)
self.conv3 = nn.Conv2d(64, 128, 3, padding=1)
self.p_fc1 = nn.Conv2d(128, 4, 1)
self.p_fc2 = nn.Linear(4 * 8 * 8, 64)
self.v_fc1 = nn.Conv2d(128, 2, 1)
self.v_fc2 = nn.Linear(2 * 8 * 8, 1)
def forward(self, x):
a1 = F.relu(self.conv1(x))
a2 = F.relu(self.conv2(a1))
a3 = F.relu(self.conv3(a2))
p1 = F.relu(self.p_fc1(a3))
p_act = p1.view(-1, 4 * 8 * 8)
p_out = F.softmax(self.p_fc2(p_act), dim=0)
v1 = F.relu(self.v_fc1(a3))
v_act = v1.view(-1, 2 * 8 * 8)
v_out = torch.tanh(self.v_fc2(v_act))
return p_out, v_out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Binarizer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/eu/ceu5kmb7clcwhnin7efniq3cgipbjbiqe7glsxnqzjv4kv5aelk4.py
# Topologically Sorted Source Nodes: [x_1, autograd_function_apply], Original ATen: [aten.tanh, aten.sign]
# Source node to ATen node mapping:
# autograd_function_apply => sign
# x_1 => tanh
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%convolution,), kwargs = {})
# %sign : [num_users=1] = call_function[target=torch.ops.aten.sign.default](args = (%tanh,), kwargs = {})
triton_poi_fused_sign_tanh_0 = async_compile.triton('triton_poi_fused_sign_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_sign_tanh_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_sign_tanh_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = libdevice.tanh(tmp0)
tmp2 = tl.full([1], 0, tl.int32)
tmp3 = tmp2 < tmp1
tmp4 = tmp3.to(tl.int8)
tmp5 = tmp1 < tmp2
tmp6 = tmp5.to(tl.int8)
tmp7 = tmp4 - tmp6
tmp8 = tmp7.to(tmp1.dtype)
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 4, 4), (64, 16, 4, 1))
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1, autograd_function_apply], Original ATen: [aten.tanh, aten.sign]
stream0 = get_raw_stream(0)
triton_poi_fused_sign_tanh_0.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
return (buf1, primals_1, primals_2, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from torch.autograd import Function
import torch
import torch.nn as nn
import torch.nn.functional as F
class SignFunction(Function):
def __init__(self):
super(SignFunction, self).__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class Sign(nn.Module):
def __init__(self):
super(Sign, self).__init__()
def forward(self, x):
return SignFunction.apply(x, self.training)
class Binarizer(nn.Module):
def __init__(self, in_channels, out_channels):
super(Binarizer, self).__init__()
self.sign = Sign()
self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=1,
bias=False)
def forward(self, x):
x = self.conv1(x)
x = F.tanh(x)
return self.sign(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
from torch.autograd import Function
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_sign_tanh_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = libdevice.tanh(tmp0)
tmp2 = tl.full([1], 0, tl.int32)
tmp3 = tmp2 < tmp1
tmp4 = tmp3.to(tl.int8)
tmp5 = tmp1 < tmp2
tmp6 = tmp5.to(tl.int8)
tmp7 = tmp4 - tmp6
tmp8 = tmp7.to(tmp1.dtype)
tl.store(out_ptr0 + x0, tmp8, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 4, 4), (64, 16, 4, 1))
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_sign_tanh_0[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
return buf1, primals_1, primals_2, buf0
class SignFunction(Function):
def __init__(self):
super(SignFunction, self).__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class Sign(nn.Module):
def __init__(self):
super(Sign, self).__init__()
def forward(self, x):
return SignFunction.apply(x, self.training)
class BinarizerNew(nn.Module):
def __init__(self, in_channels, out_channels):
super(BinarizerNew, self).__init__()
self.sign = Sign()
self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=1,
bias=False)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
| MeMihir/SuperResCompression | Binarizer | false | 5,586 | [
"MIT"
] | 1 | c76bcf6b12d56ce3ad81ebb1b204fc0425f0e633 | https://github.com/MeMihir/SuperResCompression/tree/c76bcf6b12d56ce3ad81ebb1b204fc0425f0e633 | from torch.autograd import Function
import torch
import torch.nn as nn
import torch.nn.functional as F
class SignFunction(Function):
def __init__(self):
super().__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class Sign(nn.Module):
def __init__(self):
super().__init__()
def forward(self, x):
return SignFunction.apply(x, self.training)
class Model(nn.Module):
def __init__(self, in_channels, out_channels):
super().__init__()
self.sign = Sign()
self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=1,
bias=False)
def forward(self, x):
x = self.conv1(x)
x = F.tanh(x)
return self.sign(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
Sign | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/zy/czyscnnxj3g3lyyjhg3q4airae4625yoe5ygsjf2qxbxvi3gpiqn.py
# Topologically Sorted Source Nodes: [sign], Original ATen: [aten.sign]
# Source node to ATen node mapping:
# sign => sign
# Graph fragment:
# %sign : [num_users=1] = call_function[target=torch.ops.aten.sign.default](args = (%arg0_1,), kwargs = {})
triton_poi_fused_sign_0 = async_compile.triton('triton_poi_fused_sign_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_sign_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_sign_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp1 < tmp0
tmp3 = tmp2.to(tl.int8)
tmp4 = tmp0 < tmp1
tmp5 = tmp4.to(tl.int8)
tmp6 = tmp3 - tmp5
tmp7 = tmp6.to(tmp0.dtype)
tl.store(out_ptr0 + (x0), tmp7, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sign], Original ATen: [aten.sign]
stream0 = get_raw_stream(0)
triton_poi_fused_sign_0.run(arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from torch.autograd import Function
import torch
import torch.nn as nn
class SignFunction(Function):
def __init__(self):
super(SignFunction, self).__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class Sign(nn.Module):
def __init__(self):
super(Sign, self).__init__()
def forward(self, x):
return SignFunction.apply(x, self.training)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch.autograd import Function
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_sign_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp1 < tmp0
tmp3 = tmp2.to(tl.int8)
tmp4 = tmp0 < tmp1
tmp5 = tmp4.to(tl.int8)
tmp6 = tmp3 - tmp5
tmp7 = tmp6.to(tmp0.dtype)
tl.store(out_ptr0 + x0, tmp7, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_sign_0[grid(256)](arg0_1, buf0, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class SignFunction(Function):
def __init__(self):
super(SignFunction, self).__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class SignNew(nn.Module):
def __init__(self):
super(SignNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MeMihir/SuperResCompression | Sign | false | 5,587 | [
"MIT"
] | 1 | c76bcf6b12d56ce3ad81ebb1b204fc0425f0e633 | https://github.com/MeMihir/SuperResCompression/tree/c76bcf6b12d56ce3ad81ebb1b204fc0425f0e633 | from torch.autograd import Function
import torch
import torch.nn as nn
class SignFunction(Function):
def __init__(self):
super().__init__()
@staticmethod
def forward(ctx, input, is_training=True):
if is_training:
prob = input.new(input.size()).uniform_()
x = input.clone()
x[(1 - input) / 2 <= prob] = 1
x[(1 - input) / 2 > prob] = -1
return x
else:
return input.sign()
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, x):
return SignFunction.apply(x, self.training)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GLU | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/j6/cj63ypsp5wd4xpbcgdrjj2sjbi74adsw4ajccnbd2ift6xmplwm2.py
# Topologically Sorted Source Nodes: [sigmoid, mul], Original ATen: [aten.sigmoid, aten.mul]
# Source node to ATen node mapping:
# mul => mul
# sigmoid => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_3,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, %sigmoid), kwargs = {})
triton_poi_fused_mul_sigmoid_0 = async_compile.triton('triton_poi_fused_mul_sigmoid_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_sigmoid_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_sigmoid_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + (x0), tmp3, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf1)
del primals_4
del primals_5
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sigmoid, mul], Original ATen: [aten.sigmoid, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_sigmoid_0.run(buf0, buf1, buf2, 256, grid=grid(256), stream=stream0)
return (buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0, buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class GLU(nn.Module):
def __init__(self, input_channel, output_channel):
super(GLU, self).__init__()
self.linear_left = nn.Linear(input_channel, output_channel)
self.linear_right = nn.Linear(input_channel, output_channel)
def forward(self, x):
return torch.mul(self.linear_left(x), torch.sigmoid(self.
linear_right(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_channel': 4, 'output_channel': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_sigmoid_0(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + x0, tmp3, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf1)
del primals_4
del primals_5
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_sigmoid_0[grid(256)](buf0, buf1, buf2, 256,
XBLOCK=128, num_warps=4, num_stages=1)
return buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0, buf1
class GLUNew(nn.Module):
def __init__(self, input_channel, output_channel):
super(GLUNew, self).__init__()
self.linear_left = nn.Linear(input_channel, output_channel)
self.linear_right = nn.Linear(input_channel, output_channel)
def forward(self, input_0):
primals_1 = self.linear_left.weight
primals_2 = self.linear_left.bias
primals_4 = self.linear_right.weight
primals_5 = self.linear_right.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| MichaelHopwood/GLRM | GLU | false | 5,588 | [
"MIT"
] | 1 | 80930762e6964afb8ef0db9e5ae3a10cfcc975b2 | https://github.com/MichaelHopwood/GLRM/tree/80930762e6964afb8ef0db9e5ae3a10cfcc975b2 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, input_channel, output_channel):
super().__init__()
self.linear_left = nn.Linear(input_channel, output_channel)
self.linear_right = nn.Linear(input_channel, output_channel)
def forward(self, x):
return torch.mul(self.linear_left(x), torch.sigmoid(self.
linear_right(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
AverageAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/75/c75w3rgnfmm4c7hp5div65urlkb5kzh2656pt75swmio7vzn3vp3.py
# Topologically Sorted Source Nodes: [ones, triangle, mask], Original ATen: [aten.ones, aten.tril, aten.mul]
# Source node to ATen node mapping:
# mask => mul_1
# ones => full_default
# triangle => full_default_1, le, sub, where
# Graph fragment:
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%unsqueeze, %unsqueeze_1), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%sub, 0), kwargs = {})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%le, %full_default, %full_default_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%where, %permute), kwargs = {})
triton_poi_fused_mul_ones_tril_0 = async_compile.triton('triton_poi_fused_mul_ones_tril_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_ones_tril_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_mul_ones_tril_0(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4)
x2 = xindex
tmp0 = x0 + ((-1)*x1)
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 <= tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = 1 + x1
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp3 / tmp7
tmp9 = tmp5 * tmp8
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ha/chavpwdtejkyqus2olvrr56v6fhdolpm5dx6l26ahmwfvz664fnv.py
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %bmm], -1), kwargs = {})
triton_poi_fused_cat_1 = async_compile.triton('triton_poi_fused_cat_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = (xindex // 8)
x2 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((4*x1) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((4*x1) + ((-4) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x2), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/bj/cbjkk5x2yiy67l3q4l7ooe5u7plvwkualpweocfe25rsydr62zek.py
# Topologically Sorted Source Nodes: [sigmoid, mul_1, sigmoid_1, mul_2, gating_outputs_1], Original ATen: [aten.sigmoid, aten.mul, aten.add, aten.sigmoid_backward]
# Source node to ATen node mapping:
# gating_outputs_1 => add_1
# mul_1 => mul_2
# mul_2 => mul_3
# sigmoid => sigmoid
# sigmoid_1 => sigmoid_1
# Graph fragment:
# %sigmoid : [num_users=3] = call_function[target=torch.ops.aten.sigmoid.default](args = (%getitem,), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid, %primals_1), kwargs = {})
# %sigmoid_1 : [num_users=3] = call_function[target=torch.ops.aten.sigmoid.default](args = (%getitem_1,), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid_1, %bmm), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %mul_3), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sigmoid_1), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid_1, %sub_1), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sigmoid), kwargs = {})
# %mul_8 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid, %sub_2), kwargs = {})
triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2 = async_compile.triton('triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 6, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, out_ptr1, out_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (8*x1)), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr2 + (x2), xmask)
tmp6 = tl.load(in_ptr0 + (4 + x0 + (8*x1)), xmask)
tmp7 = tl.load(in_ptr1 + (4 + x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + (x2), xmask)
tmp2 = tmp0 + tmp1
tmp3 = tl.sigmoid(tmp2)
tmp5 = tmp3 * tmp4
tmp8 = tmp6 + tmp7
tmp9 = tl.sigmoid(tmp8)
tmp11 = tmp9 * tmp10
tmp12 = tmp5 + tmp11
tmp13 = 1.0
tmp14 = tmp13 - tmp9
tmp15 = tmp9 * tmp14
tmp16 = tmp13 - tmp3
tmp17 = tmp3 * tmp16
tl.store(out_ptr0 + (x2), tmp12, xmask)
tl.store(out_ptr1 + (x2), tmp15, xmask)
tl.store(out_ptr2 + (x2), tmp17, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (8, 8), (8, 1))
assert_size_stride(primals_3, (8, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [ones, triangle, mask], Original ATen: [aten.ones, aten.tril, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_ones_tril_0.run(buf0, 16, grid=grid(16), stream=stream0)
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [average_outputs], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf0, (4, 4, 4), (0, 4, 1), 0), primals_1, out=buf1)
del buf0
buf2 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
triton_poi_fused_cat_1.run(primals_1, buf1, buf2, 128, grid=grid(128), stream=stream0)
buf3 = empty_strided_cuda((16, 8), (8, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (16, 8), (8, 1), 0), reinterpret_tensor(primals_2, (8, 8), (1, 8), 0), out=buf3)
del primals_2
buf4 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sigmoid, mul_1, sigmoid_1, mul_2, gating_outputs_1], Original ATen: [aten.sigmoid, aten.mul, aten.add, aten.sigmoid_backward]
triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2.run(buf3, primals_3, primals_1, buf1, buf4, buf5, buf6, 64, grid=grid(64), stream=stream0)
del buf3
del primals_3
return (buf4, buf1, primals_1, buf1, reinterpret_tensor(buf2, (16, 8), (8, 1), 0), buf5, buf6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((8, 8), (8, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((8, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.cuda
import torch.distributed
class ActivationFunction(object):
relu = 'relu'
gelu = 'gelu'
class PositionwiseFeedForward(nn.Module):
""" A two-layer Feed-Forward-Network with residual layer norm.
Args:
d_model (int): the size of input for the first-layer of the FFN.
d_ff (int): the hidden layer size of the second-layer
of the FNN.
dropout (float): dropout probability in :math:`[0, 1)`.
activation_fn (ActivationFunction): activation function used.
"""
def __init__(self, d_model, d_ff, dropout=0.1, activation_fn=
ActivationFunction.relu):
super(PositionwiseFeedForward, self).__init__()
self.w_1 = nn.Linear(d_model, d_ff)
self.w_2 = nn.Linear(d_ff, d_model)
self.layer_norm = nn.LayerNorm(d_model, eps=1e-06)
self.dropout_1 = nn.Dropout(dropout)
self.activation = ACTIVATION_FUNCTIONS[activation_fn]
self.dropout_2 = nn.Dropout(dropout)
def forward(self, x):
"""Layer definition.
Args:
x: ``(batch_size, input_len, model_dim)``
Returns:
(FloatTensor): Output ``(batch_size, input_len, model_dim)``.
"""
inter = self.dropout_1(self.activation(self.w_1(self.layer_norm(x))))
output = self.dropout_2(self.w_2(inter))
return output + x
def update_dropout(self, dropout):
self.dropout_1.p = dropout
self.dropout_2.p = dropout
class AverageAttention(nn.Module):
"""
Average Attention module from
"Accelerating Neural Transformer via an Average Attention Network"
:cite:`DBLP:journals/corr/abs-1805-00631`.
Args:
model_dim (int): the dimension of keys/values/queries,
must be divisible by head_count
dropout (float): dropout parameter
pos_ffn_activation_fn (ActivationFunction):
activation function choice for PositionwiseFeedForward layer
"""
def __init__(self, model_dim, dropout=0.1, aan_useffn=False,
pos_ffn_activation_fn=ActivationFunction.relu):
self.model_dim = model_dim
self.aan_useffn = aan_useffn
super(AverageAttention, self).__init__()
if aan_useffn:
self.average_layer = PositionwiseFeedForward(model_dim,
model_dim, dropout, pos_ffn_activation_fn)
self.gating_layer = nn.Linear(model_dim * 2, model_dim * 2)
def cumulative_average_mask(self, batch_size, inputs_len, device):
"""
Builds the mask to compute the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Figure 3
Args:
batch_size (int): batch size
inputs_len (int): length of the inputs
Returns:
(FloatTensor):
* A Tensor of shape ``(batch_size, input_len, input_len)``
"""
triangle = torch.tril(torch.ones(inputs_len, inputs_len, dtype=
torch.float, device=device))
weights = torch.ones(1, inputs_len, dtype=torch.float, device=device
) / torch.arange(1, inputs_len + 1, dtype=torch.float, device=
device)
mask = triangle * weights.transpose(0, 1)
return mask.unsqueeze(0).expand(batch_size, inputs_len, inputs_len)
def cumulative_average(self, inputs, mask_or_step, layer_cache=None,
step=None):
"""
Computes the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Equations (1) (5) (6)
Args:
inputs (FloatTensor): sequence to average
``(batch_size, input_len, dimension)``
mask_or_step: if cache is set, this is assumed
to be the current step of the
dynamic decoding. Otherwise, it is the mask matrix
used to compute the cumulative average.
layer_cache: a dictionary containing the cumulative average
of the previous step.
Returns:
a tensor of the same shape and type as ``inputs``.
"""
if layer_cache is not None:
step = mask_or_step
average_attention = (inputs + step * layer_cache['prev_g']) / (step
+ 1)
layer_cache['prev_g'] = average_attention
return average_attention
else:
mask = mask_or_step
return torch.matmul(mask, inputs)
def forward(self, inputs, mask=None, layer_cache=None, step=None):
"""
Args:
inputs (FloatTensor): ``(batch_size, input_len, model_dim)``
Returns:
(FloatTensor, FloatTensor):
* gating_outputs ``(batch_size, input_len, model_dim)``
* average_outputs average attention
``(batch_size, input_len, model_dim)``
"""
batch_size = inputs.size(0)
inputs_len = inputs.size(1)
average_outputs = self.cumulative_average(inputs, self.
cumulative_average_mask(batch_size, inputs_len, inputs.device) if
layer_cache is None else step, layer_cache=layer_cache)
if self.aan_useffn:
average_outputs = self.average_layer(average_outputs)
gating_outputs = self.gating_layer(torch.cat((inputs,
average_outputs), -1))
input_gate, forget_gate = torch.chunk(gating_outputs, 2, dim=2)
gating_outputs = torch.sigmoid(input_gate) * inputs + torch.sigmoid(
forget_gate) * average_outputs
return gating_outputs, average_outputs
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'model_dim': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.cuda
import torch.distributed
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_ones_tril_0(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = x0 + -1 * x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 <= tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = 1 + x1
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp3 / tmp7
tmp9 = tmp5 * tmp8
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused_cat_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2(in_ptr0, in_ptr1,
in_ptr2, in_ptr3, out_ptr0, out_ptr1, out_ptr2, xnumel, XBLOCK: tl.
constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 8 * x1), xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr2 + x2, xmask)
tmp6 = tl.load(in_ptr0 + (4 + x0 + 8 * x1), xmask)
tmp7 = tl.load(in_ptr1 + (4 + x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x2, xmask)
tmp2 = tmp0 + tmp1
tmp3 = tl.sigmoid(tmp2)
tmp5 = tmp3 * tmp4
tmp8 = tmp6 + tmp7
tmp9 = tl.sigmoid(tmp8)
tmp11 = tmp9 * tmp10
tmp12 = tmp5 + tmp11
tmp13 = 1.0
tmp14 = tmp13 - tmp9
tmp15 = tmp9 * tmp14
tmp16 = tmp13 - tmp3
tmp17 = tmp3 * tmp16
tl.store(out_ptr0 + x2, tmp12, xmask)
tl.store(out_ptr1 + x2, tmp15, xmask)
tl.store(out_ptr2 + x2, tmp17, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (8, 8), (8, 1))
assert_size_stride(primals_3, (8,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_ones_tril_0[grid(16)](buf0, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf0, (4, 4, 4), (0, 4, 1), 0
), primals_1, out=buf1)
del buf0
buf2 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
triton_poi_fused_cat_1[grid(128)](primals_1, buf1, buf2, 128,
XBLOCK=128, num_warps=4, num_stages=1)
buf3 = empty_strided_cuda((16, 8), (8, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (16, 8), (8, 1), 0),
reinterpret_tensor(primals_2, (8, 8), (1, 8), 0), out=buf3)
del primals_2
buf4 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_mul_sigmoid_sigmoid_backward_2[grid(64)](buf3,
primals_3, primals_1, buf1, buf4, buf5, buf6, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf3
del primals_3
return buf4, buf1, primals_1, buf1, reinterpret_tensor(buf2, (16, 8), (
8, 1), 0), buf5, buf6
class ActivationFunction(object):
relu = 'relu'
gelu = 'gelu'
class PositionwiseFeedForward(nn.Module):
""" A two-layer Feed-Forward-Network with residual layer norm.
Args:
d_model (int): the size of input for the first-layer of the FFN.
d_ff (int): the hidden layer size of the second-layer
of the FNN.
dropout (float): dropout probability in :math:`[0, 1)`.
activation_fn (ActivationFunction): activation function used.
"""
def __init__(self, d_model, d_ff, dropout=0.1, activation_fn=
ActivationFunction.relu):
super(PositionwiseFeedForward, self).__init__()
self.w_1 = nn.Linear(d_model, d_ff)
self.w_2 = nn.Linear(d_ff, d_model)
self.layer_norm = nn.LayerNorm(d_model, eps=1e-06)
self.dropout_1 = nn.Dropout(dropout)
self.activation = ACTIVATION_FUNCTIONS[activation_fn]
self.dropout_2 = nn.Dropout(dropout)
def forward(self, x):
"""Layer definition.
Args:
x: ``(batch_size, input_len, model_dim)``
Returns:
(FloatTensor): Output ``(batch_size, input_len, model_dim)``.
"""
inter = self.dropout_1(self.activation(self.w_1(self.layer_norm(x))))
output = self.dropout_2(self.w_2(inter))
return output + x
def update_dropout(self, dropout):
self.dropout_1.p = dropout
self.dropout_2.p = dropout
class AverageAttentionNew(nn.Module):
"""
Average Attention module from
"Accelerating Neural Transformer via an Average Attention Network"
:cite:`DBLP:journals/corr/abs-1805-00631`.
Args:
model_dim (int): the dimension of keys/values/queries,
must be divisible by head_count
dropout (float): dropout parameter
pos_ffn_activation_fn (ActivationFunction):
activation function choice for PositionwiseFeedForward layer
"""
def __init__(self, model_dim, dropout=0.1, aan_useffn=False,
pos_ffn_activation_fn=ActivationFunction.relu):
self.model_dim = model_dim
self.aan_useffn = aan_useffn
super(AverageAttentionNew, self).__init__()
if aan_useffn:
self.average_layer = PositionwiseFeedForward(model_dim,
model_dim, dropout, pos_ffn_activation_fn)
self.gating_layer = nn.Linear(model_dim * 2, model_dim * 2)
def cumulative_average_mask(self, batch_size, inputs_len, device):
"""
Builds the mask to compute the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Figure 3
Args:
batch_size (int): batch size
inputs_len (int): length of the inputs
Returns:
(FloatTensor):
* A Tensor of shape ``(batch_size, input_len, input_len)``
"""
triangle = torch.tril(torch.ones(inputs_len, inputs_len, dtype=
torch.float, device=device))
weights = torch.ones(1, inputs_len, dtype=torch.float, device=device
) / torch.arange(1, inputs_len + 1, dtype=torch.float, device=
device)
mask = triangle * weights.transpose(0, 1)
return mask.unsqueeze(0).expand(batch_size, inputs_len, inputs_len)
def cumulative_average(self, inputs, mask_or_step, layer_cache=None,
step=None):
"""
Computes the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Equations (1) (5) (6)
Args:
inputs (FloatTensor): sequence to average
``(batch_size, input_len, dimension)``
mask_or_step: if cache is set, this is assumed
to be the current step of the
dynamic decoding. Otherwise, it is the mask matrix
used to compute the cumulative average.
layer_cache: a dictionary containing the cumulative average
of the previous step.
Returns:
a tensor of the same shape and type as ``inputs``.
"""
if layer_cache is not None:
step = mask_or_step
average_attention = (inputs + step * layer_cache['prev_g']) / (step
+ 1)
layer_cache['prev_g'] = average_attention
return average_attention
else:
mask = mask_or_step
return torch.matmul(mask, inputs)
def forward(self, input_0):
primals_2 = self.gating_layer.weight
primals_3 = self.gating_layer.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0], output[1]
| MaxatTezekbayev/OpenNMT-py-lexical | AverageAttention | false | 5,589 | [
"MIT"
] | 1 | 44182999b863fc4074d67e0281c5bdab19abddfe | https://github.com/MaxatTezekbayev/OpenNMT-py-lexical/tree/44182999b863fc4074d67e0281c5bdab19abddfe | import torch
import torch.nn as nn
import torch.cuda
import torch.distributed
class ActivationFunction(object):
relu = 'relu'
gelu = 'gelu'
class PositionwiseFeedForward(nn.Module):
""" A two-layer Feed-Forward-Network with residual layer norm.
Args:
d_model (int): the size of input for the first-layer of the FFN.
d_ff (int): the hidden layer size of the second-layer
of the FNN.
dropout (float): dropout probability in :math:`[0, 1)`.
activation_fn (ActivationFunction): activation function used.
"""
def __init__(self, d_model, d_ff, dropout=0.1, activation_fn=
ActivationFunction.relu):
super().__init__()
self.w_1 = nn.Linear(d_model, d_ff)
self.w_2 = nn.Linear(d_ff, d_model)
self.layer_norm = nn.LayerNorm(d_model, eps=1e-06)
self.dropout_1 = nn.Dropout(dropout)
self.activation = ACTIVATION_FUNCTIONS[activation_fn]
self.dropout_2 = nn.Dropout(dropout)
def forward(self, x):
"""Layer definition.
Args:
x: ``(batch_size, input_len, model_dim)``
Returns:
(FloatTensor): Output ``(batch_size, input_len, model_dim)``.
"""
inter = self.dropout_1(self.activation(self.w_1(self.layer_norm(x))))
output = self.dropout_2(self.w_2(inter))
return output + x
def update_dropout(self, dropout):
self.dropout_1.p = dropout
self.dropout_2.p = dropout
class Model(nn.Module):
"""
Average Attention module from
"Accelerating Neural Transformer via an Average Attention Network"
:cite:`DBLP:journals/corr/abs-1805-00631`.
Args:
model_dim (int): the dimension of keys/values/queries,
must be divisible by head_count
dropout (float): dropout parameter
pos_ffn_activation_fn (ActivationFunction):
activation function choice for PositionwiseFeedForward layer
"""
def __init__(self, model_dim, dropout=0.1, aan_useffn=False,
pos_ffn_activation_fn=ActivationFunction.relu):
self.model_dim = model_dim
self.aan_useffn = aan_useffn
super().__init__()
if aan_useffn:
self.average_layer = PositionwiseFeedForward(model_dim,
model_dim, dropout, pos_ffn_activation_fn)
self.gating_layer = nn.Linear(model_dim * 2, model_dim * 2)
def cumulative_average_mask(self, batch_size, inputs_len, device):
"""
Builds the mask to compute the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Figure 3
Args:
batch_size (int): batch size
inputs_len (int): length of the inputs
Returns:
(FloatTensor):
* A Tensor of shape ``(batch_size, input_len, input_len)``
"""
triangle = torch.tril(torch.ones(inputs_len, inputs_len, dtype=
torch.float, device=device))
weights = torch.ones(1, inputs_len, dtype=torch.float, device=device
) / torch.arange(1, inputs_len + 1, dtype=torch.float, device=
device)
mask = triangle * weights.transpose(0, 1)
return mask.unsqueeze(0).expand(batch_size, inputs_len, inputs_len)
def cumulative_average(self, inputs, mask_or_step, layer_cache=None,
step=None):
"""
Computes the cumulative average as described in
:cite:`DBLP:journals/corr/abs-1805-00631` -- Equations (1) (5) (6)
Args:
inputs (FloatTensor): sequence to average
``(batch_size, input_len, dimension)``
mask_or_step: if cache is set, this is assumed
to be the current step of the
dynamic decoding. Otherwise, it is the mask matrix
used to compute the cumulative average.
layer_cache: a dictionary containing the cumulative average
of the previous step.
Returns:
# ... truncated (>4000 chars) for memory efficiency |
SRNet | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ej/cejfrwnzxinkchwn6symdb72fdtj7gix5hy2vuswodhbeh45mrae.py
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# x => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_convolution_relu_0 = async_compile.triton('triton_poi_fused_convolution_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1048576],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1048576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 4096) % 64
tmp0 = tl.load(in_out_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/of/coferfgb6zeb7ewd25keilubhznpkefhirehs3f5p2bmxsjdmmrt.py
# Topologically Sorted Source Nodes: [x_7, SR_img], Original ATen: [aten.convolution, aten.add]
# Source node to ATen node mapping:
# SR_img => add
# x_7 => convolution_7
# Graph fragment:
# %convolution_7 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_6, %primals_16, %primals_17, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %convolution_7), kwargs = {})
triton_poi_fused_add_convolution_1 = async_compile.triton('triton_poi_fused_add_convolution_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_convolution_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_convolution_1(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 49152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 4096) % 3
tmp0 = tl.load(in_ptr0 + (x3), None)
tmp1 = tl.load(in_out_ptr0 + (x3), None)
tmp2 = tl.load(in_ptr1 + (x1), None, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17 = args
args.clear()
assert_size_stride(primals_1, (64, 3, 3, 3), (27, 9, 3, 1))
assert_size_stride(primals_2, (64, ), (1, ))
assert_size_stride(primals_3, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64, ), (1, ))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64, ), (1, ))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (64, ), (1, ))
assert_size_stride(primals_10, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_11, (64, ), (1, ))
assert_size_stride(primals_12, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_13, (64, ), (1, ))
assert_size_stride(primals_14, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_15, (64, ), (1, ))
assert_size_stride(primals_16, (3, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_17, (3, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 1048576, grid=grid(1048576), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, x_1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf3, primals_5, 1048576, grid=grid(1048576), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [conv2d_2, x_2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf5, primals_7, 1048576, grid=grid(1048576), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [conv2d_3, x_3], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf7, primals_9, 1048576, grid=grid(1048576), stream=stream0)
del primals_9
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_10, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [conv2d_4, x_4], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf9, primals_11, 1048576, grid=grid(1048576), stream=stream0)
del primals_11
# Topologically Sorted Source Nodes: [conv2d_5], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_12, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [conv2d_5, x_5], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf11, primals_13, 1048576, grid=grid(1048576), stream=stream0)
del primals_13
# Topologically Sorted Source Nodes: [conv2d_6], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_14, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [conv2d_6, x_6], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf13, primals_15, 1048576, grid=grid(1048576), stream=stream0)
del primals_15
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_16, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf15 = buf14; del buf14 # reuse
# Topologically Sorted Source Nodes: [x_7, SR_img], Original ATen: [aten.convolution, aten.add]
triton_poi_fused_add_convolution_1.run(buf15, primals_3, primals_17, 49152, grid=grid(49152), stream=stream0)
del primals_17
return (buf15, primals_1, primals_3, primals_4, primals_6, primals_8, primals_10, primals_12, primals_14, primals_16, buf1, buf3, buf5, buf7, buf9, buf11, buf13, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((64, 3, 3, 3), (27, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 3, 64, 64), (12288, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((3, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((3, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.optim
class SRNet(nn.Module):
def __init__(self):
super(SRNet, self).__init__()
self.relu = nn.ReLU(inplace=True)
self.Conv1 = nn.Conv2d(3, 64, 3, 1, 1, bias=True)
self.Conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv8 = nn.Conv2d(64, 3, 3, 1, 1, bias=True)
def forward(self, LR_img):
x = self.relu(self.Conv1(LR_img))
x = self.relu(self.Conv2(x))
x = self.relu(self.Conv3(x))
x = self.relu(self.Conv4(x))
x = self.relu(self.Conv5(x))
x = self.relu(self.Conv6(x))
x = self.relu(self.Conv7(x))
x = self.Conv8(x)
SR_img = LR_img + x
return SR_img
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
import torch.optim
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 64
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_add_convolution_1(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 3
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_out_ptr0 + x3, None)
tmp2 = tl.load(in_ptr1 + x1, None, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + x3, tmp4, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17) = args
args.clear()
assert_size_stride(primals_1, (64, 3, 3, 3), (27, 9, 3, 1))
assert_size_stride(primals_2, (64,), (1,))
assert_size_stride(primals_3, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64,), (1,))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (64,), (1,))
assert_size_stride(primals_10, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_11, (64,), (1,))
assert_size_stride(primals_12, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_13, (64,), (1,))
assert_size_stride(primals_14, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_15, (64,), (1,))
assert_size_stride(primals_16, (3, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_17, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(1048576)](buf1, primals_2,
1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_0[grid(1048576)](buf3, primals_5,
1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_0[grid(1048576)](buf5, primals_7,
1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_7
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_0[grid(1048576)](buf7, primals_9,
1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_9
buf8 = extern_kernels.convolution(buf7, primals_10, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_convolution_relu_0[grid(1048576)](buf9, primals_11,
1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_11
buf10 = extern_kernels.convolution(buf9, primals_12, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_convolution_relu_0[grid(1048576)](buf11,
primals_13, 1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_13
buf12 = extern_kernels.convolution(buf11, primals_14, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_convolution_relu_0[grid(1048576)](buf13,
primals_15, 1048576, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_15
buf14 = extern_kernels.convolution(buf13, primals_16, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf15 = buf14
del buf14
triton_poi_fused_add_convolution_1[grid(49152)](buf15, primals_3,
primals_17, 49152, XBLOCK=256, num_warps=4, num_stages=1)
del primals_17
return (buf15, primals_1, primals_3, primals_4, primals_6, primals_8,
primals_10, primals_12, primals_14, primals_16, buf1, buf3, buf5,
buf7, buf9, buf11, buf13)
class SRNetNew(nn.Module):
def __init__(self):
super(SRNetNew, self).__init__()
self.relu = nn.ReLU(inplace=True)
self.Conv1 = nn.Conv2d(3, 64, 3, 1, 1, bias=True)
self.Conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv8 = nn.Conv2d(64, 3, 3, 1, 1, bias=True)
def forward(self, input_0):
primals_1 = self.Conv1.weight
primals_2 = self.Conv1.bias
primals_4 = self.Conv2.weight
primals_5 = self.Conv2.bias
primals_6 = self.Conv3.weight
primals_7 = self.Conv3.bias
primals_8 = self.Conv4.weight
primals_9 = self.Conv4.bias
primals_10 = self.Conv5.weight
primals_11 = self.Conv5.bias
primals_12 = self.Conv6.weight
primals_13 = self.Conv6.bias
primals_14 = self.Conv7.weight
primals_15 = self.Conv7.bias
primals_16 = self.Conv8.weight
primals_17 = self.Conv8.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17])
return output[0]
| MayankSingal/PyTorch-Zero-Shot-Super-Resolution | SRNet | false | 5,590 | [
"MIT"
] | 1 | 3521b02fd338fc90eef88c551a8bed4afc54c8c6 | https://github.com/MayankSingal/PyTorch-Zero-Shot-Super-Resolution/tree/3521b02fd338fc90eef88c551a8bed4afc54c8c6 | import torch
import torch.nn as nn
import torch.optim
class Model(nn.Module):
def __init__(self):
super().__init__()
self.relu = nn.ReLU(inplace=True)
self.Conv1 = nn.Conv2d(3, 64, 3, 1, 1, bias=True)
self.Conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=True)
self.Conv8 = nn.Conv2d(64, 3, 3, 1, 1, bias=True)
def forward(self, LR_img):
x = self.relu(self.Conv1(LR_img))
x = self.relu(self.Conv2(x))
x = self.relu(self.Conv3(x))
x = self.relu(self.Conv4(x))
x = self.relu(self.Conv5(x))
x = self.relu(self.Conv6(x))
x = self.relu(self.Conv7(x))
x = self.Conv8(x)
SR_img = LR_img + x
return SR_img
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return []
|
Sparsemax | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/je/cjetqbcnd7bc32n35tugus5ijwdxgd2rgnx55splotkzsvjkv2mu.py
# Topologically Sorted Source Nodes: [input_2, sort, cumulative_sum_zs], Original ATen: [aten.sub, aten.sort, aten.cumsum]
# Source node to ATen node mapping:
# cumulative_sum_zs => cumsum
# input_2 => sub
# sort => sort
# Graph fragment:
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view, %expand), kwargs = {})
# %sort : [num_users=1] = call_function[target=torch.ops.aten.sort.default](args = (%sub, 1, True), kwargs = {})
# %cumsum : [num_users=1] = call_function[target=torch.ops.aten.cumsum.default](args = (%getitem_2, 1), kwargs = {})
triton_per_fused_cumsum_sort_sub_0 = async_compile.triton('triton_per_fused_cumsum_sort_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton.jit
def _triton_helper_fn_add0(arg0_0, arg1_0):
tmp0 = arg0_0 + arg1_0
return tmp0
@triton_heuristics.persistent_reduction(
size_hints=[64, 4],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_cumsum_sort_sub_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_cumsum_sort_sub_0(in_ptr0, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 64
rnumel = 4
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (4*x0)), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = r1
tmp10 = tmp9.to(tl.int16)
tmp11 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp12 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13, tmp14, = triton_helpers.sort_with_index(tmp11, tmp12, None, 1, stable=False, descending=True)
tmp15 = tmp13.to(tl.float32)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp17, = tl.associative_scan((tmp16,), 1, _triton_helper_fn_add0)
tl.store(out_ptr0 + (r1 + (4*x0)), tmp8, xmask)
tl.store(out_ptr1 + (r1 + (4*x0)), tmp13, xmask)
tl.store(out_ptr2 + (r1 + (4*x0)), tmp17, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rm/crmbpyujmwscp5mkcahmeli47zzzh5qct73vmu2ldmazrzxmqpsx.py
# Topologically Sorted Source Nodes: [mul, bound, gt, is_gt, mul_1, max_2, zs_sparse, sum_1], Original ATen: [aten.mul, aten.add, aten.gt, aten._to_copy, aten.max, aten.sum]
# Source node to ATen node mapping:
# bound => add
# gt => gt
# is_gt => convert_element_type
# max_2 => max_2
# mul => mul
# mul_1 => mul_1
# sum_1 => sum_1
# zs_sparse => mul_2
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%expand_1, %getitem_2), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, 1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Tensor](args = (%add, %cumsum), kwargs = {})
# %convert_element_type : [num_users=2] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%gt, torch.float32), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convert_element_type, %expand_1), kwargs = {})
# %max_2 : [num_users=1] = call_function[target=torch.ops.aten.max.dim](args = (%mul_1, 1, True), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convert_element_type, %getitem_2), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_2, [1], True), kwargs = {})
triton_poi_fused__to_copy_add_gt_max_mul_sum_1 = async_compile.triton('triton_poi_fused__to_copy_add_gt_max_mul_sum_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_add_gt_max_mul_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 12, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__to_copy_add_gt_max_mul_sum_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (0))
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr2 + (4*x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (1))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK])
tmp12 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr0 + (2))
tmp21 = tl.broadcast_to(tmp20, [XBLOCK])
tmp22 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr2 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr0 + (3))
tmp31 = tl.broadcast_to(tmp30, [XBLOCK])
tmp32 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr2 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 * tmp2
tmp4 = 1.0
tmp5 = tmp3 + tmp4
tmp7 = tmp5 > tmp6
tmp8 = tmp7.to(tl.float32)
tmp9 = tmp8 * tmp1
tmp13 = tmp11 * tmp12
tmp14 = tmp13 + tmp4
tmp16 = tmp14 > tmp15
tmp17 = tmp16.to(tl.float32)
tmp18 = tmp17 * tmp11
tmp19 = triton_helpers.maximum(tmp9, tmp18)
tmp23 = tmp21 * tmp22
tmp24 = tmp23 + tmp4
tmp26 = tmp24 > tmp25
tmp27 = tmp26.to(tl.float32)
tmp28 = tmp27 * tmp21
tmp29 = triton_helpers.maximum(tmp19, tmp28)
tmp33 = tmp31 * tmp32
tmp34 = tmp33 + tmp4
tmp36 = tmp34 > tmp35
tmp37 = tmp36.to(tl.float32)
tmp38 = tmp37 * tmp31
tmp39 = triton_helpers.maximum(tmp29, tmp38)
tmp40 = tmp8 * tmp2
tmp41 = tmp17 * tmp12
tmp42 = tmp40 + tmp41
tmp43 = tmp27 * tmp22
tmp44 = tmp42 + tmp43
tmp45 = tmp37 * tmp32
tmp46 = tmp44 + tmp45
tl.store(out_ptr0 + (x0), tmp39, xmask)
tl.store(out_ptr1 + (x0), tmp46, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/xz/cxze32rtorhwy2jkafxfvxxeyi2jg24j22ihfkfwtfd22dqgy2go.py
# Topologically Sorted Source Nodes: [zeros_like, sub_2, max_3], Original ATen: [aten.zeros_like, aten.sub, aten.maximum]
# Source node to ATen node mapping:
# max_3 => maximum
# sub_2 => sub_2
# zeros_like => full_default
# Graph fragment:
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([64, 4], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%sub, %expand_2), kwargs = {})
# %maximum : [num_users=2] = call_function[target=torch.ops.aten.maximum.default](args = (%full_default, %sub_2), kwargs = {})
triton_poi_fused_maximum_sub_zeros_like_2 = async_compile.triton('triton_poi_fused_maximum_sub_zeros_like_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_maximum_sub_zeros_like_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_maximum_sub_zeros_like_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp2 = 1.0
tmp3 = tmp1 - tmp2
tmp5 = tmp3 / tmp4
tmp6 = tmp0 - tmp5
tmp7 = 0.0
tmp8 = triton_helpers.maximum(tmp7, tmp6)
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
buf5 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [input_2, sort, cumulative_sum_zs], Original ATen: [aten.sub, aten.sort, aten.cumsum]
stream0 = get_raw_stream(0)
triton_per_fused_cumsum_sort_sub_0.run(arg0_1, buf0, buf1, buf5, 64, 4, grid=grid(64), stream=stream0)
del arg0_1
# Topologically Sorted Source Nodes: [range_1], Original ATen: [aten.range]
buf3 = torch.ops.aten.range.step(1, 4, dtype=torch.float32, layout=torch.strided, device=device(type='cuda', index=0))
buf4 = buf3
del buf3
buf6 = empty_strided_cuda((64, 1), (1, 64), torch.float32)
buf7 = empty_strided_cuda((64, 1), (1, 64), torch.float32)
# Topologically Sorted Source Nodes: [mul, bound, gt, is_gt, mul_1, max_2, zs_sparse, sum_1], Original ATen: [aten.mul, aten.add, aten.gt, aten._to_copy, aten.max, aten.sum]
triton_poi_fused__to_copy_add_gt_max_mul_sum_1.run(buf4, buf1, buf5, buf6, buf7, 64, grid=grid(64), stream=stream0)
del buf1
del buf4
buf8 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [zeros_like, sub_2, max_3], Original ATen: [aten.zeros_like, aten.sub, aten.maximum]
triton_poi_fused_maximum_sub_zeros_like_2.run(buf0, buf7, buf6, buf8, 256, grid=grid(256), stream=stream0)
del buf0
del buf6
del buf7
return (reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf8, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Sparsemax(nn.Module):
"""Sparsemax function."""
def __init__(self, dim=None):
"""Initialize sparsemax activation
Args:
dim (int, optional): The dimension over which to apply the sparsemax function.
"""
super(Sparsemax, self).__init__()
self.dim = -1 if dim is None else dim
def forward(self, input):
"""Forward function.
Args:
input (torch.Tensor): Input tensor. First dimension should be the batch size
Returns:
torch.Tensor: [batch_size x number_of_logits] Output tensor
"""
original_size = input.size()
input = input.view(-1, input.size(self.dim))
dim = 1
number_of_logits = input.size(dim)
input = input - torch.max(input, dim=dim, keepdim=True)[0].expand_as(
input)
zs = torch.sort(input=input, dim=dim, descending=True)[0]
range = torch.range(start=1, end=number_of_logits, device=input.device
).view(1, -1)
range = range.expand_as(zs)
bound = 1 + range * zs
cumulative_sum_zs = torch.cumsum(zs, dim)
is_gt = torch.gt(bound, cumulative_sum_zs).type(input.type())
k = torch.max(is_gt * range, dim, keepdim=True)[0]
zs_sparse = is_gt * zs
taus = (torch.sum(zs_sparse, dim, keepdim=True) - 1) / k
taus = taus.expand_as(input)
self.output = torch.max(torch.zeros_like(input), input - taus)
output = self.output.view(original_size)
return output
def backward(self, grad_output):
"""Backward function."""
dim = 1
nonzeros = torch.ne(self.output, 0)
sum = torch.sum(grad_output * nonzeros, dim=dim) / torch.sum(nonzeros,
dim=dim)
self.grad_input = nonzeros * (grad_output - sum.expand_as(grad_output))
return self.grad_input
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
from torch import device
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def _triton_helper_fn_add0(arg0_0, arg1_0):
tmp0 = arg0_0 + arg1_0
return tmp0
@triton.jit
def triton_per_fused_cumsum_sort_sub_0(in_ptr0, out_ptr0, out_ptr1,
out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 64
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 4 * x0), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = r1
tmp10 = tmp9.to(tl.int16)
tmp11 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp12 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13, _tmp14 = triton_helpers.sort_with_index(tmp11, tmp12, None, 1,
stable=False, descending=True)
tmp15 = tmp13.to(tl.float32)
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp17, = tl.associative_scan((tmp16,), 1, _triton_helper_fn_add0)
tl.store(out_ptr0 + (r1 + 4 * x0), tmp8, xmask)
tl.store(out_ptr1 + (r1 + 4 * x0), tmp13, xmask)
tl.store(out_ptr2 + (r1 + 4 * x0), tmp17, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_gt_max_mul_sum_1(in_ptr0, in_ptr1,
in_ptr2, out_ptr0, out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr2 + 4 * x0, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + 1)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK])
tmp12 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr2 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr0 + 2)
tmp21 = tl.broadcast_to(tmp20, [XBLOCK])
tmp22 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp25 = tl.load(in_ptr2 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp30 = tl.load(in_ptr0 + 3)
tmp31 = tl.broadcast_to(tmp30, [XBLOCK])
tmp32 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp35 = tl.load(in_ptr2 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp3 = tmp1 * tmp2
tmp4 = 1.0
tmp5 = tmp3 + tmp4
tmp7 = tmp5 > tmp6
tmp8 = tmp7.to(tl.float32)
tmp9 = tmp8 * tmp1
tmp13 = tmp11 * tmp12
tmp14 = tmp13 + tmp4
tmp16 = tmp14 > tmp15
tmp17 = tmp16.to(tl.float32)
tmp18 = tmp17 * tmp11
tmp19 = triton_helpers.maximum(tmp9, tmp18)
tmp23 = tmp21 * tmp22
tmp24 = tmp23 + tmp4
tmp26 = tmp24 > tmp25
tmp27 = tmp26.to(tl.float32)
tmp28 = tmp27 * tmp21
tmp29 = triton_helpers.maximum(tmp19, tmp28)
tmp33 = tmp31 * tmp32
tmp34 = tmp33 + tmp4
tmp36 = tmp34 > tmp35
tmp37 = tmp36.to(tl.float32)
tmp38 = tmp37 * tmp31
tmp39 = triton_helpers.maximum(tmp29, tmp38)
tmp40 = tmp8 * tmp2
tmp41 = tmp17 * tmp12
tmp42 = tmp40 + tmp41
tmp43 = tmp27 * tmp22
tmp44 = tmp42 + tmp43
tmp45 = tmp37 * tmp32
tmp46 = tmp44 + tmp45
tl.store(out_ptr0 + x0, tmp39, xmask)
tl.store(out_ptr1 + x0, tmp46, xmask)
@triton.jit
def triton_poi_fused_maximum_sub_zeros_like_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp2 = 1.0
tmp3 = tmp1 - tmp2
tmp5 = tmp3 / tmp4
tmp6 = tmp0 - tmp5
tmp7 = 0.0
tmp8 = triton_helpers.maximum(tmp7, tmp6)
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
buf5 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_cumsum_sort_sub_0[grid(64)](arg0_1, buf0, buf1,
buf5, 64, 4, XBLOCK=8, num_warps=2, num_stages=1)
del arg0_1
buf3 = torch.ops.aten.range.step(1, 4, dtype=torch.float32, layout=
torch.strided, device=device(type='cuda', index=0))
buf4 = buf3
del buf3
buf6 = empty_strided_cuda((64, 1), (1, 64), torch.float32)
buf7 = empty_strided_cuda((64, 1), (1, 64), torch.float32)
triton_poi_fused__to_copy_add_gt_max_mul_sum_1[grid(64)](buf4, buf1,
buf5, buf6, buf7, 64, XBLOCK=64, num_warps=1, num_stages=1)
del buf1
del buf4
buf8 = buf5
del buf5
triton_poi_fused_maximum_sub_zeros_like_2[grid(256)](buf0, buf7,
buf6, buf8, 256, XBLOCK=256, num_warps=4, num_stages=1)
del buf0
del buf6
del buf7
return reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf8
class SparsemaxNew(nn.Module):
"""Sparsemax function."""
def __init__(self, dim=None):
"""Initialize sparsemax activation
Args:
dim (int, optional): The dimension over which to apply the sparsemax function.
"""
super(SparsemaxNew, self).__init__()
self.dim = -1 if dim is None else dim
def backward(self, grad_output):
"""Backward function."""
dim = 1
nonzeros = torch.ne(self.output, 0)
sum = torch.sum(grad_output * nonzeros, dim=dim) / torch.sum(nonzeros,
dim=dim)
self.grad_input = nonzeros * (grad_output - sum.expand_as(grad_output))
return self.grad_input
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| Max-luo-song/fs-map-project | Sparsemax | false | 5,591 | [
"Apache-2.0"
] | 1 | 4e9d86e182d9a4b969e86b12d72f227e4fd4fd09 | https://github.com/Max-luo-song/fs-map-project/tree/4e9d86e182d9a4b969e86b12d72f227e4fd4fd09 | import torch
import torch.nn as nn
class Model(nn.Module):
"""Sparsemax function."""
def __init__(self, dim=None):
"""Initialize sparsemax activation
Args:
dim (int, optional): The dimension over which to apply the sparsemax function.
"""
super().__init__()
self.dim = -1 if dim is None else dim
def forward(self, input):
"""Forward function.
Args:
input (torch.Tensor): Input tensor. First dimension should be the batch size
Returns:
torch.Tensor: [batch_size x number_of_logits] Output tensor
"""
original_size = input.size()
input = input.view(-1, input.size(self.dim))
dim = 1
number_of_logits = input.size(dim)
input = input - torch.max(input, dim=dim, keepdim=True)[0].expand_as(
input)
zs = torch.sort(input=input, dim=dim, descending=True)[0]
range = torch.range(start=1, end=number_of_logits, device=input.device
).view(1, -1)
range = range.expand_as(zs)
bound = 1 + range * zs
cumulative_sum_zs = torch.cumsum(zs, dim)
is_gt = torch.gt(bound, cumulative_sum_zs).type(input.type())
k = torch.max(is_gt * range, dim, keepdim=True)[0]
zs_sparse = is_gt * zs
taus = (torch.sum(zs_sparse, dim, keepdim=True) - 1) / k
taus = taus.expand_as(input)
self.output = torch.max(torch.zeros_like(input), input - taus)
output = self.output.view(original_size)
return output
def backward(self, grad_output):
"""Backward function."""
dim = 1
nonzeros = torch.ne(self.output, 0)
sum = torch.sum(grad_output * nonzeros, dim=dim) / torch.sum(nonzeros,
dim=dim)
self.grad_input = nonzeros * (grad_output - sum.expand_as(grad_output))
return self.grad_input
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
NTXent | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/fh/cfhnguw4v6uy4ysjg54ojclakwi3bj2lte6oqizl4rpf4lcxpiyp.py
# Topologically Sorted Source Nodes: [emb], Original ATen: [aten.div]
# Source node to ATen node mapping:
# emb => div
# Graph fragment:
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%arg0_1, %expand), kwargs = {})
triton_poi_fused_div_0 = async_compile.triton('triton_poi_fused_div_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_div_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_div_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + (x3), tmp15, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/lv/clvvd5zl3dk2doqn4hqb7xws2n2pplfd2c6z2bjtqzpgmwmcb6bi.py
# Topologically Sorted Source Nodes: [sim_mat], Original ATen: [aten.linalg_vector_norm, aten.clamp_min, aten.div, aten.mul]
# Source node to ATen node mapping:
# sim_mat => clamp_min_1, clamp_min_2, div_1, div_2, mul, pow_3, pow_4, pow_5, pow_6, sum_2, sum_3
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%expand_2, 2), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, [2], True), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %clamp_min_1 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%pow_4, 1e-08), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%expand_2, %clamp_min_1), kwargs = {})
# %pow_5 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%expand_1, 2), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_5, [2], True), kwargs = {})
# %pow_6 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_3, 0.5), kwargs = {})
# %clamp_min_2 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%pow_6, 1e-08), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%expand_1, %clamp_min_2), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div_2, %div_1), kwargs = {})
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1 = async_compile.triton('triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 10, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = (xindex // 256)
x4 = xindex % 64
x0 = xindex % 16
x6 = xindex % 256
x2 = (xindex // 64) % 4
x7 = xindex
tmp0 = tl.load(in_ptr0 + (x4 + (64*x3)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (x0 + (64*x3)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + (64*x3)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + (64*x3)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + (64*x3)), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr0 + (x6), xmask, eviction_policy='evict_last')
tmp17 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-08
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tmp18 = tmp17 * tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = libdevice.sqrt(tmp27)
tmp29 = triton_helpers.maximum(tmp28, tmp13)
tmp30 = tmp16 / tmp29
tmp31 = tmp15 * tmp30
tl.store(out_ptr0 + (x7), tmp31, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/zb/czbffbfo4mbxwezgrsoj7ucwdqx2cq7uw73rnvx3wpkm4bu6o5k4.py
# Topologically Sorted Source Nodes: [sim_mat, truediv, sim_mat_1, mul, pos, sum_2, inner, log, loss, loss_1], Original ATen: [aten.sum, aten.div, aten.exp, aten.mul, aten.log, aten.neg, aten.mean]
# Source node to ATen node mapping:
# inner => div_4
# log => log
# loss => neg
# loss_1 => mean
# mul => mul_1
# pos => sum_5
# sim_mat => sum_4
# sim_mat_1 => exp
# sum_2 => sum_6
# truediv => div_3
# Graph fragment:
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [2]), kwargs = {})
# %div_3 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_4, 0.5), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_3,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp, %arg1_1), kwargs = {})
# %sum_5 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul_1, [1]), kwargs = {})
# %sum_6 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1]), kwargs = {})
# %div_4 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_5, %sum_6), kwargs = {})
# %log : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%div_4,), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%log,), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%neg,), kwargs = {})
triton_per_fused_div_exp_log_mean_mul_neg_sum_2 = async_compile.triton('triton_per_fused_div_exp_log_mean_mul_neg_sum_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 64],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_div_exp_log_mean_mul_neg_sum_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 20, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_div_exp_log_mean_mul_neg_sum_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1
rnumel = 64
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
roffset = 0
rmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = (rindex // 16)
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0 + (256*r1)), None)
tmp1 = tl.load(in_ptr0 + (16 + r0 + (256*r1)), None)
tmp3 = tl.load(in_ptr0 + (32 + r0 + (256*r1)), None)
tmp5 = tl.load(in_ptr0 + (48 + r0 + (256*r1)), None)
tmp10 = tl.load(in_ptr1 + (r0 + (64*r1)), None)
tmp12 = tl.load(in_ptr0 + (64 + r0 + (256*r1)), None)
tmp13 = tl.load(in_ptr0 + (80 + r0 + (256*r1)), None)
tmp15 = tl.load(in_ptr0 + (96 + r0 + (256*r1)), None)
tmp17 = tl.load(in_ptr0 + (112 + r0 + (256*r1)), None)
tmp21 = tl.load(in_ptr1 + (16 + r0 + (64*r1)), None)
tmp24 = tl.load(in_ptr0 + (128 + r0 + (256*r1)), None)
tmp25 = tl.load(in_ptr0 + (144 + r0 + (256*r1)), None)
tmp27 = tl.load(in_ptr0 + (160 + r0 + (256*r1)), None)
tmp29 = tl.load(in_ptr0 + (176 + r0 + (256*r1)), None)
tmp33 = tl.load(in_ptr1 + (32 + r0 + (64*r1)), None)
tmp36 = tl.load(in_ptr0 + (192 + r0 + (256*r1)), None)
tmp37 = tl.load(in_ptr0 + (208 + r0 + (256*r1)), None)
tmp39 = tl.load(in_ptr0 + (224 + r0 + (256*r1)), None)
tmp41 = tl.load(in_ptr0 + (240 + r0 + (256*r1)), None)
tmp45 = tl.load(in_ptr1 + (48 + r0 + (64*r1)), None)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 2.0
tmp8 = tmp6 * tmp7
tmp9 = tl_math.exp(tmp8)
tmp11 = tmp9 * tmp10
tmp14 = tmp12 + tmp13
tmp16 = tmp14 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp18 * tmp7
tmp20 = tl_math.exp(tmp19)
tmp22 = tmp20 * tmp21
tmp23 = tmp11 + tmp22
tmp26 = tmp24 + tmp25
tmp28 = tmp26 + tmp27
tmp30 = tmp28 + tmp29
tmp31 = tmp30 * tmp7
tmp32 = tl_math.exp(tmp31)
tmp34 = tmp32 * tmp33
tmp35 = tmp23 + tmp34
tmp38 = tmp36 + tmp37
tmp40 = tmp38 + tmp39
tmp42 = tmp40 + tmp41
tmp43 = tmp42 * tmp7
tmp44 = tl_math.exp(tmp43)
tmp46 = tmp44 * tmp45
tmp47 = tmp35 + tmp46
tmp48 = tmp9 + tmp20
tmp49 = tmp48 + tmp32
tmp50 = tmp49 + tmp44
tmp51 = tmp47 / tmp50
tmp52 = tl_math.log(tmp51)
tmp53 = -tmp52
tmp54 = tl.broadcast_to(tmp53, [XBLOCK, RBLOCK])
tmp56 = tl.sum(tmp54, 1)[:, None]
tmp57 = 64.0
tmp58 = tmp56 / tmp57
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp58, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [emb], Original ATen: [aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_div_0.run(arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4, 4), (256, 64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sim_mat], Original ATen: [aten.linalg_vector_norm, aten.clamp_min, aten.div, aten.mul]
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1.run(buf0, buf1, 1024, grid=grid(1024), stream=stream0)
del buf0
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [sim_mat, truediv, sim_mat_1, mul, pos, sum_2, inner, log, loss, loss_1], Original ATen: [aten.sum, aten.div, aten.exp, aten.mul, aten.log, aten.neg, aten.mean]
triton_per_fused_div_exp_log_mean_mul_neg_sum_2.run(buf5, buf1, arg1_1, 1, 64, grid=grid(1), stream=stream0)
del arg1_1
del buf1
return (buf5, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class NTXent(nn.Module):
def __init__(self, metric: 'str'='CosineSimilarity', temperature:
'float'=0.5, reduction: 'str'='mean'):
super().__init__()
if metric not in ['CosineSimilarity']:
raise ValueError('Undefined metric!')
if reduction not in ['AvgNonInf', 'mean', 'sum', 'none']:
raise ValueError('Undefined reduction!')
self.metric = metric
self.t = temperature
self.reduction = reduction
def forward(self, embedding, label):
if self.metric == 'CosineSimilarity':
emb = F.normalize(embedding, dim=1)
sim_mat = F.cosine_similarity(emb.unsqueeze(1), emb.unsqueeze(0
), dim=2)
sim_mat = torch.exp(sim_mat / self.t)
pos = torch.sum(sim_mat * label, dim=1)
inner = pos / torch.sum(sim_mat, dim=1)
loss = -torch.log(inner)
if self.reduction == 'AvgNonInf':
non_inf = inner > 0
loss = loss * non_inf
loss = loss.sum() / torch.sum(non_inf) if torch.sum(non_inf
) > 0 else loss.mean()
elif self.reduction == 'mean':
loss = loss.mean()
elif self.reduction == 'sum':
loss = loss.sum()
elif self.reduction == 'none':
pass
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_div_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x3, tmp15, xmask)
@triton.jit
def triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1(in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 256
x4 = xindex % 64
x0 = xindex % 16
x6 = xindex % 256
x2 = xindex // 64 % 4
x7 = xindex
tmp0 = tl.load(in_ptr0 + (x4 + 64 * x3), xmask, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x3), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x3), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x3), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x3), xmask, eviction_policy=
'evict_last')
tmp16 = tl.load(in_ptr0 + x6, xmask, eviction_policy='evict_last')
tmp17 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp19 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-08
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tmp18 = tmp17 * tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = libdevice.sqrt(tmp27)
tmp29 = triton_helpers.maximum(tmp28, tmp13)
tmp30 = tmp16 / tmp29
tmp31 = tmp15 * tmp30
tl.store(out_ptr0 + x7, tmp31, xmask)
@triton.jit
def triton_per_fused_div_exp_log_mean_mul_neg_sum_2(in_out_ptr0, in_ptr0,
in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 256 * r1), None)
tmp1 = tl.load(in_ptr0 + (16 + r0 + 256 * r1), None)
tmp3 = tl.load(in_ptr0 + (32 + r0 + 256 * r1), None)
tmp5 = tl.load(in_ptr0 + (48 + r0 + 256 * r1), None)
tmp10 = tl.load(in_ptr1 + (r0 + 64 * r1), None)
tmp12 = tl.load(in_ptr0 + (64 + r0 + 256 * r1), None)
tmp13 = tl.load(in_ptr0 + (80 + r0 + 256 * r1), None)
tmp15 = tl.load(in_ptr0 + (96 + r0 + 256 * r1), None)
tmp17 = tl.load(in_ptr0 + (112 + r0 + 256 * r1), None)
tmp21 = tl.load(in_ptr1 + (16 + r0 + 64 * r1), None)
tmp24 = tl.load(in_ptr0 + (128 + r0 + 256 * r1), None)
tmp25 = tl.load(in_ptr0 + (144 + r0 + 256 * r1), None)
tmp27 = tl.load(in_ptr0 + (160 + r0 + 256 * r1), None)
tmp29 = tl.load(in_ptr0 + (176 + r0 + 256 * r1), None)
tmp33 = tl.load(in_ptr1 + (32 + r0 + 64 * r1), None)
tmp36 = tl.load(in_ptr0 + (192 + r0 + 256 * r1), None)
tmp37 = tl.load(in_ptr0 + (208 + r0 + 256 * r1), None)
tmp39 = tl.load(in_ptr0 + (224 + r0 + 256 * r1), None)
tmp41 = tl.load(in_ptr0 + (240 + r0 + 256 * r1), None)
tmp45 = tl.load(in_ptr1 + (48 + r0 + 64 * r1), None)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 2.0
tmp8 = tmp6 * tmp7
tmp9 = tl_math.exp(tmp8)
tmp11 = tmp9 * tmp10
tmp14 = tmp12 + tmp13
tmp16 = tmp14 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp18 * tmp7
tmp20 = tl_math.exp(tmp19)
tmp22 = tmp20 * tmp21
tmp23 = tmp11 + tmp22
tmp26 = tmp24 + tmp25
tmp28 = tmp26 + tmp27
tmp30 = tmp28 + tmp29
tmp31 = tmp30 * tmp7
tmp32 = tl_math.exp(tmp31)
tmp34 = tmp32 * tmp33
tmp35 = tmp23 + tmp34
tmp38 = tmp36 + tmp37
tmp40 = tmp38 + tmp39
tmp42 = tmp40 + tmp41
tmp43 = tmp42 * tmp7
tmp44 = tl_math.exp(tmp43)
tmp46 = tmp44 * tmp45
tmp47 = tmp35 + tmp46
tmp48 = tmp9 + tmp20
tmp49 = tmp48 + tmp32
tmp50 = tmp49 + tmp44
tmp51 = tmp47 / tmp50
tmp52 = tl_math.log(tmp51)
tmp53 = -tmp52
tmp54 = tl.broadcast_to(tmp53, [XBLOCK, RBLOCK])
tmp56 = tl.sum(tmp54, 1)[:, None]
tmp57 = 64.0
tmp58 = tmp56 / tmp57
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp58, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_0[grid(256)](arg0_1, buf0, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4, 4, 4, 4), (256, 64, 16, 4, 1),
torch.float32)
triton_poi_fused_clamp_min_div_linalg_vector_norm_mul_1[grid(1024)](
buf0, buf1, 1024, XBLOCK=128, num_warps=4, num_stages=1)
del buf0
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4
del buf4
triton_per_fused_div_exp_log_mean_mul_neg_sum_2[grid(1)](buf5, buf1,
arg1_1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg1_1
del buf1
return buf5,
class NTXentNew(nn.Module):
def __init__(self, metric: 'str'='CosineSimilarity', temperature:
'float'=0.5, reduction: 'str'='mean'):
super().__init__()
if metric not in ['CosineSimilarity']:
raise ValueError('Undefined metric!')
if reduction not in ['AvgNonInf', 'mean', 'sum', 'none']:
raise ValueError('Undefined reduction!')
self.metric = metric
self.t = temperature
self.reduction = reduction
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| Meteor-han/ReLMole | NTXent | false | 5,592 | [
"MIT"
] | 1 | ec8f2d3ec7b8edb6cd34aede36a980bab3dc35c2 | https://github.com/Meteor-han/ReLMole/tree/ec8f2d3ec7b8edb6cd34aede36a980bab3dc35c2 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, metric: 'str'='CosineSimilarity', temperature:
'float'=0.5, reduction: 'str'='mean'):
super().__init__()
if metric not in ['CosineSimilarity']:
raise ValueError('Undefined metric!')
if reduction not in ['AvgNonInf', 'mean', 'sum', 'none']:
raise ValueError('Undefined reduction!')
self.metric = metric
self.t = temperature
self.reduction = reduction
def forward(self, embedding, label):
if self.metric == 'CosineSimilarity':
emb = F.normalize(embedding, dim=1)
sim_mat = F.cosine_similarity(emb.unsqueeze(1), emb.unsqueeze(0
), dim=2)
sim_mat = torch.exp(sim_mat / self.t)
pos = torch.sum(sim_mat * label, dim=1)
inner = pos / torch.sum(sim_mat, dim=1)
loss = -torch.log(inner)
if self.reduction == 'AvgNonInf':
non_inf = inner > 0
loss = loss * non_inf
loss = loss.sum() / torch.sum(non_inf) if torch.sum(non_inf
) > 0 else loss.mean()
elif self.reduction == 'mean':
loss = loss.mean()
elif self.reduction == 'sum':
loss = loss.sum()
elif self.reduction == 'none':
pass
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GlobalAttention | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/r6/cr6neze6yovkog6kjrk5k2db63h47ozkojywfys6karxe7dlumrz.py
# Topologically Sorted Source Nodes: [align_vectors], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# align_vectors => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x2), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/kj/ckjtlefzavjukjsytvkak6ek26zmzexpcbnlwelx4k5kascjxlf3.py
# Topologically Sorted Source Nodes: [align_vectors], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# align_vectors => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ip/cip3p4ibqio6uu76ccsemd7wjusq5ptlow3dt2zxzouyuz2sqywf.py
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# cat => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%bmm_1, %primals_1], 2), kwargs = {})
triton_poi_fused_cat_2 = async_compile.triton('triton_poi_fused_cat_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = (xindex // 8)
x2 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((4*x1) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((4*x1) + ((-4) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x2), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/f5/cf5pnuv5il7avsmzck3quom7r6zvcfuulsdwpzlv2epzfmcgqgwb.py
# Topologically Sorted Source Nodes: [attn_h_2], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# attn_h_2 => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_2,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_3 = async_compile.triton('triton_poi_fused_clone_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4) % 4
x2 = (xindex // 16)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (4*x2) + (16*x1)), xmask)
tmp1 = libdevice.tanh(tmp0)
tl.store(out_ptr0 + (x3), tmp1, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/u4/cu4fypgfipklcxtitafatnyqdaatx5tws6qfndqotcy4qivcph6d.py
# Topologically Sorted Source Nodes: [align_vectors_2], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# align_vectors_2 => clone_1
# Graph fragment:
# %clone_1 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_3,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_4 = async_compile.triton('triton_poi_fused_clone_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_4(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = (xindex // 4) % 4
x2 = (xindex // 16)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (4*x2) + (16*x1)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [align], Original ATen: [aten.bmm]
extern_kernels.bmm(primals_1, reinterpret_tensor(primals_2, (4, 4, 4), (16, 1, 4), 0), out=buf0)
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [align_vectors], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf0, buf1, 64, grid=grid(64), stream=stream0)
buf2 = reinterpret_tensor(buf0, (16, 4), (4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [align_vectors], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf1, buf2, 64, grid=grid(64), stream=stream0)
buf3 = reinterpret_tensor(buf1, (4, 4, 4), (16, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [c], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf2, (4, 4, 4), (16, 4, 1), 0), primals_2, out=buf3)
del primals_2
buf4 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat], Original ATen: [aten.cat]
triton_poi_fused_cat_2.run(buf3, primals_1, buf4, 128, grid=grid(128), stream=stream0)
del primals_1
buf5 = reinterpret_tensor(buf3, (16, 4), (4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf4, (16, 8), (8, 1), 0), reinterpret_tensor(primals_3, (8, 4), (1, 8), 0), out=buf5)
del primals_3
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [attn_h_2], Original ATen: [aten.clone]
triton_poi_fused_clone_3.run(buf5, buf6, 64, grid=grid(64), stream=stream0)
buf7 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [align_vectors_2], Original ATen: [aten.clone]
triton_poi_fused_clone_4.run(buf2, buf7, 64, grid=grid(64), stream=stream0)
del buf2
return (buf6, buf7, reinterpret_tensor(buf4, (16, 8), (8, 1), 0), buf5, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 8), (8, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.cuda
import torch.distributed
def aeq(*args):
"""
Assert all arguments have the same value
"""
arguments = (arg for arg in args)
first = next(arguments)
assert all(arg == first for arg in arguments
), 'Not all arguments have the same value: ' + str(args)
def sequence_mask(lengths, max_len=None):
"""
Creates a boolean mask from sequence lengths.
"""
batch_size = lengths.numel()
max_len = max_len or lengths.max()
return torch.arange(0, max_len, device=lengths.device).type_as(lengths
).repeat(batch_size, 1).lt(lengths.unsqueeze(1))
class GlobalAttention(nn.Module):
"""
Global attention takes a matrix and a query vector. It
then computes a parameterized convex combination of the matrix
based on the input query.
Constructs a unit mapping a query `q` of size `dim`
and a source matrix `H` of size `n x dim`, to an output
of size `dim`.
.. mermaid::
graph BT
A[Query]
subgraph RNN
C[H 1]
D[H 2]
E[H N]
end
F[Attn]
G[Output]
A --> F
C --> F
D --> F
E --> F
C -.-> G
D -.-> G
E -.-> G
F --> G
All models compute the output as
:math:`c = \\sum_{j=1}^{\\text{SeqLength}} a_j H_j` where
:math:`a_j` is the softmax of a score function.
Then then apply a projection layer to [q, c].
However they
differ on how they compute the attention score.
* Luong Attention (dot, general):
* dot: :math:`\\text{score}(H_j,q) = H_j^T q`
* general: :math:`\\text{score}(H_j, q) = H_j^T W_a q`
* Bahdanau Attention (mlp):
* :math:`\\text{score}(H_j, q) = v_a^T \\text{tanh}(W_a q + U_a h_j)`
Args:
dim (int): dimensionality of query and key
coverage (bool): use coverage term
attn_type (str): type of attention to use, options [dot,general,mlp]
attn_func (str): attention function to use, options [softmax,sparsemax]
"""
def __init__(self, dim, coverage=False, attn_type='dot', attn_func=
'softmax'):
super(GlobalAttention, self).__init__()
self.dim = dim
assert attn_type in ['dot', 'general', 'mlp'
], 'Please select a valid attention type (got {:s}).'.format(
attn_type)
self.attn_type = attn_type
assert attn_func in ['softmax', 'sparsemax'
], 'Please select a valid attention function.'
self.attn_func = attn_func
if self.attn_type == 'general':
self.linear_in = nn.Linear(dim, dim, bias=False)
elif self.attn_type == 'mlp':
self.linear_context = nn.Linear(dim, dim, bias=False)
self.linear_query = nn.Linear(dim, dim, bias=True)
self.v = nn.Linear(dim, 1, bias=False)
out_bias = self.attn_type == 'mlp'
self.linear_out = nn.Linear(dim * 2, dim, bias=out_bias)
if coverage:
self.linear_cover = nn.Linear(1, dim, bias=False)
def score(self, h_t, h_s):
"""
Args:
h_t (FloatTensor): sequence of queries ``(batch, tgt_len, dim)``
h_s (FloatTensor): sequence of sources ``(batch, src_len, dim``
Returns:
FloatTensor: raw attention scores (unnormalized) for each src index
``(batch, tgt_len, src_len)``
"""
src_batch, src_len, src_dim = h_s.size()
tgt_batch, tgt_len, tgt_dim = h_t.size()
aeq(src_batch, tgt_batch)
aeq(src_dim, tgt_dim)
aeq(self.dim, src_dim)
if self.attn_type in ['general', 'dot']:
if self.attn_type == 'general':
h_t_ = h_t.view(tgt_batch * tgt_len, tgt_dim)
h_t_ = self.linear_in(h_t_)
h_t = h_t_.view(tgt_batch, tgt_len, tgt_dim)
h_s_ = h_s.transpose(1, 2)
return torch.bmm(h_t, h_s_)
else:
dim = self.dim
wq = self.linear_query(h_t.view(-1, dim))
wq = wq.view(tgt_batch, tgt_len, 1, dim)
wq = wq.expand(tgt_batch, tgt_len, src_len, dim)
uh = self.linear_context(h_s.contiguous().view(-1, dim))
uh = uh.view(src_batch, 1, src_len, dim)
uh = uh.expand(src_batch, tgt_len, src_len, dim)
wquh = torch.tanh(wq + uh)
return self.v(wquh.view(-1, dim)).view(tgt_batch, tgt_len, src_len)
def forward(self, source, memory_bank, memory_lengths=None, coverage=None):
"""
Args:
source (FloatTensor): query vectors ``(batch, tgt_len, dim)``
memory_bank (FloatTensor): source vectors ``(batch, src_len, dim)``
memory_lengths (LongTensor): the source context lengths ``(batch,)``
coverage (FloatTensor): None (not supported yet)
Returns:
(FloatTensor, FloatTensor):
* Computed vector ``(tgt_len, batch, dim)``
* Attention distribtutions for each query
``(tgt_len, batch, src_len)``
"""
if source.dim() == 2:
one_step = True
source = source.unsqueeze(1)
else:
one_step = False
batch, source_l, dim = memory_bank.size()
batch_, target_l, dim_ = source.size()
aeq(batch, batch_)
aeq(dim, dim_)
aeq(self.dim, dim)
if coverage is not None:
batch_, source_l_ = coverage.size()
aeq(batch, batch_)
aeq(source_l, source_l_)
if coverage is not None:
cover = coverage.view(-1).unsqueeze(1)
memory_bank += self.linear_cover(cover).view_as(memory_bank)
memory_bank = torch.tanh(memory_bank)
align = self.score(source, memory_bank)
if memory_lengths is not None:
mask = sequence_mask(memory_lengths, max_len=align.size(-1))
mask = mask.unsqueeze(1)
align.masked_fill_(~mask, -float('inf'))
if self.attn_func == 'softmax':
align_vectors = F.softmax(align.view(batch * target_l, source_l
), -1)
else:
align_vectors = sparsemax(align.view(batch * target_l, source_l
), -1)
align_vectors = align_vectors.view(batch, target_l, source_l)
c = torch.bmm(align_vectors, memory_bank)
concat_c = torch.cat([c, source], 2).view(batch * target_l, dim * 2)
attn_h = self.linear_out(concat_c).view(batch, target_l, dim)
if self.attn_type in ['general', 'dot']:
attn_h = torch.tanh(attn_h)
if one_step:
attn_h = attn_h.squeeze(1)
align_vectors = align_vectors.squeeze(1)
batch_, dim_ = attn_h.size()
aeq(batch, batch_)
aeq(dim, dim_)
batch_, source_l_ = align_vectors.size()
aeq(batch, batch_)
aeq(source_l, source_l_)
else:
attn_h = attn_h.transpose(0, 1).contiguous()
align_vectors = align_vectors.transpose(0, 1).contiguous()
target_l_, batch_, dim_ = attn_h.size()
aeq(target_l, target_l_)
aeq(batch, batch_)
aeq(dim, dim_)
target_l_, batch_, source_l_ = align_vectors.size()
aeq(target_l, target_l_)
aeq(batch, batch_)
aeq(source_l, source_l_)
return attn_h, align_vectors
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'dim': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
import torch.cuda
import torch.distributed
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_cat_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_clone_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4 % 4
x2 = xindex // 16
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 4 * x2 + 16 * x1), xmask)
tmp1 = libdevice.tanh(tmp0)
tl.store(out_ptr0 + x3, tmp1, xmask)
@triton.jit
def triton_poi_fused_clone_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4 % 4
x2 = xindex // 16
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 4 * x2 + 16 * x1), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(primals_1, reinterpret_tensor(primals_2, (4, 4,
4), (16, 1, 4), 0), out=buf0)
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(64)](buf0, buf1, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf2 = reinterpret_tensor(buf0, (16, 4), (4, 1), 0)
del buf0
triton_poi_fused__softmax_1[grid(64)](buf1, buf2, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf3 = reinterpret_tensor(buf1, (4, 4, 4), (16, 4, 1), 0)
del buf1
extern_kernels.bmm(reinterpret_tensor(buf2, (4, 4, 4), (16, 4, 1),
0), primals_2, out=buf3)
del primals_2
buf4 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
triton_poi_fused_cat_2[grid(128)](buf3, primals_1, buf4, 128,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_1
buf5 = reinterpret_tensor(buf3, (16, 4), (4, 1), 0)
del buf3
extern_kernels.mm(reinterpret_tensor(buf4, (16, 8), (8, 1), 0),
reinterpret_tensor(primals_3, (8, 4), (1, 8), 0), out=buf5)
del primals_3
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_clone_3[grid(64)](buf5, buf6, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf7 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_clone_4[grid(64)](buf2, buf7, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf2
return buf6, buf7, reinterpret_tensor(buf4, (16, 8), (8, 1), 0), buf5
def aeq(*args):
"""
Assert all arguments have the same value
"""
arguments = (arg for arg in args)
first = next(arguments)
assert all(arg == first for arg in arguments
), 'Not all arguments have the same value: ' + str(args)
def sequence_mask(lengths, max_len=None):
"""
Creates a boolean mask from sequence lengths.
"""
batch_size = lengths.numel()
max_len = max_len or lengths.max()
return torch.arange(0, max_len, device=lengths.device).type_as(lengths
).repeat(batch_size, 1).lt(lengths.unsqueeze(1))
class GlobalAttentionNew(nn.Module):
"""
Global attention takes a matrix and a query vector. It
then computes a parameterized convex combination of the matrix
based on the input query.
Constructs a unit mapping a query `q` of size `dim`
and a source matrix `H` of size `n x dim`, to an output
of size `dim`.
.. mermaid::
graph BT
A[Query]
subgraph RNN
C[H 1]
D[H 2]
E[H N]
end
F[Attn]
G[Output]
A --> F
C --> F
D --> F
E --> F
C -.-> G
D -.-> G
E -.-> G
F --> G
All models compute the output as
:math:`c = \\sum_{j=1}^{\\text{SeqLength}} a_j H_j` where
:math:`a_j` is the softmax of a score function.
Then then apply a projection layer to [q, c].
However they
differ on how they compute the attention score.
* Luong Attention (dot, general):
* dot: :math:`\\text{score}(H_j,q) = H_j^T q`
* general: :math:`\\text{score}(H_j, q) = H_j^T W_a q`
* Bahdanau Attention (mlp):
* :math:`\\text{score}(H_j, q) = v_a^T \\text{tanh}(W_a q + U_a h_j)`
Args:
dim (int): dimensionality of query and key
coverage (bool): use coverage term
attn_type (str): type of attention to use, options [dot,general,mlp]
attn_func (str): attention function to use, options [softmax,sparsemax]
"""
def __init__(self, dim, coverage=False, attn_type='dot', attn_func=
'softmax'):
super(GlobalAttentionNew, self).__init__()
self.dim = dim
assert attn_type in ['dot', 'general', 'mlp'
], 'Please select a valid attention type (got {:s}).'.format(
attn_type)
self.attn_type = attn_type
assert attn_func in ['softmax', 'sparsemax'
], 'Please select a valid attention function.'
self.attn_func = attn_func
if self.attn_type == 'general':
self.linear_in = nn.Linear(dim, dim, bias=False)
elif self.attn_type == 'mlp':
self.linear_context = nn.Linear(dim, dim, bias=False)
self.linear_query = nn.Linear(dim, dim, bias=True)
self.v = nn.Linear(dim, 1, bias=False)
out_bias = self.attn_type == 'mlp'
self.linear_out = nn.Linear(dim * 2, dim, bias=out_bias)
if coverage:
self.linear_cover = nn.Linear(1, dim, bias=False)
def score(self, h_t, h_s):
"""
Args:
h_t (FloatTensor): sequence of queries ``(batch, tgt_len, dim)``
h_s (FloatTensor): sequence of sources ``(batch, src_len, dim``
Returns:
FloatTensor: raw attention scores (unnormalized) for each src index
``(batch, tgt_len, src_len)``
"""
src_batch, src_len, src_dim = h_s.size()
tgt_batch, tgt_len, tgt_dim = h_t.size()
aeq(src_batch, tgt_batch)
aeq(src_dim, tgt_dim)
aeq(self.dim, src_dim)
if self.attn_type in ['general', 'dot']:
if self.attn_type == 'general':
h_t_ = h_t.view(tgt_batch * tgt_len, tgt_dim)
h_t_ = self.linear_in(h_t_)
h_t = h_t_.view(tgt_batch, tgt_len, tgt_dim)
h_s_ = h_s.transpose(1, 2)
return torch.bmm(h_t, h_s_)
else:
dim = self.dim
wq = self.linear_query(h_t.view(-1, dim))
wq = wq.view(tgt_batch, tgt_len, 1, dim)
wq = wq.expand(tgt_batch, tgt_len, src_len, dim)
uh = self.linear_context(h_s.contiguous().view(-1, dim))
uh = uh.view(src_batch, 1, src_len, dim)
uh = uh.expand(src_batch, tgt_len, src_len, dim)
wquh = torch.tanh(wq + uh)
return self.v(wquh.view(-1, dim)).view(tgt_batch, tgt_len, src_len)
def forward(self, input_0, input_1):
primals_3 = self.linear_out.weight
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3])
return output[0], output[1]
| MaxatTezekbayev/OpenNMT-py-lexical | GlobalAttention | false | 5,593 | [
"MIT"
] | 1 | 44182999b863fc4074d67e0281c5bdab19abddfe | https://github.com/MaxatTezekbayev/OpenNMT-py-lexical/tree/44182999b863fc4074d67e0281c5bdab19abddfe | import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.cuda
import torch.distributed
def aeq(*args):
"""
Assert all arguments have the same value
"""
arguments = (arg for arg in args)
first = next(arguments)
assert all(arg == first for arg in arguments
), 'Not all arguments have the same value: ' + str(args)
def sequence_mask(lengths, max_len=None):
"""
Creates a boolean mask from sequence lengths.
"""
batch_size = lengths.numel()
max_len = max_len or lengths.max()
return torch.arange(0, max_len, device=lengths.device).type_as(lengths
).repeat(batch_size, 1).lt(lengths.unsqueeze(1))
class Model(nn.Module):
"""
Global attention takes a matrix and a query vector. It
then computes a parameterized convex combination of the matrix
based on the input query.
Constructs a unit mapping a query `q` of size `dim`
and a source matrix `H` of size `n x dim`, to an output
of size `dim`.
.. mermaid::
graph BT
A[Query]
subgraph RNN
C[H 1]
D[H 2]
E[H N]
end
F[Attn]
G[Output]
A --> F
C --> F
D --> F
E --> F
C -.-> G
D -.-> G
E -.-> G
F --> G
All models compute the output as
:math:`c = \\sum_{j=1}^{\\text{SeqLength}} a_j H_j` where
:math:`a_j` is the softmax of a score function.
Then then apply a projection layer to [q, c].
However they
differ on how they compute the attention score.
* Luong Attention (dot, general):
* dot: :math:`\\text{score}(H_j,q) = H_j^T q`
* general: :math:`\\text{score}(H_j, q) = H_j^T W_a q`
* Bahdanau Attention (mlp):
* :math:`\\text{score}(H_j, q) = v_a^T \\text{tanh}(W_a q + U_a h_j)`
Args:
dim (int): dimensionality of query and key
coverage (bool): use coverage term
attn_type (str): type of attention to use, options [dot,general,mlp]
attn_func (str): attention function to use, options [softmax,sparsemax]
"""
def __init__(self, dim, coverage=False, attn_type='dot', attn_func=
'softmax'):
super().__init__()
self.dim = dim
assert attn_type in ['dot', 'general', 'mlp'
], 'Please select a valid attention type (got {:s}).'.format(
attn_type)
self.attn_type = attn_type
assert attn_func in ['softmax', 'sparsemax'
], 'Please select a valid attention function.'
self.attn_func = attn_func
if self.attn_type == 'general':
self.linear_in = nn.Linear(dim, dim, bias=False)
elif self.attn_type == 'mlp':
self.linear_context = nn.Linear(dim, dim, bias=False)
self.linear_query = nn.Linear(dim, dim, bias=True)
self.v = nn.Linear(dim, 1, bias=False)
out_bias = self.attn_type == 'mlp'
self.linear_out = nn.Linear(dim * 2, dim, bias=out_bias)
if coverage:
self.linear_cover = nn.Linear(1, dim, bias=False)
def score(self, h_t, h_s):
"""
Args:
h_t (FloatTensor): sequence of queries ``(batch, tgt_len, dim)``
h_s (FloatTensor): sequence of sources ``(batch, src_len, dim``
Returns:
FloatTensor: raw attention scores (unnormalized) for each src index
``(batch, tgt_len, src_len)``
"""
src_batch, src_len, src_dim = h_s.size()
tgt_batch, tgt_len, tgt_dim = h_t.size()
aeq(src_batch, tgt_batch)
aeq(src_dim, tgt_dim)
aeq(self.dim, src_dim)
if self.attn_type in ['general', 'dot']:
if self.attn_type == 'general':
h_t_ = h_t.view(tgt_batch * tgt_len, tgt_dim)
h_t_ = self.linear_in(h_t_)
h_t = h_t_.view(tgt_batch, tgt_len, tgt_dim)
h_s_ = h_s.transpose(1, 2)
return
# ... truncated (>4000 chars) for memory efficiency |
NetModel | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/gi/cgisfp7cjgya6wz3zy6wgsfrei7pmk7oec6olmpydu3ckl4g7flh.py
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_2 => relu
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_3), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1200
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 300
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 784), (784, 1))
assert_size_stride(primals_2, (300, 784), (784, 1))
assert_size_stride(primals_3, (300, ), (1, ))
assert_size_stride(primals_4, (10, 300), (300, 1))
assert_size_stride(primals_5, (10, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 300), (300, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (784, 300), (1, 784), 0), out=buf0)
del primals_2
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, primals_3, 1200, grid=grid(1200), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, buf1, reinterpret_tensor(primals_4, (300, 10), (1, 300), 0), alpha=1, beta=1, out=buf2)
del primals_5
return (buf2, primals_1, buf1, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 784), (784, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((300, 784), (784, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((300, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((10, 300), (300, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((10, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn.functional as F
import torch.utils.data.dataloader
class NetModel(torch.nn.Module):
def __init__(self):
super(NetModel, self).__init__()
self.hidden = torch.nn.Linear(28 * 28, 300)
self.output = torch.nn.Linear(300, 10)
def forward(self, x):
x = x.view(-1, 28 * 28)
x = self.hidden(x)
x = F.relu(x)
x = self.output(x)
return x
def get_inputs():
return [torch.rand([4, 784])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.utils.data.dataloader
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 1200
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 300
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 784), (784, 1))
assert_size_stride(primals_2, (300, 784), (784, 1))
assert_size_stride(primals_3, (300,), (1,))
assert_size_stride(primals_4, (10, 300), (300, 1))
assert_size_stride(primals_5, (10,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 300), (300, 1), torch.float32)
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (784,
300), (1, 784), 0), out=buf0)
del primals_2
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(1200)](buf1, primals_3, 1200, XBLOCK=
128, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
extern_kernels.addmm(primals_5, buf1, reinterpret_tensor(primals_4,
(300, 10), (1, 300), 0), alpha=1, beta=1, out=buf2)
del primals_5
return buf2, primals_1, buf1, primals_4
class NetModelNew(torch.nn.Module):
def __init__(self):
super(NetModelNew, self).__init__()
self.hidden = torch.nn.Linear(28 * 28, 300)
self.output = torch.nn.Linear(300, 10)
def forward(self, input_0):
primals_2 = self.hidden.weight
primals_3 = self.hidden.bias
primals_4 = self.output.weight
primals_5 = self.output.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| Michaelzhouisnotwhite/Learning-Gan | NetModel | false | 5,594 | [
"MIT"
] | 1 | cf1cff1f2afba296489db55f5de9ebb8405feb0e | https://github.com/Michaelzhouisnotwhite/Learning-Gan/tree/cf1cff1f2afba296489db55f5de9ebb8405feb0e | import torch
import torch.nn.functional as F
import torch.utils.data.dataloader
class Model(torch.nn.Module):
def __init__(self):
super().__init__()
self.hidden = torch.nn.Linear(28 * 28, 300)
self.output = torch.nn.Linear(300, 10)
def forward(self, x):
x = x.view(-1, 28 * 28)
x = self.hidden(x)
x = F.relu(x)
x = self.output(x)
return x
def get_inputs():
return [torch.rand([4, 784])]
def get_init_inputs():
return []
|
PairwiseLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/5e/c5eptm6xzwd32mkban3ftbqvjp4mw3gsooweshto2ug6hb7xbg5e.py
# Topologically Sorted Source Nodes: [diff, mul, sum_1], Original ATen: [aten.sub, aten.mul, aten.sum]
# Source node to ATen node mapping:
# diff => sub
# mul => mul
# sum_1 => sum_1
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %sub), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%mul,), kwargs = {})
triton_per_fused_mul_sub_sum_0 = async_compile.triton('triton_per_fused_mul_sub_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mul_sub_sum_0', 'mutated_arg_names': [], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_mul_sub_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [RBLOCK])
tmp6 = triton_helpers.promote_to_tensor(tl.sum(tmp4, 0))
tl.store(out_ptr0 + (tl.full([1], 0, tl.int32)), tmp6, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [diff, mul, sum_1], Original ATen: [aten.sub, aten.mul, aten.sum]
stream0 = get_raw_stream(0)
triton_per_fused_mul_sub_sum_0.run(arg0_1, arg1_1, buf0, 1, 256, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.utils.data
import torch.nn as nn
import torch.nn.parallel
class PairwiseLoss(nn.Module):
def __init__(self):
super(PairwiseLoss, self).__init__()
def forward(self, x, y):
diff = x - y
return torch.sum(diff * diff)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.utils.data
import torch.nn as nn
import torch.nn.parallel
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mul_sub_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [RBLOCK])
tmp6 = triton_helpers.promote_to_tensor(tl.sum(tmp4, 0))
tl.store(out_ptr0 + tl.full([1], 0, tl.int32), tmp6, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
get_raw_stream(0)
triton_per_fused_mul_sub_sum_0[grid(1)](arg0_1, arg1_1, buf0, 1,
256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf0,
class PairwiseLossNew(nn.Module):
def __init__(self):
super(PairwiseLossNew, self).__init__()
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| MinesNicaicai/large-scale-pointcloud-matching | PairwiseLoss | false | 5,595 | [
"MIT"
] | 1 | cfe140f2be1110ed75b6edd27538021e513a31c9 | https://github.com/MinesNicaicai/large-scale-pointcloud-matching/tree/cfe140f2be1110ed75b6edd27538021e513a31c9 | import torch
import torch.utils.data
import torch.nn as nn
import torch.nn.parallel
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, x, y):
diff = x - y
return torch.sum(diff * diff)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
MLP | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/ky/cky64l574tkwxzjewzevqyhty73x4t3q4p6d2tu2humfvstjwiaa.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_1 => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2048
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, None)
tl.store(out_ptr0 + (x2), tmp6, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/f4/cf4n4twxm5q5eh5lcjfpxlmjdbgr6xfkpjnkedxdctexycbqar7w.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_3 => relu_1
# Graph fragment:
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_3,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_1, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_relu_threshold_backward_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ld/cldtk5skh6gtzdd62vyilgjgd55ch7o62ebbhqgbpau5cmhd5sca.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_5 => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_5, [1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_5, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
triton_poi_fused__softmax_2 = async_compile.triton('triton_poi_fused__softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + (x3), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/7v/c7vy54xspuettz5pgulxporznj2yqlyufnh2o2cvg7er4bnu4zox.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# x_5 => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1], True), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_3 = async_compile.triton('triton_poi_fused__softmax_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_3(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x3), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (16, 32), (32, 1))
assert_size_stride(primals_5, (16, ), (1, ))
assert_size_stride(primals_6, (4, 16), (16, 1))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 32), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 32), (512, 128, 32, 1), 0); del buf0 # reuse
buf8 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf8, 2048, grid=grid(2048), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(primals_4, (32, 16), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 16), (256, 64, 16, 1), 0); del buf2 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 16), (256, 64, 16, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_1.run(buf3, primals_5, buf7, 1024, grid=grid(1024), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 16), (16, 1), 0), reinterpret_tensor(primals_6, (16, 4), (1, 16), 0), alpha=1, beta=1, out=buf4)
del primals_7
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf4, buf5, 256, grid=grid(256), stream=stream0)
buf6 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._softmax]
triton_poi_fused__softmax_3.run(buf5, buf6, 256, grid=grid(256), stream=stream0)
del buf5
return (buf6, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(buf3, (64, 16), (16, 1), 0), buf6, primals_6, buf7, primals_4, buf8, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((32, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((16, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 16), (16, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
import torch.utils
class MLP(torch.nn.Module):
def __init__(self, input_dim, output_dim):
super(MLP, self).__init__()
self.d1 = torch.nn.Linear(input_dim, 32)
self.d2 = torch.nn.Linear(32, 16)
self.d3 = torch.nn.Linear(16, output_dim)
self.relu = nn.ReLU()
self.softmax = nn.Softmax(dim=1)
def forward(self, x):
x = self.d1(x)
x = self.relu(x)
x = self.d2(x)
x = self.relu(x)
x = self.d3(x)
x = self.softmax(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_dim': 4, 'output_dim': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch import nn
import torch.utils
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x3, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (16, 32), (32, 1))
assert_size_stride(primals_5, (16,), (1,))
assert_size_stride(primals_6, (4, 16), (16, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 32), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf0
buf8 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf1,
primals_2, buf8, 2048, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 16), (16, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_4, (32, 16), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 16), (256, 64, 16, 1), 0)
del buf2
buf7 = empty_strided_cuda((4, 4, 4, 16), (256, 64, 16, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_1[grid(1024)](buf3,
primals_5, buf7, 1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 16),
(16, 1), 0), reinterpret_tensor(primals_6, (16, 4), (1, 16), 0),
alpha=1, beta=1, out=buf4)
del primals_7
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_2[grid(256)](buf4, buf5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf6 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf4
triton_poi_fused__softmax_3[grid(256)](buf5, buf6, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf5
return buf6, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(
buf3, (64, 16), (16, 1), 0), buf6, primals_6, buf7, primals_4, buf8
class MLPNew(torch.nn.Module):
def __init__(self, input_dim, output_dim):
super(MLPNew, self).__init__()
self.d1 = torch.nn.Linear(input_dim, 32)
self.d2 = torch.nn.Linear(32, 16)
self.d3 = torch.nn.Linear(16, output_dim)
self.relu = nn.ReLU()
self.softmax = nn.Softmax(dim=1)
def forward(self, input_0):
primals_1 = self.d1.weight
primals_2 = self.d1.bias
primals_4 = self.d2.weight
primals_5 = self.d2.bias
primals_6 = self.d3.weight
primals_7 = self.d3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| MichaelLee-ceo/FedSAUC | MLP | false | 5,596 | [
"Apache-2.0"
] | 1 | 8c00008772213562ff6a07bf9fa92c3831713118 | https://github.com/MichaelLee-ceo/FedSAUC/tree/8c00008772213562ff6a07bf9fa92c3831713118 | import torch
from torch import nn
import torch.utils
class Model(torch.nn.Module):
def __init__(self, input_dim, output_dim):
super().__init__()
self.d1 = torch.nn.Linear(input_dim, 32)
self.d2 = torch.nn.Linear(32, 16)
self.d3 = torch.nn.Linear(16, output_dim)
self.relu = nn.ReLU()
self.softmax = nn.Softmax(dim=1)
def forward(self, x):
x = self.d1(x)
x = self.relu(x)
x = self.d2(x)
x = self.relu(x)
x = self.d3(x)
x = self.softmax(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
CNN_DropOut | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/25/c25jycy7e5gprlbgtiyrrn7pxanfat7ddoyg4geoxxk5b3yr55vp.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# x_2 => convolution
# x_3 => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%unsqueeze, %primals_2, %primals_3, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_convolution_relu_0 = async_compile.triton('triton_poi_fused_convolution_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32768],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 21632
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 676) % 8
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ac/cacw4wiptbynpfgqcbrkvsw3g5rwqpp5mybqq7c5biz7ktl56fjr.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# x_4 => getitem, getitem_1
# Graph fragment:
# %getitem : [num_users=2] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets, 0), kwargs = {})
# %getitem_1 : [num_users=1] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets, 1), kwargs = {})
triton_poi_fused_max_pool2d_with_indices_1 = async_compile.triton('triton_poi_fused_max_pool2d_with_indices_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[8192],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i8', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_max_pool2d_with_indices_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 5408
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 13
x3 = (xindex // 13)
x2 = (xindex // 1352)
x4 = xindex % 1352
tmp0 = tl.load(in_ptr0 + ((2*x0) + (52*x3)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (2*x0) + (52*x3)), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (26 + (2*x0) + (52*x3)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (27 + (2*x0) + (52*x3)), xmask, eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = tmp1 > tmp0
tmp8 = tl.full([1], 1, tl.int8)
tmp9 = tl.full([1], 0, tl.int8)
tmp10 = tl.where(tmp7, tmp8, tmp9)
tmp11 = tmp3 > tmp2
tmp12 = tl.full([1], 2, tl.int8)
tmp13 = tl.where(tmp11, tmp12, tmp10)
tmp14 = tmp5 > tmp4
tmp15 = tl.full([1], 3, tl.int8)
tmp16 = tl.where(tmp14, tmp15, tmp13)
tl.store(out_ptr0 + (x4 + (1376*x2)), tmp6, xmask)
tl.store(out_ptr1 + (x4 + (1408*x2)), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/dg/cdgydpu4uv4skm66vxq4n3ixbpxejadpitghdy5bn7jlo5vnue25.py
# Topologically Sorted Source Nodes: [x_5, x_6], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# x_5 => convolution_1
# x_6 => relu_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%getitem, %primals_4, %primals_5, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_1,), kwargs = {})
triton_poi_fused_convolution_relu_2 = async_compile.triton('triton_poi_fused_convolution_relu_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[8192],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 7744
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 121) % 16
x2 = (xindex // 1936)
x4 = xindex % 1936
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x4 + (1952*x2)), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/tk/ctkacddt2xdaiu5gos5eoszgznxszehacgidrhwuf6hou6fyt3aj.py
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# x_7 => _low_memory_max_pool2d_with_offsets_1, getitem_3
# Graph fragment:
# %_low_memory_max_pool2d_with_offsets_1 : [num_users=2] = call_function[target=torch.ops.prims._low_memory_max_pool2d_with_offsets.default](args = (%relu_1, [2, 2], [2, 2], [0, 0], [1, 1], False), kwargs = {})
# %getitem_3 : [num_users=1] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets_1, 1), kwargs = {})
triton_poi_fused_max_pool2d_with_indices_3 = async_compile.triton('triton_poi_fused_max_pool2d_with_indices_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[2048],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i8', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_max_pool2d_with_indices_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_3(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 1600
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 5
x1 = (xindex // 5) % 5
x2 = (xindex // 25) % 16
x3 = (xindex // 400)
x4 = xindex
tmp0 = tl.load(in_ptr0 + ((2*x0) + (22*x1) + (121*x2) + (1952*x3)), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (2*x0) + (22*x1) + (121*x2) + (1952*x3)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (11 + (2*x0) + (22*x1) + (121*x2) + (1952*x3)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (12 + (2*x0) + (22*x1) + (121*x2) + (1952*x3)), xmask, eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tmp8 = tmp7 > tmp6
tmp9 = tl.full([1], 2, tl.int8)
tmp10 = tl.where(tmp8, tmp9, tmp5)
tmp11 = triton_helpers.maximum(tmp7, tmp6)
tmp13 = tmp12 > tmp11
tmp14 = tl.full([1], 3, tl.int8)
tmp15 = tl.where(tmp13, tmp14, tmp10)
tmp16 = triton_helpers.maximum(tmp12, tmp11)
tl.store(out_ptr0 + (x4), tmp15, xmask)
tl.store(out_ptr1 + (x4), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/2y/c2y7nraxg57efqjgaab4g2ywiolmgkkaxdzjcqfjeb2guhagama5.py
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_11 => relu_2
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_7), kwargs = {})
# %relu_2 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor,), kwargs = {})
triton_poi_fused_relu_4 = async_compile.triton('triton_poi_fused_relu_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_4', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_4(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9 = args
args.clear()
assert_size_stride(primals_1, (4, 28, 28), (784, 28, 1))
assert_size_stride(primals_2, (8, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_3, (8, ), (1, ))
assert_size_stride(primals_4, (16, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_5, (16, ), (1, ))
assert_size_stride(primals_6, (64, 400), (400, 1))
assert_size_stride(primals_7, (64, ), (1, ))
assert_size_stride(primals_8, (10, 64), (64, 1))
assert_size_stride(primals_9, (10, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_1, (4, 1, 28, 28), (784, 784, 28, 1), 0), primals_2, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 8, 26, 26), (5408, 676, 26, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_3, 21632, grid=grid(21632), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((4, 8, 13, 13), (1376, 169, 13, 1), torch.float32)
buf3 = empty_strided_cuda((4, 8, 13, 13), (1408, 169, 13, 1), torch.int8)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.max_pool2d_with_indices]
triton_poi_fused_max_pool2d_with_indices_1.run(buf1, buf2, buf3, 5408, grid=grid(5408), stream=stream0)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf2, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 16, 11, 11), (1936, 121, 11, 1))
buf5 = empty_strided_cuda((4, 16, 11, 11), (1952, 121, 11, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_5, x_6], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_2.run(buf4, primals_5, buf5, 7744, grid=grid(7744), stream=stream0)
del buf4
del primals_5
buf6 = empty_strided_cuda((4, 16, 5, 5), (400, 25, 5, 1), torch.int8)
buf7 = empty_strided_cuda((4, 16, 5, 5), (400, 25, 5, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.max_pool2d_with_indices]
triton_poi_fused_max_pool2d_with_indices_3.run(buf5, buf6, buf7, 1600, grid=grid(1600), stream=stream0)
buf8 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf7, (4, 400), (400, 1), 0), reinterpret_tensor(primals_6, (400, 64), (1, 400), 0), out=buf8)
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [x_11], Original ATen: [aten.relu]
triton_poi_fused_relu_4.run(buf9, primals_7, 256, grid=grid(256), stream=stream0)
del primals_7
buf10 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_13], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_9, buf9, reinterpret_tensor(primals_8, (64, 10), (1, 64), 0), alpha=1, beta=1, out=buf10)
del primals_9
return (buf10, primals_2, primals_4, reinterpret_tensor(primals_1, (4, 1, 28, 28), (784, 784, 28, 1), 0), buf1, buf2, buf3, buf5, buf6, reinterpret_tensor(buf7, (4, 400), (400, 1), 0), buf9, primals_8, primals_6, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 28, 28), (784, 28, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((8, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((8, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((16, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 400), (400, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((10, 64), (64, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((10, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
import torch.utils
class CNN_DropOut(torch.nn.Module):
"""
Recommended model by "Adaptive Federated Optimization" (https://arxiv.org/pdf/2003.00295.pdf)
Used for EMNIST experiments.
When `only_digits=True`, the summary of returned model is
```
Model:
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
reshape (Reshape) (None, 28, 28, 1) 0
_________________________________________________________________
conv2d (Conv2D) (None, 26, 26, 32) 320
_________________________________________________________________
conv2d_1 (Conv2D) (None, 24, 24, 64) 18496
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 12, 12, 64) 0
_________________________________________________________________
dropout (Dropout) (None, 12, 12, 64) 0
_________________________________________________________________
flatten (Flatten) (None, 9216) 0
_________________________________________________________________
dense (Dense) (None, 128) 1179776
_________________________________________________________________
dropout_1 (Dropout) (None, 128) 0
_________________________________________________________________
dense_1 (Dense) (None, 10) 1290
=================================================================
Total params: 1,199,882
Trainable params: 1,199,882
Non-trainable params: 0
```
Args:
only_digits: If True, uses a final layer with 10 outputs, for use with the
digits only MNIST dataset (http://yann.lecun.com/exdb/mnist/).
If False, uses 62 outputs for Federated Extended MNIST (FEMNIST)
EMNIST: Extending MNIST to handwritten letters: https://arxiv.org/abs/1702.05373.
Returns:
A `torch.nn.Module`.
"""
def __init__(self, only_digits=True):
super(CNN_DropOut, self).__init__()
self.conv2d_1 = torch.nn.Conv2d(1, 8, kernel_size=3)
self.max_pooling = nn.MaxPool2d(2, stride=2)
self.conv2d_2 = torch.nn.Conv2d(8, 16, kernel_size=3)
self.dropout_1 = nn.Dropout(0.25)
self.flatten = nn.Flatten()
self.linear_1 = nn.Linear(16 * 5 * 5, 64)
self.dropout_2 = nn.Dropout(0.25)
self.linear_2 = nn.Linear(64, 10)
self.relu = nn.ReLU()
def forward(self, x):
x = x.view(x.shape[0], 28, 28)
x = torch.unsqueeze(x, 1)
x = self.conv2d_1(x)
x = self.relu(x)
x = self.max_pooling(x)
x = self.conv2d_2(x)
x = self.relu(x)
x = self.max_pooling(x)
x = self.dropout_1(x)
x = self.flatten(x)
x = self.linear_1(x)
x = self.relu(x)
x = self.dropout_2(x)
x = self.linear_2(x)
return x
def get_inputs():
return [torch.rand([4, 28, 28])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
import torch.utils
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 21632
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 676 % 8
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 5408
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 13
x3 = xindex // 13
x2 = xindex // 1352
x4 = xindex % 1352
tmp0 = tl.load(in_ptr0 + (2 * x0 + 52 * x3), xmask, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 52 * x3), xmask, eviction_policy
='evict_last')
tmp3 = tl.load(in_ptr0 + (26 + 2 * x0 + 52 * x3), xmask,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (27 + 2 * x0 + 52 * x3), xmask,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = tmp1 > tmp0
tmp8 = tl.full([1], 1, tl.int8)
tmp9 = tl.full([1], 0, tl.int8)
tmp10 = tl.where(tmp7, tmp8, tmp9)
tmp11 = tmp3 > tmp2
tmp12 = tl.full([1], 2, tl.int8)
tmp13 = tl.where(tmp11, tmp12, tmp10)
tmp14 = tmp5 > tmp4
tmp15 = tl.full([1], 3, tl.int8)
tmp16 = tl.where(tmp14, tmp15, tmp13)
tl.store(out_ptr0 + (x4 + 1376 * x2), tmp6, xmask)
tl.store(out_ptr1 + (x4 + 1408 * x2), tmp16, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 7744
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 121 % 16
x2 = xindex // 1936
x4 = xindex % 1936
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x4 + 1952 * x2), tmp4, xmask)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_3(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1600
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 5
x1 = xindex // 5 % 5
x2 = xindex // 25 % 16
x3 = xindex // 400
x4 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 22 * x1 + 121 * x2 + 1952 * x3),
xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 22 * x1 + 121 * x2 + 1952 * x3),
xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (11 + 2 * x0 + 22 * x1 + 121 * x2 + 1952 * x3),
xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (12 + 2 * x0 + 22 * x1 + 121 * x2 + 1952 * x3
), xmask, eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tmp8 = tmp7 > tmp6
tmp9 = tl.full([1], 2, tl.int8)
tmp10 = tl.where(tmp8, tmp9, tmp5)
tmp11 = triton_helpers.maximum(tmp7, tmp6)
tmp13 = tmp12 > tmp11
tmp14 = tl.full([1], 3, tl.int8)
tmp15 = tl.where(tmp13, tmp14, tmp10)
tmp16 = triton_helpers.maximum(tmp12, tmp11)
tl.store(out_ptr0 + x4, tmp15, xmask)
tl.store(out_ptr1 + x4, tmp16, xmask)
@triton.jit
def triton_poi_fused_relu_4(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (4, 28, 28), (784, 28, 1))
assert_size_stride(primals_2, (8, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_3, (8,), (1,))
assert_size_stride(primals_4, (16, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_5, (16,), (1,))
assert_size_stride(primals_6, (64, 400), (400, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (10, 64), (64, 1))
assert_size_stride(primals_9, (10,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_1, (4,
1, 28, 28), (784, 784, 28, 1), 0), primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 8, 26, 26), (5408, 676, 26, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(21632)](buf1, primals_3,
21632, XBLOCK=256, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((4, 8, 13, 13), (1376, 169, 13, 1), torch
.float32)
buf3 = empty_strided_cuda((4, 8, 13, 13), (1408, 169, 13, 1), torch
.int8)
triton_poi_fused_max_pool2d_with_indices_1[grid(5408)](buf1, buf2,
buf3, 5408, XBLOCK=128, num_warps=4, num_stages=1)
buf4 = extern_kernels.convolution(buf2, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 16, 11, 11), (1936, 121, 11, 1))
buf5 = empty_strided_cuda((4, 16, 11, 11), (1952, 121, 11, 1),
torch.float32)
triton_poi_fused_convolution_relu_2[grid(7744)](buf4, primals_5,
buf5, 7744, XBLOCK=256, num_warps=4, num_stages=1)
del buf4
del primals_5
buf6 = empty_strided_cuda((4, 16, 5, 5), (400, 25, 5, 1), torch.int8)
buf7 = empty_strided_cuda((4, 16, 5, 5), (400, 25, 5, 1), torch.float32
)
triton_poi_fused_max_pool2d_with_indices_3[grid(1600)](buf5, buf6,
buf7, 1600, XBLOCK=256, num_warps=4, num_stages=1)
buf8 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (4, 400), (400, 1), 0),
reinterpret_tensor(primals_6, (400, 64), (1, 400), 0), out=buf8)
buf9 = buf8
del buf8
triton_poi_fused_relu_4[grid(256)](buf9, primals_7, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_7
buf10 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
extern_kernels.addmm(primals_9, buf9, reinterpret_tensor(primals_8,
(64, 10), (1, 64), 0), alpha=1, beta=1, out=buf10)
del primals_9
return buf10, primals_2, primals_4, reinterpret_tensor(primals_1, (4, 1,
28, 28), (784, 784, 28, 1), 0
), buf1, buf2, buf3, buf5, buf6, reinterpret_tensor(buf7, (4, 400),
(400, 1), 0), buf9, primals_8, primals_6
class CNN_DropOutNew(torch.nn.Module):
"""
Recommended model by "Adaptive Federated Optimization" (https://arxiv.org/pdf/2003.00295.pdf)
Used for EMNIST experiments.
When `only_digits=True`, the summary of returned model is
```
Model:
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
reshape (Reshape) (None, 28, 28, 1) 0
_________________________________________________________________
conv2d (Conv2D) (None, 26, 26, 32) 320
_________________________________________________________________
conv2d_1 (Conv2D) (None, 24, 24, 64) 18496
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 12, 12, 64) 0
_________________________________________________________________
dropout (Dropout) (None, 12, 12, 64) 0
_________________________________________________________________
flatten (Flatten) (None, 9216) 0
_________________________________________________________________
dense (Dense) (None, 128) 1179776
_________________________________________________________________
dropout_1 (Dropout) (None, 128) 0
_________________________________________________________________
dense_1 (Dense) (None, 10) 1290
=================================================================
Total params: 1,199,882
Trainable params: 1,199,882
Non-trainable params: 0
```
Args:
only_digits: If True, uses a final layer with 10 outputs, for use with the
digits only MNIST dataset (http://yann.lecun.com/exdb/mnist/).
If False, uses 62 outputs for Federated Extended MNIST (FEMNIST)
EMNIST: Extending MNIST to handwritten letters: https://arxiv.org/abs/1702.05373.
Returns:
A `torch.nn.Module`.
"""
def __init__(self, only_digits=True):
super(CNN_DropOutNew, self).__init__()
self.conv2d_1 = torch.nn.Conv2d(1, 8, kernel_size=3)
self.max_pooling = nn.MaxPool2d(2, stride=2)
self.conv2d_2 = torch.nn.Conv2d(8, 16, kernel_size=3)
self.dropout_1 = nn.Dropout(0.25)
self.flatten = nn.Flatten()
self.linear_1 = nn.Linear(16 * 5 * 5, 64)
self.dropout_2 = nn.Dropout(0.25)
self.linear_2 = nn.Linear(64, 10)
self.relu = nn.ReLU()
def forward(self, input_0):
primals_2 = self.conv2d_1.weight
primals_3 = self.conv2d_1.bias
primals_4 = self.conv2d_2.weight
primals_5 = self.conv2d_2.bias
primals_6 = self.linear_1.weight
primals_7 = self.linear_1.bias
primals_8 = self.linear_2.weight
primals_9 = self.linear_2.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
| MichaelLee-ceo/FedSAUC | CNN_DropOut | false | 5,597 | [
"Apache-2.0"
] | 1 | 8c00008772213562ff6a07bf9fa92c3831713118 | https://github.com/MichaelLee-ceo/FedSAUC/tree/8c00008772213562ff6a07bf9fa92c3831713118 | import torch
from torch import nn
import torch.utils
class Model(torch.nn.Module):
"""
Recommended model by "Adaptive Federated Optimization" (https://arxiv.org/pdf/2003.00295.pdf)
Used for EMNIST experiments.
When `only_digits=True`, the summary of returned model is
```
Model:
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
reshape (Reshape) (None, 28, 28, 1) 0
_________________________________________________________________
conv2d (Conv2D) (None, 26, 26, 32) 320
_________________________________________________________________
conv2d_1 (Conv2D) (None, 24, 24, 64) 18496
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 12, 12, 64) 0
_________________________________________________________________
dropout (Dropout) (None, 12, 12, 64) 0
_________________________________________________________________
flatten (Flatten) (None, 9216) 0
_________________________________________________________________
dense (Dense) (None, 128) 1179776
_________________________________________________________________
dropout_1 (Dropout) (None, 128) 0
_________________________________________________________________
dense_1 (Dense) (None, 10) 1290
=================================================================
Total params: 1,199,882
Trainable params: 1,199,882
Non-trainable params: 0
```
Args:
only_digits: If True, uses a final layer with 10 outputs, for use with the
digits only MNIST dataset (http://yann.lecun.com/exdb/mnist/).
If False, uses 62 outputs for Federated Extended MNIST (FEMNIST)
EMNIST: Extending MNIST to handwritten letters: https://arxiv.org/abs/1702.05373.
Returns:
A `torch.nn.Module`.
"""
def __init__(self, only_digits=True):
super().__init__()
self.conv2d_1 = torch.nn.Conv2d(1, 8, kernel_size=3)
self.max_pooling = nn.MaxPool2d(2, stride=2)
self.conv2d_2 = torch.nn.Conv2d(8, 16, kernel_size=3)
self.dropout_1 = nn.Dropout(0.25)
self.flatten = nn.Flatten()
self.linear_1 = nn.Linear(16 * 5 * 5, 64)
self.dropout_2 = nn.Dropout(0.25)
self.linear_2 = nn.Linear(64, 10)
self.relu = nn.ReLU()
def forward(self, x):
x = x.view(x.shape[0], 28, 28)
x = torch.unsqueeze(x, 1)
x = self.conv2d_1(x)
x = self.relu(x)
x = self.max_pooling(x)
x = self.conv2d_2(x)
x = self.relu(x)
x = self.max_pooling(x)
x = self.dropout_1(x)
x = self.flatten(x)
x = self.linear_1(x)
x = self.relu(x)
x = self.dropout_2(x)
x = self.linear_2(x)
return x
def get_inputs():
return [torch.rand([4, 28, 28])]
def get_init_inputs():
return []
|
ContrastiveLoss | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/lh/clhtaboxxs526aw4bqcb7s6xoig5vzwco55tfg6waaga3ao3elgd.py
# Topologically Sorted Source Nodes: [euclidean_distance], Original ATen: [aten.sub, aten.add, aten.norm]
# Source node to ATen node mapping:
# euclidean_distance => add, pow_1, pow_2, sub, sum_1
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg1_1, %arg0_1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Scalar](args = (%sub, 1e-06), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%add, 2.0), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [3]), kwargs = {})
# %pow_2 : [num_users=2] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 0.5), kwargs = {})
triton_poi_fused_add_norm_sub_0 = async_compile.triton('triton_poi_fused_add_norm_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_norm_sub_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 8, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_norm_sub_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp3 = 1e-06
tmp4 = tmp2 + tmp3
tmp5 = tmp4 * tmp4
tmp8 = tmp6 - tmp7
tmp9 = tmp8 + tmp3
tmp10 = tmp9 * tmp9
tmp11 = tmp5 + tmp10
tmp14 = tmp12 - tmp13
tmp15 = tmp14 + tmp3
tmp16 = tmp15 * tmp15
tmp17 = tmp11 + tmp16
tmp20 = tmp18 - tmp19
tmp21 = tmp20 + tmp3
tmp22 = tmp21 * tmp21
tmp23 = tmp17 + tmp22
tmp24 = libdevice.sqrt(tmp23)
tl.store(out_ptr0 + (x0), tmp24, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/2l/c2l2efhstxzxtce6u6qjwkvh7vawevxafq4szecegv75l7dernch.py
# Topologically Sorted Source Nodes: [sub, pow_1, mul, sub_1, clamp, pow_2, mul_1, add, loss_contrastive], Original ATen: [aten.rsub, aten.pow, aten.mul, aten.clamp, aten.add, aten.mean]
# Source node to ATen node mapping:
# add => add_1
# clamp => clamp_min
# loss_contrastive => mean
# mul => mul
# mul_1 => mul_1
# pow_1 => pow_3
# pow_2 => pow_4
# sub => sub_1
# sub_1 => sub_2
# Graph fragment:
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %arg2_1), kwargs = {})
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%pow_2, 2), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %pow_3), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (2.0, %pow_2), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_2, 0.0), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%clamp_min, 2), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg2_1, %pow_4), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %mul_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%add_1,), kwargs = {})
triton_per_fused_add_clamp_mean_mul_pow_rsub_1 = async_compile.triton('triton_per_fused_add_clamp_mean_mul_pow_rsub_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.persistent_reduction(
size_hints=[1, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {3: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=(3,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_clamp_mean_mul_pow_rsub_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, 'num_load': 2, 'num_reduction': 1, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False}
)
@triton.jit
def triton_per_fused_add_clamp_mean_mul_pow_rsub_1(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
xnumel = 1
XBLOCK: tl.constexpr = 1
rnumel = 256
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
xmask = tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
roffset = 0
rmask = tl.full([RBLOCK], True, tl.int1)
r2 = rindex
r0 = rindex % 64
tmp0 = tl.load(in_ptr0 + (r2), None)
tmp3 = tl.load(in_ptr1 + (r0), None, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp3 * tmp3
tmp5 = tmp2 * tmp4
tmp6 = 2.0
tmp7 = tmp6 - tmp3
tmp8 = 0.0
tmp9 = triton_helpers.maximum(tmp7, tmp8)
tmp10 = tmp9 * tmp9
tmp11 = tmp0 * tmp10
tmp12 = tmp5 + tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 256.0
tmp17 = tmp15 / tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp17, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [euclidean_distance], Original ATen: [aten.sub, aten.add, aten.norm]
stream0 = get_raw_stream(0)
triton_poi_fused_add_norm_sub_0.run(arg1_1, arg0_1, buf0, 64, grid=grid(64), stream=stream0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [sub, pow_1, mul, sub_1, clamp, pow_2, mul_1, add, loss_contrastive], Original ATen: [aten.rsub, aten.pow, aten.mul, aten.clamp, aten.add, aten.mean]
triton_per_fused_add_clamp_mean_mul_pow_rsub_1.run(buf2, arg2_1, buf0, 1, 256, grid=grid(1), stream=stream0)
del arg2_1
del buf0
return (buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg2_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1, arg2_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.utils.data
import torch.nn.functional as F
import torch.nn.parallel
class ContrastiveLoss(torch.nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=2.0):
super(ContrastiveLoss, self).__init__()
self.margin = margin
def forward(self, output1, output2, is_negative):
euclidean_distance = F.pairwise_distance(output1, output2)
loss_contrastive = torch.mean((1 - is_negative) * torch.pow(
euclidean_distance, 2) + is_negative * torch.pow(torch.clamp(
self.margin - euclidean_distance, min=0.0), 2))
return loss_contrastive
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.utils.data
import torch.nn.parallel
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_norm_sub_0(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp18 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 - tmp1
tmp3 = 1e-06
tmp4 = tmp2 + tmp3
tmp5 = tmp4 * tmp4
tmp8 = tmp6 - tmp7
tmp9 = tmp8 + tmp3
tmp10 = tmp9 * tmp9
tmp11 = tmp5 + tmp10
tmp14 = tmp12 - tmp13
tmp15 = tmp14 + tmp3
tmp16 = tmp15 * tmp15
tmp17 = tmp11 + tmp16
tmp20 = tmp18 - tmp19
tmp21 = tmp20 + tmp3
tmp22 = tmp21 * tmp21
tmp23 = tmp17 + tmp22
tmp24 = libdevice.sqrt(tmp23)
tl.store(out_ptr0 + x0, tmp24, xmask)
@triton.jit
def triton_per_fused_add_clamp_mean_mul_pow_rsub_1(in_out_ptr0, in_ptr0,
in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r2 = rindex
r0 = rindex % 64
tmp0 = tl.load(in_ptr0 + r2, None)
tmp3 = tl.load(in_ptr1 + r0, None, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp3 * tmp3
tmp5 = tmp2 * tmp4
tmp6 = 2.0
tmp7 = tmp6 - tmp3
tmp8 = 0.0
tmp9 = triton_helpers.maximum(tmp7, tmp8)
tmp10 = tmp9 * tmp9
tmp11 = tmp0 * tmp10
tmp12 = tmp5 + tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 256.0
tmp17 = tmp15 / tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp17, None)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_norm_sub_0[grid(64)](arg1_1, arg0_1, buf0, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1
del buf1
triton_per_fused_add_clamp_mean_mul_pow_rsub_1[grid(1)](buf2,
arg2_1, buf0, 1, 256, num_warps=2, num_stages=1)
del arg2_1
del buf0
return buf2,
class ContrastiveLossNew(torch.nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=2.0):
super(ContrastiveLossNew, self).__init__()
self.margin = margin
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0]
| MinesNicaicai/large-scale-pointcloud-matching | ContrastiveLoss | false | 5,598 | [
"MIT"
] | 1 | cfe140f2be1110ed75b6edd27538021e513a31c9 | https://github.com/MinesNicaicai/large-scale-pointcloud-matching/tree/cfe140f2be1110ed75b6edd27538021e513a31c9 | import torch
import torch.utils.data
import torch.nn.functional as F
import torch.nn.parallel
class Model(torch.nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=2.0):
super().__init__()
self.margin = margin
def forward(self, output1, output2, is_negative):
euclidean_distance = F.pairwise_distance(output1, output2)
loss_contrastive = torch.mean((1 - is_negative) * torch.pow(
euclidean_distance, 2) + is_negative * torch.pow(torch.clamp(
self.margin - euclidean_distance, min=0.0), 2))
return loss_contrastive
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return []
|
Attention | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/nu/cnuc7ivckuuly7yn2763pwt3sw72jd6vuwpeeu4sfespm5iz7fq4.py
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# p_attn => exp
# Graph fragment:
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_2, 1), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor, %amax_default), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor, 2.0), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp3 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + (x2), tmp17, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/fj/cfjl47pvhwbpfbvh6rfehwy5ijxc5p3zgkld2lwf3mw5bl6pbkak.py
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# p_attn => div_1, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(arg0_1, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(arg1_1, (16, 4, 4), (16, 1, 4), 0), out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf1, buf2, 256, grid=grid(256), stream=stream0)
buf3 = reinterpret_tensor(buf1, (16, 4, 4), (16, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [matmul_1], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf3)
del arg2_1
return (reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg2_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1, arg2_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class Attention(nn.Module):
def forward(self, query, key, value, mask=None, dropout=None):
"""Compute 'Scaled Dot Product Attention'"""
d_k = query.size(-1)
scores = torch.matmul(query, key.transpose(-2, -1)) / math.sqrt(d_k)
if mask is not None:
scores = scores.masked_fill(mask == 0, -1000000000.0)
p_attn = F.softmax(scores, dim=-1)
if dropout is not None:
p_attn = dropout(p_attn)
return torch.matmul(p_attn, value), p_attn
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp3 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + x2, tmp17, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(arg0_1, (16, 4, 4), (16, 4, 1
), 0), reinterpret_tensor(arg1_1, (16, 4, 4), (16, 1, 4), 0),
out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
triton_poi_fused__softmax_1[grid(256)](buf1, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf3 = reinterpret_tensor(buf1, (16, 4, 4), (16, 4, 1), 0)
del buf1
extern_kernels.bmm(reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf3
)
del arg2_1
return reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf2
class AttentionNew(nn.Module):
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0], output[1]
| Moymix/BERT-pytorch | Attention | false | 5,599 | [
"Apache-2.0"
] | 1 | f0b9c3ae53e05c00adcc761e0422e4222d8b5619 | https://github.com/Moymix/BERT-pytorch/tree/f0b9c3ae53e05c00adcc761e0422e4222d8b5619 | import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def forward(self, query, key, value, mask=None, dropout=None):
"""Compute 'Scaled Dot Product Attention'"""
d_k = query.size(-1)
scores = torch.matmul(query, key.transpose(-2, -1)) / math.sqrt(d_k)
if mask is not None:
scores = scores.masked_fill(mask == 0, -1000000000.0)
p_attn = F.softmax(scores, dim=-1)
if dropout is not None:
p_attn = dropout(p_attn)
return torch.matmul(p_attn, value), p_attn
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return []
|
VDSR_F64B6 | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/jj/cjjr5irxfotmh4xreee5lxonmkib4qscmbuzwp2xvaafsncwvu3q.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# y => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1048576],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1048576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + (x0), tmp2, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9 = args
args.clear()
assert_size_stride(primals_1, (64, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (1, 64, 3, 3), (576, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [y_1], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf3, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [y_2], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf5, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [y_3], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf7, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [y_4], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf9, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_5], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf11, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_6], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [y_6], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf13, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [y_7], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf14, primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, buf1, buf3, buf5, buf7, buf9, buf11, buf13, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((64, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 1, 64, 64), (4096, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((1, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class VDSR_F64B6(nn.Module):
def __init__(self, model=False, fixed=False):
super(VDSR_F64B6, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.conv8(y)
return y
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + x0, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (64, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (1, 64, 3, 3), (576, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(1048576)](buf1, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_relu_0[grid(1048576)](buf3, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_relu_0[grid(1048576)](buf5, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_relu_0[grid(1048576)](buf7, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_relu_0[grid(1048576)](buf9, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_relu_0[grid(1048576)](buf11, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_relu_0[grid(1048576)](buf13, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf14, primals_1, primals_2, primals_3, primals_4, primals_5,
primals_6, primals_7, primals_8, primals_9, buf1, buf3, buf5, buf7,
buf9, buf11, buf13)
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class VDSR_F64B6New(nn.Module):
def __init__(self, model=False, fixed=False):
super(VDSR_F64B6New, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_3 = self.conv2.weight
primals_4 = self.conv3.weight
primals_5 = self.conv4.weight
primals_6 = self.conv5.weight
primals_7 = self.conv6.weight
primals_8 = self.conv7.weight
primals_9 = self.conv8.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
| MingSun-Tse/pytorch-vdsr | VDSR_F64B6 | false | 5,600 | [
"MIT"
] | 1 | 597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | https://github.com/MingSun-Tse/pytorch-vdsr/tree/597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class Model(nn.Module):
def __init__(self, model=False, fixed=False):
super().__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.conv8(y)
return y
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.conv8(y)
out8 = y
return out1, out2, out3, out4, out5, out6, out7, out8
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return []
|
SmallVDSR_16x | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/xt/cxtkxbuil7mp4w2vi2sujbaxa6dkeig5rsoswqo7mepdr66ra5cf.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# y => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[262144],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 262144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + (x0), tmp2, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21 = args
args.clear()
assert_size_stride(primals_1, (16, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_4, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_5, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_7, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_8, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_9, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_10, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_11, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_12, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_13, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_14, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_15, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_16, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_17, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_18, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_19, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_20, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_21, (1, 16, 3, 3), (144, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [y_1], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf3, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [y_2], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf5, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [y_3], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf7, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [y_4], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf9, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_5], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf11, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_6], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [y_6], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf13, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_7], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf15 = buf14; del buf14 # reuse
# Topologically Sorted Source Nodes: [y_7], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf15, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_8], Original ATen: [aten.convolution]
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [y_8], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf17, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_9], Original ATen: [aten.convolution]
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf19, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_10], Original ATen: [aten.convolution]
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf21 = buf20; del buf20 # reuse
# Topologically Sorted Source Nodes: [y_10], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf21, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_11], Original ATen: [aten.convolution]
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf23 = buf22; del buf22 # reuse
# Topologically Sorted Source Nodes: [y_11], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf23, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_12], Original ATen: [aten.convolution]
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf25 = buf24; del buf24 # reuse
# Topologically Sorted Source Nodes: [y_12], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf25, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_13], Original ATen: [aten.convolution]
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf27 = buf26; del buf26 # reuse
# Topologically Sorted Source Nodes: [y_13], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf27, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_14], Original ATen: [aten.convolution]
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf29 = buf28; del buf28 # reuse
# Topologically Sorted Source Nodes: [y_14], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf29, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_15], Original ATen: [aten.convolution]
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf31 = buf30; del buf30 # reuse
# Topologically Sorted Source Nodes: [y_15], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf31, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_16], Original ATen: [aten.convolution]
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf33 = buf32; del buf32 # reuse
# Topologically Sorted Source Nodes: [y_16], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf33, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_17], Original ATen: [aten.convolution]
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf35 = buf34; del buf34 # reuse
# Topologically Sorted Source Nodes: [y_17], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf35, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_18], Original ATen: [aten.convolution]
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf37 = buf36; del buf36 # reuse
# Topologically Sorted Source Nodes: [y_18], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf37, 262144, grid=grid(262144), stream=stream0)
# Topologically Sorted Source Nodes: [y_19], Original ATen: [aten.convolution]
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21, buf1, buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21, buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((16, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 1, 64, 64), (4096, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_18 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_19 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_20 = rand_strided((16, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_21 = rand_strided((1, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class SmallVDSR_16x(nn.Module):
def __init__(self, model=False, fixed=False):
super(SmallVDSR_16x, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 16, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(16, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.relu(self.conv8(y))
y = self.relu(self.conv9(y))
y = self.relu(self.conv10(y))
y = self.relu(self.conv11(y))
y = self.relu(self.conv12(y))
y = self.relu(self.conv13(y))
y = self.relu(self.conv14(y))
y = self.relu(self.conv15(y))
y = self.relu(self.conv16(y))
y = self.relu(self.conv17(y))
y = self.relu(self.conv18(y))
y = self.relu(self.conv19(y))
y = self.conv20(y)
return y
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + x0, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17,
primals_18, primals_19, primals_20, primals_21) = args
args.clear()
assert_size_stride(primals_1, (16, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_4, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_5, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_6, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_7, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_8, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_9, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_10, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_11, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_12, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_13, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_14, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_15, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_16, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_17, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_18, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_19, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_20, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_21, (1, 16, 3, 3), (144, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(262144)](buf1, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_relu_0[grid(262144)](buf3, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_relu_0[grid(262144)](buf5, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_relu_0[grid(262144)](buf7, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_relu_0[grid(262144)](buf9, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_relu_0[grid(262144)](buf11, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_relu_0[grid(262144)](buf13, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf15 = buf14
del buf14
triton_poi_fused_relu_0[grid(262144)](buf15, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf17 = buf16
del buf16
triton_poi_fused_relu_0[grid(262144)](buf17, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf19 = buf18
del buf18
triton_poi_fused_relu_0[grid(262144)](buf19, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf21 = buf20
del buf20
triton_poi_fused_relu_0[grid(262144)](buf21, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf23 = buf22
del buf22
triton_poi_fused_relu_0[grid(262144)](buf23, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf25 = buf24
del buf24
triton_poi_fused_relu_0[grid(262144)](buf25, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf27 = buf26
del buf26
triton_poi_fused_relu_0[grid(262144)](buf27, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf29 = buf28
del buf28
triton_poi_fused_relu_0[grid(262144)](buf29, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf31 = buf30
del buf30
triton_poi_fused_relu_0[grid(262144)](buf31, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf33 = buf32
del buf32
triton_poi_fused_relu_0[grid(262144)](buf33, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf35 = buf34
del buf34
triton_poi_fused_relu_0[grid(262144)](buf35, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf37 = buf36
del buf36
triton_poi_fused_relu_0[grid(262144)](buf37, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5,
primals_6, primals_7, primals_8, primals_9, primals_10, primals_11,
primals_12, primals_13, primals_14, primals_15, primals_16,
primals_17, primals_18, primals_19, primals_20, primals_21, buf1,
buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21,
buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37)
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class SmallVDSR_16xNew(nn.Module):
def __init__(self, model=False, fixed=False):
super(SmallVDSR_16xNew, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 16, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(16, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_3 = self.conv2.weight
primals_4 = self.conv3.weight
primals_5 = self.conv4.weight
primals_6 = self.conv5.weight
primals_7 = self.conv6.weight
primals_8 = self.conv7.weight
primals_9 = self.conv8.weight
primals_10 = self.conv9.weight
primals_11 = self.conv10.weight
primals_12 = self.conv11.weight
primals_13 = self.conv12.weight
primals_14 = self.conv13.weight
primals_15 = self.conv14.weight
primals_16 = self.conv15.weight
primals_17 = self.conv16.weight
primals_18 = self.conv17.weight
primals_19 = self.conv18.weight
primals_20 = self.conv19.weight
primals_21 = self.conv20.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21])
return output[0]
| MingSun-Tse/pytorch-vdsr | SmallVDSR_16x | false | 5,601 | [
"MIT"
] | 1 | 597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | https://github.com/MingSun-Tse/pytorch-vdsr/tree/597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class Model(nn.Module):
def __init__(self, model=False, fixed=False):
super().__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 16, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(16, 16, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(16, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
# ... truncated (>4000 chars) for memory efficiency |
Squash | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/aq/caqnx7bnsdfn2e6tzirgxkuwohtkneweobuzqcu3252knhg7lubg.py
# Topologically Sorted Source Nodes: [pow_1, squared_norm, add, scale, mul, sqrt, add_1, truediv_1], Original ATen: [aten.pow, aten.sum, aten.add, aten.div, aten.mul, aten.sqrt]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# mul => mul
# pow_1 => pow_1
# scale => div
# sqrt => sqrt
# squared_norm => sum_1
# truediv_1 => div_1
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%arg0_1, 2), kwargs = {})
# %sum_1 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [-1], True), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_1, 1), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sum_1, %add), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %arg0_1), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%sum_1,), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sqrt, 1e-08), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul, %add_1), kwargs = {})
triton_poi_fused_add_div_mul_pow_sqrt_sum_0 = async_compile.triton('triton_poi_fused_add_div_mul_pow_sqrt_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_mul_pow_sqrt_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_div_mul_pow_sqrt_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp10 / tmp12
tmp15 = tmp13 * tmp14
tmp16 = libdevice.sqrt(tmp10)
tmp17 = 1e-08
tmp18 = tmp16 + tmp17
tmp19 = tmp15 / tmp18
tl.store(out_ptr0 + (x2), tmp19, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_1, squared_norm, add, scale, mul, sqrt, add_1, truediv_1], Original ATen: [aten.pow, aten.sum, aten.add, aten.div, aten.mul, aten.sqrt]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_mul_pow_sqrt_sum_0.run(arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
return (buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Squash(nn.Module):
def forward(self, x, dim=-1):
squared_norm = (x ** 2).sum(dim=dim, keepdim=True)
scale = squared_norm / (1 + squared_norm)
return scale * x / (squared_norm.sqrt() + 1e-08)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_div_mul_pow_sqrt_sum_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp10 / tmp12
tmp15 = tmp13 * tmp14
tmp16 = libdevice.sqrt(tmp10)
tmp17 = 1e-08
tmp18 = tmp16 + tmp17
tmp19 = tmp15 / tmp18
tl.store(out_ptr0 + x2, tmp19, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_mul_pow_sqrt_sum_0[grid(256)](arg0_1, buf0,
256, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class SquashNew(nn.Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MobtgZhang/MWMLNet | Squash | false | 5,602 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import torch
import torch.nn as nn
class Model(nn.Module):
def forward(self, x, dim=-1):
squared_norm = (x ** 2).sum(dim=dim, keepdim=True)
scale = squared_norm / (1 + squared_norm)
return scale * x / (squared_norm.sqrt() + 1e-08)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
ResConnectionLayer | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/df/cdfcie57v6pcdd6oeaz4mvlgksxgyuxzmlv5bklwemyulqhtcxta.py
# Topologically Sorted Source Nodes: [mean, std, sub, mul, add, truediv, add_1], Original ATen: [aten.mean, aten.std, aten.sub, aten.mul, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# mean => mean
# mul => mul
# std => sqrt, var
# sub => sub
# truediv => div
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%primals_1, [-1], True), kwargs = {})
# %var : [num_users=1] = call_function[target=torch.ops.aten.var.correction](args = (%primals_1, [-1]), kwargs = {correction: 1.0, keepdim: True})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%var,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %mean), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_2, %sub), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sqrt, 1e-06), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul, %add), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, %primals_3), kwargs = {})
triton_poi_fused_add_div_mean_mul_std_sub_0 = async_compile.triton('triton_poi_fused_add_div_mean_mul_std_sub_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_mean_mul_std_sub_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 7, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tl.load(in_ptr1 + (4*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp8 = tmp6 + tmp7
tmp9 = 4.0
tmp10 = tmp8 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp0 * tmp11
tmp13 = tmp2 - tmp10
tmp14 = tmp13 * tmp13
tmp15 = tmp3 - tmp10
tmp16 = tmp15 * tmp15
tmp17 = tmp14 + tmp16
tmp18 = tmp5 - tmp10
tmp19 = tmp18 * tmp18
tmp20 = tmp17 + tmp19
tmp21 = tmp7 - tmp10
tmp22 = tmp21 * tmp21
tmp23 = tmp20 + tmp22
tmp24 = 3.0
tmp25 = tmp23 / tmp24
tmp26 = libdevice.sqrt(tmp25)
tmp27 = 1e-06
tmp28 = tmp26 + tmp27
tmp29 = tmp12 / tmp28
tmp31 = tmp29 + tmp30
tl.store(out_ptr0 + (x2), tmp31, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ej/cejqrrx6xg2ykx3zpkubs7k47thei4h2ldnfxjlnconk37jvowpc.py
# Topologically Sorted Source Nodes: [mul_1, pow_1, mul_2, add_2, mul_3, tanh, add_3, hid], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
# Source node to ATen node mapping:
# add_2 => add_2
# add_3 => add_3
# hid => mul_4
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# pow_1 => pow_1
# tanh => tanh
# Graph fragment:
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, 0.5), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_1, 3), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_1, 0.044715), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %mul_2), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_2, 0.7978845608028654), kwargs = {})
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%mul_3,), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%tanh, 1), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_1, %add_3), kwargs = {})
triton_poi_fused_add_mul_pow_tanh_1 = async_compile.triton('triton_poi_fused_add_mul_pow_tanh_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_pow_tanh_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = tmp0 * tmp0
tmp4 = tmp3 * tmp0
tmp5 = 0.044715
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tmp8 = 0.7978845608028654
tmp9 = tmp7 * tmp8
tmp10 = libdevice.tanh(tmp9)
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp2 * tmp12
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/b3/cb3g6fwupaz5a5j23ckgaqji56bsmt4ixc37lwt344u76m75fqhf.py
# Topologically Sorted Source Nodes: [add_4], Original ATen: [aten.add]
# Source node to ATen node mapping:
# add_4 => add_4
# Graph fragment:
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_1, %view_3), kwargs = {})
triton_poi_fused_add_2 = async_compile.triton('triton_poi_fused_add_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_out_ptr0 + (x2), xmask)
tmp2 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mean, std, sub, mul, add, truediv, add_1], Original ATen: [aten.mean, aten.std, aten.sub, aten.mul, aten.add, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_mean_mul_std_sub_0.run(primals_2, primals_1, primals_3, buf0, 256, grid=grid(256), stream=stream0)
del primals_2
del primals_3
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf0, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf1)
del primals_5
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul_1, pow_1, mul_2, add_2, mul_3, tanh, add_3, hid], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
triton_poi_fused_add_mul_pow_tanh_1.run(buf1, buf2, 256, grid=grid(256), stream=stream0)
buf3 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf3)
buf4 = reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [add_4], Original ATen: [aten.add]
triton_poi_fused_add_2.run(buf4, primals_1, primals_7, 256, grid=grid(256), stream=stream0)
del primals_7
return (buf4, primals_1, reinterpret_tensor(buf0, (64, 4), (4, 1), 0), buf1, reinterpret_tensor(buf2, (64, 4), (4, 1), 0), primals_6, primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
class LayerNorm(nn.Module):
"""Construct a layernorm module (See citation for details)."""
def __init__(self, features, eps=1e-06):
super(LayerNorm, self).__init__()
self.a_2 = nn.Parameter(torch.ones(features))
self.b_2 = nn.Parameter(torch.zeros(features))
self.eps = eps
def forward(self, x):
mean = x.mean(-1, keepdim=True)
std = x.std(-1, keepdim=True)
return self.a_2 * (x - mean) / (std + self.eps) + self.b_2
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class FeedForwardNetwork(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, inputs):
hid = self.gleu(self.lin1(inputs))
return self.lin2(self.dropout(hid))
class ResConnectionLayer(nn.Module):
def __init__(self, in_dim, dropout):
super(ResConnectionLayer, self).__init__()
self.norm = LayerNorm(in_dim)
self.dropout = nn.Dropout(dropout)
self.ffn = FeedForwardNetwork(in_dim, in_dim)
def forward(self, x):
"""Apply residual connection to any sublayer with the same size."""
return x + self.dropout(self.ffn(self.norm(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_dim': 4, 'dropout': 0.5}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_div_mean_mul_std_sub_0(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp8 = tmp6 + tmp7
tmp9 = 4.0
tmp10 = tmp8 / tmp9
tmp11 = tmp1 - tmp10
tmp12 = tmp0 * tmp11
tmp13 = tmp2 - tmp10
tmp14 = tmp13 * tmp13
tmp15 = tmp3 - tmp10
tmp16 = tmp15 * tmp15
tmp17 = tmp14 + tmp16
tmp18 = tmp5 - tmp10
tmp19 = tmp18 * tmp18
tmp20 = tmp17 + tmp19
tmp21 = tmp7 - tmp10
tmp22 = tmp21 * tmp21
tmp23 = tmp20 + tmp22
tmp24 = 3.0
tmp25 = tmp23 / tmp24
tmp26 = libdevice.sqrt(tmp25)
tmp27 = 1e-06
tmp28 = tmp26 + tmp27
tmp29 = tmp12 / tmp28
tmp31 = tmp29 + tmp30
tl.store(out_ptr0 + x2, tmp31, xmask)
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_1(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = tmp0 * tmp0
tmp4 = tmp3 * tmp0
tmp5 = 0.044715
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tmp8 = 0.7978845608028654
tmp9 = tmp7 * tmp8
tmp10 = libdevice.tanh(tmp9)
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp2 * tmp12
tl.store(out_ptr0 + x0, tmp13, xmask)
@triton.jit
def triton_poi_fused_add_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_out_ptr0 + x2, xmask)
tmp2 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_mean_mul_std_sub_0[grid(256)](primals_2,
primals_1, primals_3, buf0, 256, XBLOCK=128, num_warps=4,
num_stages=1)
del primals_2
del primals_3
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf0, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf1)
del primals_5
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_mul_pow_tanh_1[grid(256)](buf1, buf2, 256,
XBLOCK=256, num_warps=4, num_stages=1)
buf3 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf3)
buf4 = reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf3
triton_poi_fused_add_2[grid(256)](buf4, primals_1, primals_7, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_7
return buf4, primals_1, reinterpret_tensor(buf0, (64, 4), (4, 1), 0
), buf1, reinterpret_tensor(buf2, (64, 4), (4, 1), 0
), primals_6, primals_4
class LayerNorm(nn.Module):
"""Construct a layernorm module (See citation for details)."""
def __init__(self, features, eps=1e-06):
super(LayerNorm, self).__init__()
self.a_2 = nn.Parameter(torch.ones(features))
self.b_2 = nn.Parameter(torch.zeros(features))
self.eps = eps
def forward(self, x):
mean = x.mean(-1, keepdim=True)
std = x.std(-1, keepdim=True)
return self.a_2 * (x - mean) / (std + self.eps) + self.b_2
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class FeedForwardNetwork(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, inputs):
hid = self.gleu(self.lin1(inputs))
return self.lin2(self.dropout(hid))
class ResConnectionLayerNew(nn.Module):
def __init__(self, in_dim, dropout):
super(ResConnectionLayerNew, self).__init__()
self.norm = LayerNorm(in_dim)
self.dropout = nn.Dropout(dropout)
self.ffn = FeedForwardNetwork(in_dim, in_dim)
def forward(self, input_0):
primals_2 = self.norm.a_2
primals_3 = self.norm.b_2
primals_4 = self.ffn.lin1.weight
primals_5 = self.ffn.lin1.bias
primals_6 = self.ffn.lin2.weight
primals_7 = self.ffn.lin2.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| MobtgZhang/MWMLNet | ResConnectionLayer | false | 5,603 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import math
import torch
import torch.nn as nn
class LayerNorm(nn.Module):
"""Construct a layernorm module (See citation for details)."""
def __init__(self, features, eps=1e-06):
super().__init__()
self.a_2 = nn.Parameter(torch.ones(features))
self.b_2 = nn.Parameter(torch.zeros(features))
self.eps = eps
def forward(self, x):
mean = x.mean(-1, keepdim=True)
std = x.std(-1, keepdim=True)
return self.a_2 * (x - mean) / (std + self.eps) + self.b_2
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class FeedForwardNetwork(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, inputs):
hid = self.gleu(self.lin1(inputs))
return self.lin2(self.dropout(hid))
class Model(nn.Module):
def __init__(self, in_dim, dropout):
super().__init__()
self.norm = LayerNorm(in_dim)
self.dropout = nn.Dropout(dropout)
self.ffn = FeedForwardNetwork(in_dim, in_dim)
def forward(self, x):
"""Apply residual connection to any sublayer with the same size."""
return x + self.dropout(self.ffn(self.norm(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 0.5]
|
AE | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/lp/clp5td7lbqtje3pt7v6xbcp766swgazqemomz2nzsxtdtmjesxht.py
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# x => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1], [9, 9], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_convolution_relu_0 = async_compile.triton('triton_poi_fused_convolution_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[262144],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 262144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 4096) % 16
tmp0 = tl.load(in_out_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/fx/cfxuycpiorfjquslszyvaiul2h6o52zfes7nl5luwduiyxejuafg.py
# Topologically Sorted Source Nodes: [max_pool2d], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# max_pool2d => _low_memory_max_pool2d_offsets_to_indices, _low_memory_max_pool2d_with_offsets, getitem
# Graph fragment:
# %_low_memory_max_pool2d_with_offsets : [num_users=2] = call_function[target=torch.ops.prims._low_memory_max_pool2d_with_offsets.default](args = (%relu, [2, 2], [2, 2], [0, 0], [1, 1], False), kwargs = {})
# %getitem : [num_users=2] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets, 0), kwargs = {})
# %_low_memory_max_pool2d_offsets_to_indices : [num_users=2] = call_function[target=torch.ops.prims._low_memory_max_pool2d_offsets_to_indices.default](args = (%getitem_1, 2, 64, [2, 2], [0, 0]), kwargs = {})
triton_poi_fused_max_pool2d_with_indices_1 = async_compile.triton('triton_poi_fused_max_pool2d_with_indices_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i64', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_max_pool2d_with_indices_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 65536
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 32
x1 = (xindex // 32)
x4 = xindex
x2 = (xindex // 32) % 32
tmp0 = tl.load(in_ptr0 + ((2*x0) + (128*x1)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (2*x0) + (128*x1)), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (64 + (2*x0) + (128*x1)), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (65 + (2*x0) + (128*x1)), None, eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = tmp1 > tmp0
tmp8 = tl.full([1], 1, tl.int8)
tmp9 = tl.full([1], 0, tl.int8)
tmp10 = tl.where(tmp7, tmp8, tmp9)
tmp11 = tmp3 > tmp2
tmp12 = tl.full([1], 2, tl.int8)
tmp13 = tl.where(tmp11, tmp12, tmp10)
tmp14 = tmp5 > tmp4
tmp15 = tl.full([1], 3, tl.int8)
tmp16 = tl.where(tmp14, tmp15, tmp13)
tmp17 = tl.full([1], 2, tl.int32)
tmp18 = tl.where((tmp16 < 0) != (tmp17 < 0), tl.where(tmp16 % tmp17 != 0, tmp16 // tmp17 - 1, tmp16 // tmp17), tmp16 // tmp17)
tmp19 = tmp18 * tmp17
tmp20 = tmp16 - tmp19
tmp21 = 2*x2
tmp22 = tmp21 + tmp18
tmp23 = 2*x0
tmp24 = tmp23 + tmp20
tmp25 = tl.full([1], 64, tl.int64)
tmp26 = tmp22 * tmp25
tmp27 = tmp26 + tmp24
tl.store(out_ptr0 + (x4), tmp6, None)
tl.store(out_ptr1 + (x4), tmp27, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/n2/cn23y7jhz2xygynfpumprse4torkd7nlohv4kacxis56jwb54k5e.py
# Topologically Sorted Source Nodes: [conv2d_1, x_2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_1 => convolution_1
# x_2 => relu_1
# Graph fragment:
# %convolution_1 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%getitem, %primals_4, %primals_5, [1, 1], [7, 7], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_1,), kwargs = {})
triton_poi_fused_convolution_relu_2 = async_compile.triton('triton_poi_fused_convolution_relu_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16384],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16384
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 1024) % 4
tmp0 = tl.load(in_out_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/b5/cb53yctlr3gz2rfsuo4fslnritazzf5vgsiae5sejhqagushb4hc.py
# Topologically Sorted Source Nodes: [max_pool2d_1], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# max_pool2d_1 => _low_memory_max_pool2d_offsets_to_indices_1, _low_memory_max_pool2d_with_offsets_1
# Graph fragment:
# %_low_memory_max_pool2d_with_offsets_1 : [num_users=2] = call_function[target=torch.ops.prims._low_memory_max_pool2d_with_offsets.default](args = (%relu_1, [2, 2], [2, 2], [0, 0], [1, 1], False), kwargs = {})
# %_low_memory_max_pool2d_offsets_to_indices_1 : [num_users=2] = call_function[target=torch.ops.prims._low_memory_max_pool2d_offsets_to_indices.default](args = (%getitem_3, 2, 32, [2, 2], [0, 0]), kwargs = {})
triton_poi_fused_max_pool2d_with_indices_3 = async_compile.triton('triton_poi_fused_max_pool2d_with_indices_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_max_pool2d_with_indices_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_3(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 16
x3 = (xindex // 16)
x1 = (xindex // 16) % 16
x4 = xindex
tmp0 = tl.load(in_ptr0 + ((2*x0) + (64*x3)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (2*x0) + (64*x3)), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (32 + (2*x0) + (64*x3)), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (33 + (2*x0) + (64*x3)), None, eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tmp8 = tmp7 > tmp6
tmp9 = tl.full([1], 2, tl.int8)
tmp10 = tl.where(tmp8, tmp9, tmp5)
tmp11 = triton_helpers.maximum(tmp7, tmp6)
tmp13 = tmp12 > tmp11
tmp14 = tl.full([1], 3, tl.int8)
tmp15 = tl.where(tmp13, tmp14, tmp10)
tmp16 = triton_helpers.maximum(tmp12, tmp11)
tmp17 = tl.full([1], 2, tl.int32)
tmp18 = tl.where((tmp15 < 0) != (tmp17 < 0), tl.where(tmp15 % tmp17 != 0, tmp15 // tmp17 - 1, tmp15 // tmp17), tmp15 // tmp17)
tmp19 = tmp18 * tmp17
tmp20 = tmp15 - tmp19
tmp21 = 2*x1
tmp22 = tmp21 + tmp18
tmp23 = 2*x0
tmp24 = tmp23 + tmp20
tmp25 = tl.full([1], 32, tl.int64)
tmp26 = tmp22 * tmp25
tmp27 = tmp26 + tmp24
tl.store(out_ptr0 + (x4), tmp27, None)
tl.store(out_ptr1 + (x4), tmp16, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/rb/crbjzrv5jiob43hnwmv5pasmjf4rrkawp7xunizdmnue24skyybh.py
# Topologically Sorted Source Nodes: [conv_transpose2d, x_5], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv_transpose2d => convolution_2
# x_5 => relu_2
# Graph fragment:
# %convolution_2 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%max_unpool2d, %primals_6, %primals_7, [1, 1], [7, 7], [1, 1], True, [0, 0], 1), kwargs = {})
# %relu_2 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_2,), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_2, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_4 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[65536],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_4', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_4(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 65536
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 1024) % 16
tmp0 = tl.load(in_out_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr0 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x3), tmp4, None)
tl.store(out_ptr0 + (x3), tmp6, None)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/pg/cpgg4oz3hvqpwyi5s6ig3dfrsdbicyjavvtp2wvyfnoivug2bvhp.py
# Topologically Sorted Source Nodes: [conv_transpose2d_1, x_7], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv_transpose2d_1 => convolution_3
# x_7 => relu_3
# Graph fragment:
# %convolution_3 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%max_unpool2d_1, %primals_8, %primals_9, [1, 1], [9, 9], [1, 1], True, [0, 0], 1), kwargs = {})
# %relu_3 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_3,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_3, 0), kwargs = {})
triton_poi_fused_convolution_relu_threshold_backward_5 = async_compile.triton('triton_poi_fused_convolution_relu_threshold_backward_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16384],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_relu_threshold_backward_5', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_5(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16384
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), None)
tmp1 = tl.load(in_ptr0 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 0.0
tmp7 = tmp5 <= tmp6
tl.store(in_out_ptr0 + (x0), tmp5, None)
tl.store(out_ptr0 + (x0), tmp7, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9 = args
args.clear()
assert_size_stride(primals_1, (16, 1, 19, 19), (361, 361, 19, 1))
assert_size_stride(primals_2, (16, ), (1, ))
assert_size_stride(primals_3, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_4, (4, 16, 15, 15), (3600, 225, 15, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 16, 15, 15), (3600, 225, 15, 1))
assert_size_stride(primals_7, (16, ), (1, ))
assert_size_stride(primals_8, (16, 1, 19, 19), (361, 361, 19, 1))
assert_size_stride(primals_9, (1, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1), padding=(9, 9), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d, x], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 262144, grid=grid(262144), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1), torch.float32)
buf3 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1), torch.int64)
# Topologically Sorted Source Nodes: [max_pool2d], Original ATen: [aten.max_pool2d_with_indices]
triton_poi_fused_max_pool2d_with_indices_1.run(buf1, buf2, buf3, 65536, grid=grid(65536), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf2, primals_4, stride=(1, 1), padding=(7, 7), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 4, 32, 32), (4096, 1024, 32, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, x_2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_2.run(buf5, primals_5, 16384, grid=grid(16384), stream=stream0)
del primals_5
buf6 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch.int64)
buf7 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch.float32)
# Topologically Sorted Source Nodes: [max_pool2d_1], Original ATen: [aten.max_pool2d_with_indices]
triton_poi_fused_max_pool2d_with_indices_3.run(buf5, buf6, buf7, 4096, grid=grid(4096), stream=stream0)
# Topologically Sorted Source Nodes: [max_pool2d_1, x_4], Original ATen: [aten.max_pool2d_with_indices, aten.max_unpool2d]
buf8 = torch.ops.aten.max_unpool2d.default(buf7, buf6, [32, 32])
del buf7
buf9 = buf8
del buf8
# Topologically Sorted Source Nodes: [conv_transpose2d], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_6, stride=(1, 1), padding=(7, 7), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 16, 32, 32), (16384, 1024, 32, 1))
buf11 = buf10; del buf10 # reuse
buf17 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [conv_transpose2d, x_5], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_4.run(buf11, primals_7, buf17, 65536, grid=grid(65536), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.max_unpool2d]
buf12 = torch.ops.aten.max_unpool2d.default(buf11, buf3, [64, 64])
del buf11
buf13 = buf12
del buf12
# Topologically Sorted Source Nodes: [conv_transpose2d_1], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_8, stride=(1, 1), padding=(9, 9), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 1, 64, 64), (4096, 4096, 64, 1))
buf15 = buf14; del buf14 # reuse
buf16 = empty_strided_cuda((4, 1, 64, 64), (4096, 4096, 64, 1), torch.bool)
# Topologically Sorted Source Nodes: [conv_transpose2d_1, x_7], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_5.run(buf15, primals_9, buf16, 16384, grid=grid(16384), stream=stream0)
del primals_9
return (buf15, primals_1, primals_3, primals_4, primals_6, primals_8, buf1, buf2, buf3, buf5, buf6, buf9, buf13, buf16, buf17, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((16, 1, 19, 19), (361, 361, 19, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 1, 64, 64), (4096, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 16, 15, 15), (3600, 225, 15, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 16, 15, 15), (3600, 225, 15, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((16, 1, 19, 19), (361, 361, 19, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((1, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
from torch import nn
import torch.utils.data
import torch.utils.data.distributed
class AE(nn.Module):
def __init__(self):
super(AE, self).__init__()
self.conv1 = nn.Conv2d(1, 16, kernel_size=19, padding=9)
self.conv2 = nn.Conv2d(16, 4, kernel_size=15, padding=7)
self.pool = nn.MaxPool2d(kernel_size=2, stride=2, return_indices=True)
self.t_conv1 = nn.ConvTranspose2d(4, 16, kernel_size=15, padding=7)
self.t_conv2 = nn.ConvTranspose2d(16, 1, kernel_size=19, padding=9)
self.unpool = nn.MaxUnpool2d(kernel_size=2, stride=2)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
x = self.relu(self.conv1(x))
x, pool1 = self.pool(x)
x = self.relu(self.conv2(x))
x, pool2 = self.pool(x)
x = self.unpool(x, pool2)
x = self.relu(self.t_conv1(x))
x = self.unpool(x, pool1)
x = self.relu(self.t_conv2(x))
return x
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
import torch.utils.data
import torch.utils.data.distributed
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 16
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 32
x1 = xindex // 32
x4 = xindex
x2 = xindex // 32 % 32
tmp0 = tl.load(in_ptr0 + (2 * x0 + 128 * x1), None, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 128 * x1), None, eviction_policy
='evict_last')
tmp3 = tl.load(in_ptr0 + (64 + 2 * x0 + 128 * x1), None,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (65 + 2 * x0 + 128 * x1), None,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = tmp1 > tmp0
tmp8 = tl.full([1], 1, tl.int8)
tmp9 = tl.full([1], 0, tl.int8)
tmp10 = tl.where(tmp7, tmp8, tmp9)
tmp11 = tmp3 > tmp2
tmp12 = tl.full([1], 2, tl.int8)
tmp13 = tl.where(tmp11, tmp12, tmp10)
tmp14 = tmp5 > tmp4
tmp15 = tl.full([1], 3, tl.int8)
tmp16 = tl.where(tmp14, tmp15, tmp13)
tmp17 = tl.full([1], 2, tl.int32)
tmp18 = tl.where((tmp16 < 0) != (tmp17 < 0), tl.where(tmp16 % tmp17 !=
0, tmp16 // tmp17 - 1, tmp16 // tmp17), tmp16 // tmp17)
tmp19 = tmp18 * tmp17
tmp20 = tmp16 - tmp19
tmp21 = 2 * x2
tmp22 = tmp21 + tmp18
tmp23 = 2 * x0
tmp24 = tmp23 + tmp20
tmp25 = tl.full([1], 64, tl.int64)
tmp26 = tmp22 * tmp25
tmp27 = tmp26 + tmp24
tl.store(out_ptr0 + x4, tmp6, None)
tl.store(out_ptr1 + x4, tmp27, None)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 1024 % 4
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_3(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 16
x3 = xindex // 16
x1 = xindex // 16 % 16
x4 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 64 * x3), None, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 64 * x3), None, eviction_policy=
'evict_last')
tmp7 = tl.load(in_ptr0 + (32 + 2 * x0 + 64 * x3), None, eviction_policy
='evict_last')
tmp12 = tl.load(in_ptr0 + (33 + 2 * x0 + 64 * x3), None,
eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tmp8 = tmp7 > tmp6
tmp9 = tl.full([1], 2, tl.int8)
tmp10 = tl.where(tmp8, tmp9, tmp5)
tmp11 = triton_helpers.maximum(tmp7, tmp6)
tmp13 = tmp12 > tmp11
tmp14 = tl.full([1], 3, tl.int8)
tmp15 = tl.where(tmp13, tmp14, tmp10)
tmp16 = triton_helpers.maximum(tmp12, tmp11)
tmp17 = tl.full([1], 2, tl.int32)
tmp18 = tl.where((tmp15 < 0) != (tmp17 < 0), tl.where(tmp15 % tmp17 !=
0, tmp15 // tmp17 - 1, tmp15 // tmp17), tmp15 // tmp17)
tmp19 = tmp18 * tmp17
tmp20 = tmp15 - tmp19
tmp21 = 2 * x1
tmp22 = tmp21 + tmp18
tmp23 = 2 * x0
tmp24 = tmp23 + tmp20
tmp25 = tl.full([1], 32, tl.int64)
tmp26 = tmp22 * tmp25
tmp27 = tmp26 + tmp24
tl.store(out_ptr0 + x4, tmp27, None)
tl.store(out_ptr1 + x4, tmp16, None)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_4(in_out_ptr0,
in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 1024 % 16
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x3, tmp4, None)
tl.store(out_ptr0 + x3, tmp6, None)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_5(in_out_ptr0,
in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 0.0
tmp7 = tmp5 <= tmp6
tl.store(in_out_ptr0 + x0, tmp5, None)
tl.store(out_ptr0 + x0, tmp7, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (16, 1, 19, 19), (361, 361, 19, 1))
assert_size_stride(primals_2, (16,), (1,))
assert_size_stride(primals_3, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_4, (4, 16, 15, 15), (3600, 225, 15, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 16, 15, 15), (3600, 225, 15, 1))
assert_size_stride(primals_7, (16,), (1,))
assert_size_stride(primals_8, (16, 1, 19, 19), (361, 361, 19, 1))
assert_size_stride(primals_9, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(9, 9), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(262144)](buf1, primals_2,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1),
torch.float32)
buf3 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1),
torch.int64)
triton_poi_fused_max_pool2d_with_indices_1[grid(65536)](buf1, buf2,
buf3, 65536, XBLOCK=256, num_warps=4, num_stages=1)
buf4 = extern_kernels.convolution(buf2, primals_4, stride=(1, 1),
padding=(7, 7), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 4, 32, 32), (4096, 1024, 32, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_2[grid(16384)](buf5, primals_5,
16384, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf6 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch
.int64)
buf7 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch
.float32)
triton_poi_fused_max_pool2d_with_indices_3[grid(4096)](buf5, buf6,
buf7, 4096, XBLOCK=128, num_warps=4, num_stages=1)
buf8 = torch.ops.aten.max_unpool2d.default(buf7, buf6, [32, 32])
del buf7
buf9 = buf8
del buf8
buf10 = extern_kernels.convolution(buf9, primals_6, stride=(1, 1),
padding=(7, 7), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 16, 32, 32), (16384, 1024, 32, 1))
buf11 = buf10
del buf10
buf17 = empty_strided_cuda((4, 16, 32, 32), (16384, 1024, 32, 1),
torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_4[grid(65536)](
buf11, primals_7, buf17, 65536, XBLOCK=512, num_warps=4,
num_stages=1)
del primals_7
buf12 = torch.ops.aten.max_unpool2d.default(buf11, buf3, [64, 64])
del buf11
buf13 = buf12
del buf12
buf14 = extern_kernels.convolution(buf13, primals_8, stride=(1, 1),
padding=(9, 9), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 1, 64, 64), (4096, 4096, 64, 1))
buf15 = buf14
del buf14
buf16 = empty_strided_cuda((4, 1, 64, 64), (4096, 4096, 64, 1),
torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_5[grid(16384)](
buf15, primals_9, buf16, 16384, XBLOCK=256, num_warps=4,
num_stages=1)
del primals_9
return (buf15, primals_1, primals_3, primals_4, primals_6, primals_8,
buf1, buf2, buf3, buf5, buf6, buf9, buf13, buf16, buf17)
class AENew(nn.Module):
def __init__(self):
super(AENew, self).__init__()
self.conv1 = nn.Conv2d(1, 16, kernel_size=19, padding=9)
self.conv2 = nn.Conv2d(16, 4, kernel_size=15, padding=7)
self.pool = nn.MaxPool2d(kernel_size=2, stride=2, return_indices=True)
self.t_conv1 = nn.ConvTranspose2d(4, 16, kernel_size=15, padding=7)
self.t_conv2 = nn.ConvTranspose2d(16, 1, kernel_size=19, padding=9)
self.unpool = nn.MaxUnpool2d(kernel_size=2, stride=2)
self.relu = nn.ReLU(inplace=True)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_6 = self.t_conv1.weight
primals_7 = self.t_conv1.bias
primals_8 = self.t_conv2.weight
primals_9 = self.t_conv2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
| Minauras/deepdefresneling | AE | false | 5,604 | [
"BSD-2-Clause"
] | 1 | e17168e9a8d322201998c73da54efbd334b0ffb9 | https://github.com/Minauras/deepdefresneling/tree/e17168e9a8d322201998c73da54efbd334b0ffb9 | import torch
from torch import nn
import torch.utils.data
import torch.utils.data.distributed
class Model(nn.Module):
def __init__(self):
super().__init__()
self.conv1 = nn.Conv2d(1, 16, kernel_size=19, padding=9)
self.conv2 = nn.Conv2d(16, 4, kernel_size=15, padding=7)
self.pool = nn.MaxPool2d(kernel_size=2, stride=2, return_indices=True)
self.t_conv1 = nn.ConvTranspose2d(4, 16, kernel_size=15, padding=7)
self.t_conv2 = nn.ConvTranspose2d(16, 1, kernel_size=19, padding=9)
self.unpool = nn.MaxUnpool2d(kernel_size=2, stride=2)
self.relu = nn.ReLU(inplace=True)
def forward(self, x):
x = self.relu(self.conv1(x))
x, pool1 = self.pool(x)
x = self.relu(self.conv2(x))
x, pool2 = self.pool(x)
x = self.unpool(x, pool2)
x = self.relu(self.t_conv1(x))
x = self.unpool(x, pool1)
x = self.relu(self.t_conv2(x))
return x
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return []
|
Lookahead | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/xf/cxfjorpj7nzakemkplfiue4lxn3wssubujqanweih6rojbuoolb7.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# x_1 => constant_pad_nd
# Graph fragment:
# %constant_pad_nd : [num_users=2] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%permute_1, [0, 3], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16, 8], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
xnumel = 7
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = x1
tmp1 = tl.full([1, 1], 4, tl.int64)
tmp2 = tmp0 < tmp1
tmp3 = tl.load(in_ptr0 + (y0 + (16*x1)), tmp2 & xmask & ymask, eviction_policy='evict_last', other=0.0)
tl.store(out_ptr0 + (x1 + (7*y0)), tmp3, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/r7/cr7evfztwmdzfk7dfomwxcc4g457gqiibpdthzqr2y7vyuw4ptpg.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x_3 => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_3,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_1 = async_compile.triton('triton_poi_fused_clone_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4, 16], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 4
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x1)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x1 + (16*y0)), tmp0, xmask & ymask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 1, 4), (4, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 7), (28, 7, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(primals_1, buf0, 16, 7, grid=grid(16, 7), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=4, bias=None)
assert_size_stride(buf1, (4, 4, 4), (16, 4, 1))
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.clone]
triton_poi_fused_clone_1.run(buf1, buf2, 4, 16, grid=grid(4, 16), stream=stream0)
del buf1
return (buf2, primals_2, buf0, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 1, 4), (4, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.utils.data.distributed
import torch.nn as nn
import torch.nn.functional as F
class Lookahead(nn.Module):
def __init__(self, n_features, context):
super(Lookahead, self).__init__()
assert context > 0
self.context = context
self.n_features = n_features
self.pad = 0, self.context - 1
self.conv = nn.Conv1d(self.n_features, self.n_features, kernel_size
=self.context, stride=1, groups=self.n_features, padding=0,
bias=None)
def forward(self, x):
x = x.transpose(0, 1).transpose(1, 2)
x = F.pad(x, pad=self.pad, value=0)
x = self.conv(x)
x = x.transpose(1, 2).transpose(0, 1).contiguous()
return x
def __repr__(self):
return self.__class__.__name__ + '(' + 'n_features=' + str(self.
n_features) + ', context=' + str(self.context) + ')'
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'n_features': 4, 'context': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.utils.data.distributed
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 7
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = x1
tmp1 = tl.full([1, 1], 4, tl.int64)
tmp2 = tmp0 < tmp1
tmp3 = tl.load(in_ptr0 + (y0 + 16 * x1), tmp2 & xmask & ymask,
eviction_policy='evict_last', other=0.0)
tl.store(out_ptr0 + (x1 + 7 * y0), tmp3, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 4
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x1), xmask & ymask, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (x1 + 16 * y0), tmp0, xmask & ymask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 1, 4), (4, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 7), (28, 7, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(16, 7)](primals_1, buf0, 16,
7, XBLOCK=8, YBLOCK=16, num_warps=4, num_stages=1)
del primals_1
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=4, bias=None)
assert_size_stride(buf1, (4, 4, 4), (16, 4, 1))
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_clone_1[grid(4, 16)](buf1, buf2, 4, 16, XBLOCK=16,
YBLOCK=4, num_warps=1, num_stages=1)
del buf1
return buf2, primals_2, buf0
class LookaheadNew(nn.Module):
def __init__(self, n_features, context):
super(LookaheadNew, self).__init__()
assert context > 0
self.context = context
self.n_features = n_features
self.pad = 0, self.context - 1
self.conv = nn.Conv1d(self.n_features, self.n_features, kernel_size
=self.context, stride=1, groups=self.n_features, padding=0,
bias=None)
def __repr__(self):
return self.__class__.__name__ + '(' + 'n_features=' + str(self.
n_features) + ', context=' + str(self.context) + ')'
def forward(self, input_0):
primals_2 = self.conv.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| MrXJC/deepspeech.pytorch | Lookahead | false | 5,605 | [
"MIT"
] | 1 | 6379c18d3f56cad8896a51d45166ea979423e0bf | https://github.com/MrXJC/deepspeech.pytorch/tree/6379c18d3f56cad8896a51d45166ea979423e0bf | import torch
import torch.utils.data.distributed
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, n_features, context):
super().__init__()
assert context > 0
self.context = context
self.n_features = n_features
self.pad = 0, self.context - 1
self.conv = nn.Conv1d(self.n_features, self.n_features, kernel_size
=self.context, stride=1, groups=self.n_features, padding=0,
bias=None)
def forward(self, x):
x = x.transpose(0, 1).transpose(1, 2)
x = F.pad(x, pad=self.pad, value=0)
x = self.conv(x)
x = x.transpose(1, 2).transpose(0, 1).contiguous()
return x
def __repr__(self):
return self.__class__.__name__ + '(' + 'n_features=' + str(self.
n_features) + ', context=' + str(self.context) + ')'
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
VDSR | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/jj/cjjr5irxfotmh4xreee5lxonmkib4qscmbuzwp2xvaafsncwvu3q.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# y => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1048576],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1048576
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + (x0), tmp2, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21 = args
args.clear()
assert_size_stride(primals_1, (64, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_10, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_11, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_12, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_13, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_14, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_15, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_16, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_17, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_18, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_19, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_20, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_21, (1, 64, 3, 3), (576, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [y_1], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf3, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [y_2], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf5, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [y_3], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf7, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [y_4], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf9, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_5], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf11, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_6], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [y_6], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf13, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_7], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf15 = buf14; del buf14 # reuse
# Topologically Sorted Source Nodes: [y_7], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf15, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_8], Original ATen: [aten.convolution]
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [y_8], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf17, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_9], Original ATen: [aten.convolution]
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf19, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_10], Original ATen: [aten.convolution]
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf21 = buf20; del buf20 # reuse
# Topologically Sorted Source Nodes: [y_10], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf21, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_11], Original ATen: [aten.convolution]
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf23 = buf22; del buf22 # reuse
# Topologically Sorted Source Nodes: [y_11], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf23, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_12], Original ATen: [aten.convolution]
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf25 = buf24; del buf24 # reuse
# Topologically Sorted Source Nodes: [y_12], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf25, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_13], Original ATen: [aten.convolution]
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf27 = buf26; del buf26 # reuse
# Topologically Sorted Source Nodes: [y_13], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf27, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_14], Original ATen: [aten.convolution]
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf29 = buf28; del buf28 # reuse
# Topologically Sorted Source Nodes: [y_14], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf29, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_15], Original ATen: [aten.convolution]
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf31 = buf30; del buf30 # reuse
# Topologically Sorted Source Nodes: [y_15], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf31, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_16], Original ATen: [aten.convolution]
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf33 = buf32; del buf32 # reuse
# Topologically Sorted Source Nodes: [y_16], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf33, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_17], Original ATen: [aten.convolution]
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf35 = buf34; del buf34 # reuse
# Topologically Sorted Source Nodes: [y_17], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf35, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_18], Original ATen: [aten.convolution]
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf37 = buf36; del buf36 # reuse
# Topologically Sorted Source Nodes: [y_18], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf37, 1048576, grid=grid(1048576), stream=stream0)
# Topologically Sorted Source Nodes: [y_19], Original ATen: [aten.convolution]
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21, buf1, buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21, buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((64, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 1, 64, 64), (4096, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_18 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_19 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_20 = rand_strided((64, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_21 = rand_strided((1, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class VDSR(nn.Module):
def __init__(self, model=False, fixed=False):
super(VDSR, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.relu(self.conv8(y))
y = self.relu(self.conv9(y))
y = self.relu(self.conv10(y))
y = self.relu(self.conv11(y))
y = self.relu(self.conv12(y))
y = self.relu(self.conv13(y))
y = self.relu(self.conv14(y))
y = self.relu(self.conv15(y))
y = self.relu(self.conv16(y))
y = self.relu(self.conv17(y))
y = self.relu(self.conv18(y))
y = self.relu(self.conv19(y))
y = self.conv20(y)
return y
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + x0, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17,
primals_18, primals_19, primals_20, primals_21) = args
args.clear()
assert_size_stride(primals_1, (64, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_8, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_10, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_11, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_12, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_13, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_14, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_15, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_16, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_17, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_18, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_19, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_20, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_21, (1, 64, 3, 3), (576, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(1048576)](buf1, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_relu_0[grid(1048576)](buf3, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_relu_0[grid(1048576)](buf5, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_relu_0[grid(1048576)](buf7, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_relu_0[grid(1048576)](buf9, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_relu_0[grid(1048576)](buf11, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_relu_0[grid(1048576)](buf13, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf15 = buf14
del buf14
triton_poi_fused_relu_0[grid(1048576)](buf15, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf17 = buf16
del buf16
triton_poi_fused_relu_0[grid(1048576)](buf17, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf19 = buf18
del buf18
triton_poi_fused_relu_0[grid(1048576)](buf19, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf21 = buf20
del buf20
triton_poi_fused_relu_0[grid(1048576)](buf21, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf23 = buf22
del buf22
triton_poi_fused_relu_0[grid(1048576)](buf23, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf25 = buf24
del buf24
triton_poi_fused_relu_0[grid(1048576)](buf25, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf27 = buf26
del buf26
triton_poi_fused_relu_0[grid(1048576)](buf27, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf29 = buf28
del buf28
triton_poi_fused_relu_0[grid(1048576)](buf29, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf31 = buf30
del buf30
triton_poi_fused_relu_0[grid(1048576)](buf31, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf33 = buf32
del buf32
triton_poi_fused_relu_0[grid(1048576)](buf33, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf35 = buf34
del buf34
triton_poi_fused_relu_0[grid(1048576)](buf35, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 64, 64, 64), (262144, 4096, 64, 1))
buf37 = buf36
del buf36
triton_poi_fused_relu_0[grid(1048576)](buf37, 1048576, XBLOCK=512,
num_warps=8, num_stages=1)
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5,
primals_6, primals_7, primals_8, primals_9, primals_10, primals_11,
primals_12, primals_13, primals_14, primals_15, primals_16,
primals_17, primals_18, primals_19, primals_20, primals_21, buf1,
buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21,
buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37)
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class VDSRNew(nn.Module):
def __init__(self, model=False, fixed=False):
super(VDSRNew, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_3 = self.conv2.weight
primals_4 = self.conv3.weight
primals_5 = self.conv4.weight
primals_6 = self.conv5.weight
primals_7 = self.conv6.weight
primals_8 = self.conv7.weight
primals_9 = self.conv8.weight
primals_10 = self.conv9.weight
primals_11 = self.conv10.weight
primals_12 = self.conv11.weight
primals_13 = self.conv12.weight
primals_14 = self.conv13.weight
primals_15 = self.conv14.weight
primals_16 = self.conv15.weight
primals_17 = self.conv16.weight
primals_18 = self.conv17.weight
primals_19 = self.conv18.weight
primals_20 = self.conv19.weight
primals_21 = self.conv20.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21])
return output[0]
| MingSun-Tse/pytorch-vdsr | VDSR | false | 5,606 | [
"MIT"
] | 1 | 597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | https://github.com/MingSun-Tse/pytorch-vdsr/tree/597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class Model(nn.Module):
def __init__(self, model=False, fixed=False):
super().__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 64, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(64, 64, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(64, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU(inplace=True)
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.relu(self.conv8(y))
y = self.relu(self.conv9(y))
y = self.relu(self.conv10(y))
y = self.relu(self.conv11(y))
y = self.relu(self.conv12(y))
y = self.relu(self.conv13(y))
y = self.relu(self.conv14(y))
y = self.relu(self.conv15(y))
y = self.relu(self.conv16(y))
y = self.relu(self.conv17(y))
y = self.relu(self.conv18(y))
y = self.relu(self.conv19(y))
y = self.conv20(y)
return y
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.
# ... truncated (>4000 chars) for memory efficiency |
DepthwiseSeparableConv | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/53/c53345w2ogarycgzyrcothtqrrb7taubpprhokfthwhic4knqepc.py
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# relu => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%squeeze_1,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_relu_threshold_backward_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_threshold_backward_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 20
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 5)
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 1, 4), (4, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_4, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv1d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_2, (1, 4, 4), (16, 4, 1), 0), primals_1, stride=(1,), padding=(2,), dilation=(1,), transposed=False, output_padding=(0,), groups=4, bias=None)
assert_size_stride(buf0, (1, 4, 5), (20, 5, 1))
# Topologically Sorted Source Nodes: [conv1d_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_3, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf1, (1, 4, 5), (20, 5, 1))
buf2 = reinterpret_tensor(buf1, (4, 5), (5, 1), 0); del buf1 # reuse
buf3 = empty_strided_cuda((4, 5), (5, 1), torch.bool)
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf2, primals_4, buf3, 20, grid=grid(20), stream=stream0)
del primals_4
return (buf2, primals_1, primals_3, reinterpret_tensor(primals_2, (1, 4, 4), (16, 4, 1), 0), buf0, buf3, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 1, 4), (4, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 1), (4, 1, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn.functional as F
import torch.nn as nn
class DepthwiseSeparableConv(nn.Module):
def __init__(self, in_ch, out_ch, k, bias=True):
super().__init__()
self.depthwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
in_ch, kernel_size=k, groups=in_ch, padding=k // 2, bias=False)
self.pointwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
out_ch, kernel_size=1, padding=0, bias=bias)
def forward(self, x):
return F.relu(self.pointwise_conv(self.depthwise_conv(x)))
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'in_ch': 4, 'out_ch': 4, 'k': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 20
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 5
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 1, 4), (4, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_4, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_2, (1,
4, 4), (16, 4, 1), 0), primals_1, stride=(1,), padding=(2,),
dilation=(1,), transposed=False, output_padding=(0,), groups=4,
bias=None)
assert_size_stride(buf0, (1, 4, 5), (20, 5, 1))
buf1 = extern_kernels.convolution(buf0, primals_3, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf1, (1, 4, 5), (20, 5, 1))
buf2 = reinterpret_tensor(buf1, (4, 5), (5, 1), 0)
del buf1
buf3 = empty_strided_cuda((4, 5), (5, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(20)](buf2,
primals_4, buf3, 20, XBLOCK=32, num_warps=1, num_stages=1)
del primals_4
return buf2, primals_1, primals_3, reinterpret_tensor(primals_2, (1, 4,
4), (16, 4, 1), 0), buf0, buf3
class DepthwiseSeparableConvNew(nn.Module):
def __init__(self, in_ch, out_ch, k, bias=True):
super().__init__()
self.depthwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
in_ch, kernel_size=k, groups=in_ch, padding=k // 2, bias=False)
self.pointwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
out_ch, kernel_size=1, padding=0, bias=bias)
def forward(self, input_0):
primals_1 = self.depthwise_conv.weight
primals_3 = self.pointwise_conv.weight
primals_4 = self.pointwise_conv.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| MobtgZhang/MWMLNet | DepthwiseSeparableConv | false | 5,607 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import torch
import torch.nn.functional as F
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_ch, out_ch, k, bias=True):
super().__init__()
self.depthwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
in_ch, kernel_size=k, groups=in_ch, padding=k // 2, bias=False)
self.pointwise_conv = nn.Conv1d(in_channels=in_ch, out_channels=
out_ch, kernel_size=1, padding=0, bias=bias)
def forward(self, x):
return F.relu(self.pointwise_conv(self.depthwise_conv(x)))
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
SFU | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/3l/c3lo77c7wjxasxrhtr6wesb72ods2d2rxnxhbfieun7j2wukm3wn.py
# Topologically Sorted Source Nodes: [r_f], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# r_f => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_1, %primals_2], 2), kwargs = {})
triton_poi_fused_cat_0 = async_compile.triton('triton_poi_fused_cat_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = (xindex // 8)
x2 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((4*x1) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((4*x1) + ((-4) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x2), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4h/c4hvd3vhapir7uardh64cxnc57lcta2i2xqwszrnyxam4nh7uanb.py
# Topologically Sorted Source Nodes: [r, g, mul, sub, mul_1, o], Original ATen: [aten.tanh, aten.sigmoid, aten.mul, aten.rsub, aten.add]
# Source node to ATen node mapping:
# g => sigmoid
# mul => mul
# mul_1 => mul_1
# o => add
# r => tanh
# sub => sub
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%view_1,), kwargs = {})
# %sigmoid : [num_users=2] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_3,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid, %tanh), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %sigmoid), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %primals_1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %mul_1), kwargs = {})
triton_poi_fused_add_mul_rsub_sigmoid_tanh_1 = async_compile.triton('triton_poi_fused_add_mul_rsub_sigmoid_tanh_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_rsub_sigmoid_tanh_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_rsub_sigmoid_tanh_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp2 = tl.load(in_ptr1 + (x0), xmask)
tmp7 = tl.load(in_ptr2 + (x0), xmask)
tmp1 = tl.sigmoid(tmp0)
tmp3 = libdevice.tanh(tmp2)
tmp4 = tmp1 * tmp3
tmp5 = 1.0
tmp6 = tmp5 - tmp1
tmp8 = tmp6 * tmp7
tmp9 = tmp4 + tmp8
tl.store(out_ptr0 + (x0), tmp9, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (4, 8), (8, 1))
assert_size_stride(primals_6, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [r_f], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(primals_1, primals_2, buf0, 128, grid=grid(128), stream=stream0)
del primals_2
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_4, reinterpret_tensor(buf0, (16, 8), (8, 1), 0), reinterpret_tensor(primals_3, (8, 4), (1, 8), 0), alpha=1, beta=1, out=buf1)
del primals_3
del primals_4
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_6, reinterpret_tensor(buf0, (16, 8), (8, 1), 0), reinterpret_tensor(primals_5, (8, 4), (1, 8), 0), alpha=1, beta=1, out=buf2)
del primals_5
del primals_6
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [r, g, mul, sub, mul_1, o], Original ATen: [aten.tanh, aten.sigmoid, aten.mul, aten.rsub, aten.add]
triton_poi_fused_add_mul_rsub_sigmoid_tanh_1.run(buf2, buf1, primals_1, buf3, 64, grid=grid(64), stream=stream0)
return (buf3, primals_1, reinterpret_tensor(buf0, (16, 8), (8, 1), 0), buf1, buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 8), (8, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 8), (8, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class SFU(nn.Module):
"""Semantic Fusion Unit
The ouput vector is expected to not only retrieve correlative information from fusion vectors,
but also retain partly unchange as the input vector
"""
def __init__(self, input_size, fusion_size):
super(SFU, self).__init__()
self.linear_r = nn.Linear(input_size + fusion_size, input_size)
self.linear_g = nn.Linear(input_size + fusion_size, input_size)
def forward(self, x, fusions):
r_f = torch.cat([x, fusions], 2)
r = torch.tanh(self.linear_r(r_f))
g = torch.sigmoid(self.linear_g(r_f))
o = g * r + (1 - g) * x
return o
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'input_size': 4, 'fusion_size': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_add_mul_rsub_sigmoid_tanh_1(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp2 = tl.load(in_ptr1 + x0, xmask)
tmp7 = tl.load(in_ptr2 + x0, xmask)
tmp1 = tl.sigmoid(tmp0)
tmp3 = libdevice.tanh(tmp2)
tmp4 = tmp1 * tmp3
tmp5 = 1.0
tmp6 = tmp5 - tmp1
tmp8 = tmp6 * tmp7
tmp9 = tmp4 + tmp8
tl.store(out_ptr0 + x0, tmp9, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4, 8), (8, 1))
assert_size_stride(primals_6, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(128)](primals_1, primals_2, buf0, 128,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_4, reinterpret_tensor(buf0, (16, 8), (
8, 1), 0), reinterpret_tensor(primals_3, (8, 4), (1, 8), 0),
alpha=1, beta=1, out=buf1)
del primals_3
del primals_4
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_6, reinterpret_tensor(buf0, (16, 8), (
8, 1), 0), reinterpret_tensor(primals_5, (8, 4), (1, 8), 0),
alpha=1, beta=1, out=buf2)
del primals_5
del primals_6
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_mul_rsub_sigmoid_tanh_1[grid(64)](buf2, buf1,
primals_1, buf3, 64, XBLOCK=64, num_warps=1, num_stages=1)
return buf3, primals_1, reinterpret_tensor(buf0, (16, 8), (8, 1), 0
), buf1, buf2
class SFUNew(nn.Module):
"""Semantic Fusion Unit
The ouput vector is expected to not only retrieve correlative information from fusion vectors,
but also retain partly unchange as the input vector
"""
def __init__(self, input_size, fusion_size):
super(SFUNew, self).__init__()
self.linear_r = nn.Linear(input_size + fusion_size, input_size)
self.linear_g = nn.Linear(input_size + fusion_size, input_size)
def forward(self, input_0, input_1):
primals_3 = self.linear_r.weight
primals_4 = self.linear_r.bias
primals_5 = self.linear_g.weight
primals_6 = self.linear_g.bias
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| MobtgZhang/MWMLNet | SFU | false | 5,608 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import torch
import torch.nn as nn
class Model(nn.Module):
"""Semantic Fusion Unit
The ouput vector is expected to not only retrieve correlative information from fusion vectors,
but also retain partly unchange as the input vector
"""
def __init__(self, input_size, fusion_size):
super().__init__()
self.linear_r = nn.Linear(input_size + fusion_size, input_size)
self.linear_g = nn.Linear(input_size + fusion_size, input_size)
def forward(self, x, fusions):
r_f = torch.cat([x, fusions], 2)
r = torch.tanh(self.linear_r(r_f))
g = torch.sigmoid(self.linear_g(r_f))
o = g * r + (1 - g) * x
return o
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
Attention | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/nu/cnuc7ivckuuly7yn2763pwt3sw72jd6vuwpeeu4sfespm5iz7fq4.py
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# p_attn => exp
# Graph fragment:
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_2, 1), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%mul_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_tensor, %amax_default), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor, 2.0), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor,), kwargs = {})
triton_poi_fused__softmax_0 = async_compile.triton('triton_poi_fused__softmax_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp3 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + (x2), tmp17, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/fj/cfjl47pvhwbpfbvh6rfehwy5ijxc5p3zgkld2lwf3mw5bl6pbkak.py
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# p_attn => div_1, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_1 = async_compile.triton('triton_poi_fused__softmax_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(arg1_1, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(arg0_1, (16, 4, 4), (16, 1, 4), 0), out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [p_attn], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf1, buf2, 256, grid=grid(256), stream=stream0)
buf3 = reinterpret_tensor(buf1, (16, 4, 4), (16, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [matmul_1], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf3)
del arg2_1
return (reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf2, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg2_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1, arg2_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn.functional as F
import torch.nn as nn
class Attention(nn.Module):
"""
Compute 'Scaled Dot Product Attention
"""
def forward(self, query, key, value, mask=None, dropout=None):
scores = torch.matmul(query, key.transpose(-2, -1)) / math.sqrt(query
.size(-1))
if mask is not None:
scores = scores.masked_fill(mask == 0, -1000000000.0)
p_attn = F.softmax(scores, dim=-1)
if dropout is not None:
p_attn = dropout(p_attn)
return torch.matmul(p_attn, value), p_attn
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp3 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp3 * tmp1
tmp6 = tmp5 * tmp1
tmp7 = triton_helpers.maximum(tmp4, tmp6)
tmp9 = tmp8 * tmp1
tmp10 = triton_helpers.maximum(tmp7, tmp9)
tmp12 = tmp11 * tmp1
tmp13 = triton_helpers.maximum(tmp10, tmp12)
tmp14 = tmp2 - tmp13
tmp15 = 0.5
tmp16 = tmp14 * tmp15
tmp17 = tl_math.exp(tmp16)
tl.store(out_ptr0 + x2, tmp17, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(arg1_1, (16, 4, 4), (16, 4, 1
), 0), reinterpret_tensor(arg0_1, (16, 4, 4), (16, 1, 4), 0),
out=buf0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
triton_poi_fused__softmax_1[grid(256)](buf1, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf3 = reinterpret_tensor(buf1, (16, 4, 4), (16, 4, 1), 0)
del buf1
extern_kernels.bmm(reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf3
)
del arg2_1
return reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf2
class AttentionNew(nn.Module):
"""
Compute 'Scaled Dot Product Attention
"""
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0], output[1]
| MobtgZhang/MWMLNet | Attention | false | 5,609 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import math
import torch
import torch.nn.functional as F
import torch.nn as nn
class Model(nn.Module):
"""
Compute 'Scaled Dot Product Attention
"""
def forward(self, query, key, value, mask=None, dropout=None):
scores = torch.matmul(query, key.transpose(-2, -1)) / math.sqrt(query
.size(-1))
if mask is not None:
scores = scores.masked_fill(mask == 0, -1000000000.0)
p_attn = F.softmax(scores, dim=-1)
if dropout is not None:
p_attn = dropout(p_attn)
return torch.matmul(p_attn, value), p_attn
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return []
|
Sharpen_Block | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/hi/chieqt2p7y3xojubgmoetn3r5tmgtaq7aft5irb7n2uvwm6zuiqy.py
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.reflection_pad2d]
# Source node to ATen node mapping:
# pad => _unsafe_index, _unsafe_index_1
# Graph fragment:
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%arg0_1, [None, None, %sub_1, None]), kwargs = {})
# %_unsafe_index_1 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index, [None, None, None, %sub_3]), kwargs = {})
triton_poi_fused_reflection_pad2d_0 = async_compile.triton('triton_poi_fused_reflection_pad2d_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_reflection_pad2d_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_reflection_pad2d_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = (xindex // 6) % 6
x2 = (xindex // 36)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + ((-1)*(tl_math.abs((-3) + (tl_math.abs((-1) + x0))))) + ((-4)*(tl_math.abs((-3) + (tl_math.abs((-1) + x1))))) + (16*x2)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 1, 4, 4), (16, 16, 4, 1))
assert_size_stride(arg1_1, (1, 1, 3, 3), (9, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 6, 6), (36, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.reflection_pad2d]
stream0 = get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0.run(arg0_1, buf0, 144, grid=grid(144), stream=stream0)
del arg0_1
# Topologically Sorted Source Nodes: [pad, conv2d], Original ATen: [aten.reflection_pad2d, aten.convolution]
buf1 = extern_kernels.convolution(buf0, arg1_1, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 1, 4, 4), (16, 16, 4, 1))
del arg1_1
del buf0
return (buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 1, 4, 4), (16, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((1, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import numpy as np
import torch.nn as nn
class Sharpen_Block(nn.Module):
def __init__(self):
super(Sharpen_Block, self).__init__()
self.pad = nn.ReflectionPad2d((1, 1, 1, 1))
self.conv = nn.Conv2d(1, 1, 3, 1, 0, bias=False)
self.conv.weight = nn.Parameter(torch.from_numpy(np.array([[[[0, -
0.4, 0], [0, 2.6, 0], [0, -0.4, 0]]]])).float())
self.conv.weight.requires_grad = False
def forward(self, x):
return self.conv(self.pad(x))
def get_inputs():
return [torch.rand([4, 1, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_reflection_pad2d_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = xindex // 6 % 6
x2 = xindex // 36
x3 = xindex
tmp0 = tl.load(in_ptr0 + (15 + -1 * tl_math.abs(-3 + tl_math.abs(-1 +
x0)) + -4 * tl_math.abs(-3 + tl_math.abs(-1 + x1)) + 16 * x2),
xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + x3, tmp0, xmask)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 1, 4, 4), (16, 16, 4, 1))
assert_size_stride(arg1_1, (1, 1, 3, 3), (9, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 6, 6), (36, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0[grid(144)](arg0_1, buf0, 144,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
buf1 = extern_kernels.convolution(buf0, arg1_1, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 1, 4, 4), (16, 16, 4, 1))
del arg1_1
del buf0
return buf1,
class Sharpen_BlockNew(nn.Module):
def __init__(self):
super(Sharpen_BlockNew, self).__init__()
self.pad = nn.ReflectionPad2d((1, 1, 1, 1))
self.conv = nn.Conv2d(1, 1, 3, 1, 0, bias=False)
self.conv.weight = nn.Parameter(torch.from_numpy(np.array([[[[0, -
0.4, 0], [0, 2.6, 0], [0, -0.4, 0]]]])).float())
self.conv.weight.requires_grad = False
def forward(self, input_0):
arg1_1 = self.conv.weight
arg0_1 = input_0
output = call([arg0_1, arg1_1])
return output[0]
| MingSun-Tse/pytorch-vdsr | Sharpen_Block | false | 5,610 | [
"MIT"
] | 1 | 597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | https://github.com/MingSun-Tse/pytorch-vdsr/tree/597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | import torch
import numpy as np
import torch.nn as nn
class Model(nn.Module):
def __init__(self):
super().__init__()
self.pad = nn.ReflectionPad2d((1, 1, 1, 1))
self.conv = nn.Conv2d(1, 1, 3, 1, 0, bias=False)
self.conv.weight = nn.Parameter(torch.from_numpy(np.array([[[[0, -
0.4, 0], [0, 2.6, 0], [0, -0.4, 0]]]])).float())
self.conv.weight.requires_grad = False
def forward(self, x):
return self.conv(self.pad(x))
def get_inputs():
return [torch.rand([4, 1, 4, 4])]
def get_init_inputs():
return []
|
SmallVDSR_F8 | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/of/cofqdm26j473y3qugzsd27huz6mlmh4piozcssdqtsv6l7udq3bj.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# y => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution,), kwargs = {})
triton_poi_fused_relu_0 = async_compile.triton('triton_poi_fused_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[131072],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_relu_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 131072
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + (x0), tmp2, None)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21 = args
args.clear()
assert_size_stride(primals_1, (8, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_5, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_6, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_7, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_8, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_9, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_10, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_11, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_12, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_13, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_14, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_15, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_16, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_17, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_18, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_19, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_20, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_21, (1, 8, 3, 3), (72, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [y_1], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf3, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [y_2], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf5, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_3], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [y_3], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf7, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [y_4], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf9, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_5], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf11, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_6], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [y_6], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf13, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_7], Original ATen: [aten.convolution]
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf15 = buf14; del buf14 # reuse
# Topologically Sorted Source Nodes: [y_7], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf15, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_8], Original ATen: [aten.convolution]
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [y_8], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf17, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_9], Original ATen: [aten.convolution]
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf19, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_10], Original ATen: [aten.convolution]
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf21 = buf20; del buf20 # reuse
# Topologically Sorted Source Nodes: [y_10], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf21, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_11], Original ATen: [aten.convolution]
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf23 = buf22; del buf22 # reuse
# Topologically Sorted Source Nodes: [y_11], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf23, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_12], Original ATen: [aten.convolution]
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf25 = buf24; del buf24 # reuse
# Topologically Sorted Source Nodes: [y_12], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf25, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_13], Original ATen: [aten.convolution]
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf27 = buf26; del buf26 # reuse
# Topologically Sorted Source Nodes: [y_13], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf27, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_14], Original ATen: [aten.convolution]
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf29 = buf28; del buf28 # reuse
# Topologically Sorted Source Nodes: [y_14], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf29, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_15], Original ATen: [aten.convolution]
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf31 = buf30; del buf30 # reuse
# Topologically Sorted Source Nodes: [y_15], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf31, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_16], Original ATen: [aten.convolution]
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf33 = buf32; del buf32 # reuse
# Topologically Sorted Source Nodes: [y_16], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf33, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_17], Original ATen: [aten.convolution]
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf35 = buf34; del buf34 # reuse
# Topologically Sorted Source Nodes: [y_17], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf35, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [conv2d_18], Original ATen: [aten.convolution]
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf37 = buf36; del buf36 # reuse
# Topologically Sorted Source Nodes: [y_18], Original ATen: [aten.relu]
triton_poi_fused_relu_0.run(buf37, 131072, grid=grid(131072), stream=stream0)
# Topologically Sorted Source Nodes: [y_19], Original ATen: [aten.convolution]
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21, buf1, buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21, buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((8, 1, 3, 3), (9, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 1, 64, 64), (4096, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_18 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_19 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_20 = rand_strided((8, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_21 = rand_strided((1, 8, 3, 3), (72, 9, 3, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17, primals_18, primals_19, primals_20, primals_21])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class SmallVDSR_F8(nn.Module):
def __init__(self, model=False, fixed=False):
super(SmallVDSR_F8, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 8, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(8, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def forward(self, y):
y = self.relu(self.conv1(y))
y = self.relu(self.conv2(y))
y = self.relu(self.conv3(y))
y = self.relu(self.conv4(y))
y = self.relu(self.conv5(y))
y = self.relu(self.conv6(y))
y = self.relu(self.conv7(y))
y = self.relu(self.conv8(y))
y = self.relu(self.conv9(y))
y = self.relu(self.conv10(y))
y = self.relu(self.conv11(y))
y = self.relu(self.conv12(y))
y = self.relu(self.conv13(y))
y = self.relu(self.conv14(y))
y = self.relu(self.conv15(y))
y = self.relu(self.conv16(y))
y = self.relu(self.conv17(y))
y = self.relu(self.conv18(y))
y = self.relu(self.conv19(y))
y = self.conv20(y)
return y
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_relu_0(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(in_out_ptr0 + x0, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17,
primals_18, primals_19, primals_20, primals_21) = args
args.clear()
assert_size_stride(primals_1, (8, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_2, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_3, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_4, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_5, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_6, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_7, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_8, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_9, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_10, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_11, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_12, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_13, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_14, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_15, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_16, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_17, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_18, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_19, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_20, (8, 8, 3, 3), (72, 9, 3, 1))
assert_size_stride(primals_21, (1, 8, 3, 3), (72, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(131072)](buf1, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf2 = extern_kernels.convolution(buf1, primals_3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_relu_0[grid(131072)](buf3, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_relu_0[grid(131072)](buf5, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf6 = extern_kernels.convolution(buf5, primals_5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_relu_0[grid(131072)](buf7, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf8 = extern_kernels.convolution(buf7, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_relu_0[grid(131072)](buf9, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf10 = extern_kernels.convolution(buf9, primals_7, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_relu_0[grid(131072)](buf11, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_relu_0[grid(131072)](buf13, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf14 = extern_kernels.convolution(buf13, primals_9, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf15 = buf14
del buf14
triton_poi_fused_relu_0[grid(131072)](buf15, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf17 = buf16
del buf16
triton_poi_fused_relu_0[grid(131072)](buf17, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf18 = extern_kernels.convolution(buf17, primals_11, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf18, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf19 = buf18
del buf18
triton_poi_fused_relu_0[grid(131072)](buf19, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf21 = buf20
del buf20
triton_poi_fused_relu_0[grid(131072)](buf21, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf22 = extern_kernels.convolution(buf21, primals_13, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf23 = buf22
del buf22
triton_poi_fused_relu_0[grid(131072)](buf23, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf24 = extern_kernels.convolution(buf23, primals_14, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf24, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf25 = buf24
del buf24
triton_poi_fused_relu_0[grid(131072)](buf25, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf26 = extern_kernels.convolution(buf25, primals_15, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf26, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf27 = buf26
del buf26
triton_poi_fused_relu_0[grid(131072)](buf27, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf28 = extern_kernels.convolution(buf27, primals_16, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf28, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf29 = buf28
del buf28
triton_poi_fused_relu_0[grid(131072)](buf29, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf30 = extern_kernels.convolution(buf29, primals_17, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf30, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf31 = buf30
del buf30
triton_poi_fused_relu_0[grid(131072)](buf31, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf32 = extern_kernels.convolution(buf31, primals_18, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf32, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf33 = buf32
del buf32
triton_poi_fused_relu_0[grid(131072)](buf33, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf34 = extern_kernels.convolution(buf33, primals_19, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf34, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf35 = buf34
del buf34
triton_poi_fused_relu_0[grid(131072)](buf35, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf36 = extern_kernels.convolution(buf35, primals_20, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 8, 64, 64), (32768, 4096, 64, 1))
buf37 = buf36
del buf36
triton_poi_fused_relu_0[grid(131072)](buf37, 131072, XBLOCK=512,
num_warps=8, num_stages=1)
buf38 = extern_kernels.convolution(buf37, primals_21, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 1, 64, 64), (4096, 4096, 64, 1))
return (buf38, primals_1, primals_2, primals_3, primals_4, primals_5,
primals_6, primals_7, primals_8, primals_9, primals_10, primals_11,
primals_12, primals_13, primals_14, primals_15, primals_16,
primals_17, primals_18, primals_19, primals_20, primals_21, buf1,
buf3, buf5, buf7, buf9, buf11, buf13, buf15, buf17, buf19, buf21,
buf23, buf25, buf27, buf29, buf31, buf33, buf35, buf37)
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class SmallVDSR_F8New(nn.Module):
def __init__(self, model=False, fixed=False):
super(SmallVDSR_F8New, self).__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 8, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(8, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
out20 = y
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19,
out20)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_3 = self.conv2.weight
primals_4 = self.conv3.weight
primals_5 = self.conv4.weight
primals_6 = self.conv5.weight
primals_7 = self.conv6.weight
primals_8 = self.conv7.weight
primals_9 = self.conv8.weight
primals_10 = self.conv9.weight
primals_11 = self.conv10.weight
primals_12 = self.conv11.weight
primals_13 = self.conv12.weight
primals_14 = self.conv13.weight
primals_15 = self.conv14.weight
primals_16 = self.conv15.weight
primals_17 = self.conv16.weight
primals_18 = self.conv17.weight
primals_19 = self.conv18.weight
primals_20 = self.conv19.weight
primals_21 = self.conv20.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21])
return output[0]
| MingSun-Tse/pytorch-vdsr | SmallVDSR_F8 | false | 5,611 | [
"MIT"
] | 1 | 597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | https://github.com/MingSun-Tse/pytorch-vdsr/tree/597bacb4ec7385c8cc6cdf91e26e64ef2e6808b7 | import torch
import torch.nn as nn
def load_param(model1_path, model2):
dict_param1 = torch.load(model1_path)
dict_param2 = dict(model2.named_parameters())
for name2 in dict_param2:
if name2 in dict_param1:
dict_param2[name2].data.copy_(dict_param1[name2].data)
model2.load_state_dict(dict_param2)
return model2
class Model(nn.Module):
def __init__(self, model=False, fixed=False):
super().__init__()
self.fixed = fixed
self.conv1 = nn.Conv2d(1, 8, 3, 1, 1, bias=False)
self.conv2 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv3 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv4 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv5 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv6 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv7 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv8 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv9 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv10 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv11 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv12 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv13 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv14 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv15 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv16 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv17 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv18 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv19 = nn.Conv2d(8, 8, 3, 1, 1, bias=False)
self.conv20 = nn.Conv2d(8, 1, 3, 1, 1, bias=False)
self.relu = nn.ReLU()
if model:
load_param(model, self)
if fixed:
for param in self.parameters():
param.requires_grad = False
def forward_stem(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
out16 = y
y = self.relu(self.conv17(y))
out17 = y
y = self.relu(self.conv18(y))
out18 = y
y = self.relu(self.conv19(y))
out19 = y
y = self.conv20(y)
return (out1, out2, out3, out4, out5, out6, out7, out8, out9, out10,
out11, out12, out13, out14, out15, out16, out17, out18, out19, y)
def forward_dense(self, y):
y = self.relu(self.conv1(y))
out1 = y
y = self.relu(self.conv2(y))
out2 = y
y = self.relu(self.conv3(y))
out3 = y
y = self.relu(self.conv4(y))
out4 = y
y = self.relu(self.conv5(y))
out5 = y
y = self.relu(self.conv6(y))
out6 = y
y = self.relu(self.conv7(y))
out7 = y
y = self.relu(self.conv8(y))
out8 = y
y = self.relu(self.conv9(y))
out9 = y
y = self.relu(self.conv10(y))
out10 = y
y = self.relu(self.conv11(y))
out11 = y
y = self.relu(self.conv12(y))
out12 = y
y = self.relu(self.conv13(y))
out13 = y
y = self.relu(self.conv14(y))
out14 = y
y = self.relu(self.conv15(y))
out15 = y
y = self.relu(self.conv16(y))
# ... truncated (>4000 chars) for memory efficiency |
Dense_block | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/o4/co4zyxcl2qkqwco2hpqjmusaq55kf7f6hyk5tkf5vs6buom4p47w.py
# Topologically Sorted Source Nodes: [leaky_relu], Original ATen: [aten.leaky_relu]
# Source node to ATen node mapping:
# leaky_relu => gt, mul, where
# Graph fragment:
# %gt : [num_users=2] = call_function[target=torch.ops.aten.gt.Scalar](args = (%view_1, 0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, 0.2), kwargs = {})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%gt, %view_1, %mul), kwargs = {})
triton_poi_fused_leaky_relu_0 = async_compile.triton('triton_poi_fused_leaky_relu_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i1', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_leaky_relu_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 2, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_leaky_relu_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.2
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(out_ptr0 + (x2), tmp4, xmask)
tl.store(out_ptr1 + (x2), tmp7, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [leaky_relu], Original ATen: [aten.leaky_relu]
stream0 = get_raw_stream(0)
triton_poi_fused_leaky_relu_0.run(buf0, primals_2, buf1, buf2, 256, grid=grid(256), stream=stream0)
del buf0
del primals_2
return (buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class Dense_block(nn.Module):
""" This is the initial dense block as in the paper """
def __init__(self, in_channels, out_channels):
super(Dense_block, self).__init__()
self.Dense = torch.nn.Linear(in_channels, out_channels)
nn.init.xavier_uniform(self.Dense.weight.data, 1.0)
self.activation = torch.nn.LeakyReLU(0.2)
def forward(self, x):
return self.activation(self.Dense(x))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_leaky_relu_0(in_ptr0, in_ptr1, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.2
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr1 + x2, tmp7, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_leaky_relu_0[grid(256)](buf0, primals_2, buf1,
buf2, 256, XBLOCK=256, num_warps=4, num_stages=1)
del buf0
del primals_2
return buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1
class Dense_blockNew(nn.Module):
""" This is the initial dense block as in the paper """
def __init__(self, in_channels, out_channels):
super(Dense_blockNew, self).__init__()
self.Dense = torch.nn.Linear(in_channels, out_channels)
nn.init.xavier_uniform(self.Dense.weight.data, 1.0)
self.activation = torch.nn.LeakyReLU(0.2)
def forward(self, input_0):
primals_1 = self.Dense.weight
primals_2 = self.Dense.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| Mohanned-Elkholy/ResNet-GAN | Dense_block | false | 5,612 | [
"MIT"
] | 1 | 81b01294d8b5035131aee24d486e2cb879030832 | https://github.com/Mohanned-Elkholy/ResNet-GAN/tree/81b01294d8b5035131aee24d486e2cb879030832 | import torch
import torch.nn as nn
class Model(nn.Module):
""" This is the initial dense block as in the paper """
def __init__(self, in_channels, out_channels):
super().__init__()
self.Dense = torch.nn.Linear(in_channels, out_channels)
nn.init.xavier_uniform(self.Dense.weight.data, 1.0)
self.activation = torch.nn.LeakyReLU(0.2)
def forward(self, x):
return self.activation(self.Dense(x))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
MatrixTree | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/mw/cmwulpxxw2jt763yw6dihufmcwawkztohwf6my2rtf33kl2nbgds.py
# Topologically Sorted Source Nodes: [eye, ne, lap, sum_1], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
# Source node to ATen node mapping:
# eye => eq, full_default, full_default_1, iota_1, where
# lap => full_default_2, where_1
# ne => ne
# sum_1 => sum_1
# Graph fragment:
# %iota_1 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze, %iota_1), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq, %full_default, %full_default_1), kwargs = {})
# %ne : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where, 0), kwargs = {})
# %full_default_2 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_1 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne, %full_default_2, %select), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%where_1, [0]), kwargs = {})
triton_poi_fused_eye_masked_fill_ne_sum_0 = async_compile.triton('triton_poi_fused_eye_masked_fill_ne_sum_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_eye_masked_fill_ne_sum_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (x0), xmask)
tmp16 = tl.load(in_ptr0 + (4 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (8 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (12 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + (x0), tmp38, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5f/c5fvishccr56l7e7obdlxzywjkbh3davnf5cdvnuqxihtgywozfd.py
# Topologically Sorted Source Nodes: [eye, ne, lap, neg, diag, lap_1, diag_1, exp_1], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
# Source node to ATen node mapping:
# diag => eq_1, full_default_3, iota_2, where_2
# diag_1 => diagonal_copy
# exp_1 => exp_1
# eye => eq, full_default, full_default_1, iota_1, where
# lap => full_default_2, where_1
# lap_1 => add_1
# ne => ne
# neg => neg
# Graph fragment:
# %iota_1 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze, %iota_1), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq, %full_default, %full_default_1), kwargs = {})
# %ne : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where, 0), kwargs = {})
# %full_default_2 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_1 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne, %full_default_2, %select), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%where_1,), kwargs = {})
# %iota_2 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_1 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_2, %unsqueeze_2), kwargs = {})
# %full_default_3 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_2 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_1, %permute, %full_default_3), kwargs = {})
# %add_1 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%neg, %where_2), kwargs = {})
# %diagonal_copy : [num_users=1] = call_function[target=torch.ops.aten.diagonal_copy.default](args = (%select_1,), kwargs = {})
# %exp_1 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%diagonal_copy,), kwargs = {})
# %select_scatter_default : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%add_1, %exp_1, 0, 0), kwargs = {})
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1 = async_compile.triton('triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (5*x0), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (x2), xmask)
tmp18 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + (x2), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/5k/c5kir7ouxg6vi6ka3fqtxb3atjlynhq3dcipimsepx55zvv5ls6k.py
# Topologically Sorted Source Nodes: [eye_1, ne_1, lap_2, sum_2], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
# Source node to ATen node mapping:
# eye_1 => eq_3, full_default_7, full_default_8, iota_7, where_4
# lap_2 => full_default_9, where_5
# ne_1 => ne_1
# sum_2 => sum_2
# Graph fragment:
# %iota_7 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_3 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_6, %iota_7), kwargs = {})
# %full_default_7 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_8 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_4 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_3, %full_default_7, %full_default_8), kwargs = {})
# %ne_1 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_4, 0), kwargs = {})
# %full_default_9 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_5 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_1, %full_default_9, %select_20), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%where_5, [0]), kwargs = {})
triton_poi_fused_eye_masked_fill_ne_sum_2 = async_compile.triton('triton_poi_fused_eye_masked_fill_ne_sum_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_eye_masked_fill_ne_sum_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (16 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (20 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (24 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (28 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + (x0), tmp38, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ny/cnyxgqp3tmqmvliho4wkdr5wi6tnqd2oycsfgq3cjapc2phvva3v.py
# Topologically Sorted Source Nodes: [eye_1, ne_1, lap_2, neg_1, diag_5, lap_3, diag_6, exp_5], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
# Source node to ATen node mapping:
# diag_5 => eq_4, full_default_10, iota_8, where_6
# diag_6 => diagonal_copy_3
# exp_5 => exp_5
# eye_1 => eq_3, full_default_7, full_default_8, iota_7, where_4
# lap_2 => full_default_9, where_5
# lap_3 => add_3
# ne_1 => ne_1
# neg_1 => neg_1
# Graph fragment:
# %iota_7 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_3 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_6, %iota_7), kwargs = {})
# %full_default_7 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_8 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_4 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_3, %full_default_7, %full_default_8), kwargs = {})
# %ne_1 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_4, 0), kwargs = {})
# %full_default_9 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_5 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_1, %full_default_9, %select_20), kwargs = {})
# %neg_1 : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%where_5,), kwargs = {})
# %iota_8 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_4 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_8, %unsqueeze_8), kwargs = {})
# %full_default_10 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_6 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_4, %permute_5, %full_default_10), kwargs = {})
# %add_3 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%neg_1, %where_6), kwargs = {})
# %diagonal_copy_3 : [num_users=1] = call_function[target=torch.ops.aten.diagonal_copy.default](args = (%select_21,), kwargs = {})
# %exp_5 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%diagonal_copy_3,), kwargs = {})
# %select_scatter_default_4 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%add_3, %exp_5, 0, 0), kwargs = {})
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3 = async_compile.triton('triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (16 + (5*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (16 + x2), xmask)
tmp18 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + (x2), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/76/c764os7jr65oallcf3eupkgmn4rajwnkb7qa2a4qrpedfegb2xji.py
# Topologically Sorted Source Nodes: [diag_4, add_2], Original ATen: [aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_2 => add_2
# diag_4 => eq_2, full_default_6, iota_4, where_3
# Graph fragment:
# %iota_4 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_2 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_4, %unsqueeze_5), kwargs = {})
# %full_default_6 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_3 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_2, %permute_4, %full_default_6), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_16, %where_3), kwargs = {})
triton_poi_fused_add_diag_embed_4 = async_compile.triton('triton_poi_fused_add_diag_embed_4', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_4', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp4 = tl.load(in_ptr0 + (x2), xmask)
tmp6 = tl.load(in_ptr1 + (5*x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (x2), xmask)
tmp18 = tl.load(in_ptr0 + (5*x0), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 0, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tmp2 == tmp0
tmp5 = tl_math.exp(tmp4)
tmp7 = tmp5 * tmp6
tmp8 = 0.0
tmp9 = tl.where(tmp3, tmp8, tmp7)
tmp10 = x1
tmp11 = tmp10 == tmp0
tmp13 = tmp5 * tmp12
tmp14 = tl.where(tmp11, tmp8, tmp13)
tmp15 = tmp9 - tmp14
tmp16 = tl.where(tmp1, tmp15, tmp4)
tmp17 = tmp2 == tmp10
tmp19 = tl_math.exp(tmp18)
tmp21 = tmp19 * tmp20
tmp22 = tl.where(tmp17, tmp21, tmp8)
tmp23 = tmp16 + tmp22
tl.store(out_ptr0 + (x2), tmp23, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ry/crysshrrigakeht5r6neboffuaxbkjaegs3cqmvdij2ytziz5vck.py
# Topologically Sorted Source Nodes: [exp_2, mul, setitem_1, exp_3, mul_1, setitem_2, sub, diag_4, add_2], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_2 => add_2
# diag_4 => eq_2, full_default_6, iota_4, where_3
# exp_2 => exp_2
# exp_3 => exp_3
# mul => mul
# mul_1 => mul_1
# setitem_1 => copy_1, full_default_4
# setitem_2 => copy_2, full_default_5
# sub => sub
# Graph fragment:
# %exp_2 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_5,), kwargs = {})
# %mul : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_2, %permute_1), kwargs = {})
# %full_default_4 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_1 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_7, %full_default_4), kwargs = {})
# %select_scatter_default_1 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul, %copy_1, 1, 0), kwargs = {})
# %exp_3 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_6,), kwargs = {})
# %mul_1 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_3, %permute_2), kwargs = {})
# %full_default_5 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_2 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_9, %full_default_5), kwargs = {})
# %select_scatter_default_2 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_1, %copy_2, 0, 0), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%select_scatter_default_1, %select_scatter_default_2), kwargs = {})
# %select_scatter_default_3 : [num_users=3] = call_function[target=torch.ops.aten.select_scatter.default](args = (%arg0_1, %sub, 0, 0), kwargs = {})
# %iota_4 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_2 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_4, %unsqueeze_5), kwargs = {})
# %full_default_6 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_3 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_2, %permute_4, %full_default_6), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_16, %where_3), kwargs = {})
# %select_scatter_default_5 : [num_users=2] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_3, %add_2, 0, 0), kwargs = {})
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5 = async_compile.triton('triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 16)
x3 = xindex % 16
x0 = xindex % 4
x1 = (xindex // 4) % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr1 + (x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr2 + (5*x0), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr1 + (x5), xmask)
tmp0 = x2
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tmp4 == tmp1
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 * tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp5, tmp10, tmp9)
tmp12 = x1
tmp13 = tmp12 == tmp1
tmp15 = tmp7 * tmp14
tmp16 = tl.where(tmp13, tmp10, tmp15)
tmp17 = tmp11 - tmp16
tmp19 = tl.where(tmp2, tmp17, tmp18)
tmp20 = tl.where(tmp2, tmp3, tmp19)
tl.store(out_ptr0 + (x5), tmp20, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ds/cdsgimfrigpgjvo5w5eo5m4acelh6kweus6l5g6oa5euxfo5zryv.py
# Topologically Sorted Source Nodes: [eye_2, ne_2, lap_4, sum_3], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
# Source node to ATen node mapping:
# eye_2 => eq_6, full_default_14, full_default_15, iota_13, where_8
# lap_4 => full_default_16, where_9
# ne_2 => ne_2
# sum_3 => sum_3
# Graph fragment:
# %iota_13 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_6 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_12, %iota_13), kwargs = {})
# %full_default_14 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_15 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_8 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_6, %full_default_14, %full_default_15), kwargs = {})
# %ne_2 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_8, 0), kwargs = {})
# %full_default_16 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_9 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_2, %full_default_16, %select_41), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%where_9, [0]), kwargs = {})
triton_poi_fused_eye_masked_fill_ne_sum_6 = async_compile.triton('triton_poi_fused_eye_masked_fill_ne_sum_6', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_eye_masked_fill_ne_sum_6', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_6(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (32 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (36 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (40 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (44 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + (x0), tmp38, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/bx/cbx52hvmunlgryyn5hbdsqyfni653yiknzgubzvil2q6yaseomvy.py
# Topologically Sorted Source Nodes: [eye_2, ne_2, lap_4, neg_2, diag_10, lap_5, diag_11, exp_9], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
# Source node to ATen node mapping:
# diag_10 => eq_7, full_default_17, iota_14, where_10
# diag_11 => diagonal_copy_6
# exp_9 => exp_9
# eye_2 => eq_6, full_default_14, full_default_15, iota_13, where_8
# lap_4 => full_default_16, where_9
# lap_5 => add_5
# ne_2 => ne_2
# neg_2 => neg_2
# Graph fragment:
# %iota_13 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_6 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_12, %iota_13), kwargs = {})
# %full_default_14 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_15 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_8 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_6, %full_default_14, %full_default_15), kwargs = {})
# %ne_2 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_8, 0), kwargs = {})
# %full_default_16 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_9 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_2, %full_default_16, %select_41), kwargs = {})
# %neg_2 : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%where_9,), kwargs = {})
# %iota_14 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_7 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_14, %unsqueeze_14), kwargs = {})
# %full_default_17 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_10 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_7, %permute_10, %full_default_17), kwargs = {})
# %add_5 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%neg_2, %where_10), kwargs = {})
# %diagonal_copy_6 : [num_users=1] = call_function[target=torch.ops.aten.diagonal_copy.default](args = (%select_42,), kwargs = {})
# %exp_9 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%diagonal_copy_6,), kwargs = {})
# %select_scatter_default_9 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%add_5, %exp_9, 0, 0), kwargs = {})
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7 = async_compile.triton('triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (32 + (5*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (32 + x2), xmask)
tmp18 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + (x2), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/mz/cmzl5zoke7j2kqgzcvjpkpdzkg2kwyb47nzibimhi2hklzb5s6ut.py
# Topologically Sorted Source Nodes: [diag_9, add_4], Original ATen: [aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_4 => add_4
# diag_9 => eq_5, full_default_13, iota_10, where_7
# Graph fragment:
# %iota_10 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_5 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_10, %unsqueeze_11), kwargs = {})
# %full_default_13 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_7 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_5, %permute_9, %full_default_13), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_37, %where_7), kwargs = {})
triton_poi_fused_add_diag_embed_8 = async_compile.triton('triton_poi_fused_add_diag_embed_8', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_8', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_8(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp5 = tl.load(in_ptr0 + (16 + x2), xmask)
tmp7 = tl.load(in_ptr1 + (5*x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (x2), xmask)
tmp17 = tl.load(in_ptr2 + (16 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (16 + (5*x0)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 1, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + (x2), tmp25, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/4m/c4mfewqtg74k4hoz5gcjnz5z3hfa6tzmoe5hz6cil2jwbmx7cosw.py
# Topologically Sorted Source Nodes: [exp_6, mul_3, setitem_6, exp_7, mul_4, setitem_7, sub_1, diag_9, add_4], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_4 => add_4
# diag_9 => eq_5, full_default_13, iota_10, where_7
# exp_6 => exp_6
# exp_7 => exp_7
# mul_3 => mul_3
# mul_4 => mul_4
# setitem_6 => copy_6, full_default_11
# setitem_7 => copy_7, full_default_12
# sub_1 => sub_1
# Graph fragment:
# %exp_6 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_25,), kwargs = {})
# %mul_3 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_6, %permute_6), kwargs = {})
# %full_default_11 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_6 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_27, %full_default_11), kwargs = {})
# %select_scatter_default_6 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_3, %copy_6, 1, 0), kwargs = {})
# %exp_7 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_26,), kwargs = {})
# %mul_4 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_7, %permute_7), kwargs = {})
# %full_default_12 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_7 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_29, %full_default_12), kwargs = {})
# %select_scatter_default_7 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_4, %copy_7, 0, 0), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%select_scatter_default_6, %select_scatter_default_7), kwargs = {})
# %select_scatter_default_8 : [num_users=3] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_5, %sub_1, 0, 1), kwargs = {})
# %iota_10 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_5 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_10, %unsqueeze_11), kwargs = {})
# %full_default_13 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_7 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_5, %permute_9, %full_default_13), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_37, %where_7), kwargs = {})
# %select_scatter_default_10 : [num_users=2] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_8, %add_4, 0, 1), kwargs = {})
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9 = async_compile.triton('triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 16)
x3 = xindex % 16
x0 = xindex % 4
x1 = (xindex // 4) % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (16 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + (5*x0), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + (x5), xmask)
tmp0 = x2
tmp1 = tl.full([1], 1, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + (x5), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/f4/cf43dymhxq3xler6ijzfxsjgm3oemfpzsjaoarspoimxqod3bkiv.py
# Topologically Sorted Source Nodes: [eye_3, ne_3, lap_6, sum_4], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
# Source node to ATen node mapping:
# eye_3 => eq_9, full_default_21, full_default_22, iota_19, where_12
# lap_6 => full_default_23, where_13
# ne_3 => ne_3
# sum_4 => sum_4
# Graph fragment:
# %iota_19 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_9 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_18, %iota_19), kwargs = {})
# %full_default_21 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_22 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_12 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_9, %full_default_21, %full_default_22), kwargs = {})
# %ne_3 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_12, 0), kwargs = {})
# %full_default_23 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_13 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_3, %full_default_23, %select_62), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%where_13, [0]), kwargs = {})
triton_poi_fused_eye_masked_fill_ne_sum_10 = async_compile.triton('triton_poi_fused_eye_masked_fill_ne_sum_10', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_eye_masked_fill_ne_sum_10', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_10(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (48 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (52 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (56 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (60 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + (x0), tmp38, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/sg/csgmzo5h24yeta5phehu6lzwmojkxl6w5o6pxa7fu4tywve6h72o.py
# Topologically Sorted Source Nodes: [eye_3, ne_3, lap_6, neg_3, diag_15, lap_7, diag_16, exp_13], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
# Source node to ATen node mapping:
# diag_15 => eq_10, full_default_24, iota_20, where_14
# diag_16 => diagonal_copy_9
# exp_13 => exp_13
# eye_3 => eq_9, full_default_21, full_default_22, iota_19, where_12
# lap_6 => full_default_23, where_13
# lap_7 => add_7
# ne_3 => ne_3
# neg_3 => neg_3
# Graph fragment:
# %iota_19 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_9 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze_18, %iota_19), kwargs = {})
# %full_default_21 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([1], 1), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %full_default_22 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_12 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_9, %full_default_21, %full_default_22), kwargs = {})
# %ne_3 : [num_users=1] = call_function[target=torch.ops.aten.ne.Scalar](args = (%where_12, 0), kwargs = {})
# %full_default_23 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_13 : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ne_3, %full_default_23, %select_62), kwargs = {})
# %neg_3 : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%where_13,), kwargs = {})
# %iota_20 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_10 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_20, %unsqueeze_20), kwargs = {})
# %full_default_24 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_14 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_10, %permute_15, %full_default_24), kwargs = {})
# %add_7 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%neg_3, %where_14), kwargs = {})
# %diagonal_copy_9 : [num_users=1] = call_function[target=torch.ops.aten.diagonal_copy.default](args = (%select_63,), kwargs = {})
# %exp_13 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%diagonal_copy_9,), kwargs = {})
# %select_scatter_default_14 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%add_7, %exp_13, 0, 0), kwargs = {})
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11 = async_compile.triton('triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 4)
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (48 + (5*x0)), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (48 + x2), xmask)
tmp18 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + (x2), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/s3/cs3gvd4vxrcwwln23ccoasr7omajc7scigkbutbxw7gq332qtsp2.py
# Topologically Sorted Source Nodes: [diag_14, add_6], Original ATen: [aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_6 => add_6
# diag_14 => eq_8, full_default_20, iota_16, where_11
# Graph fragment:
# %iota_16 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_8 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_16, %unsqueeze_17), kwargs = {})
# %full_default_20 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_11 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_8, %permute_14, %full_default_20), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_58, %where_11), kwargs = {})
triton_poi_fused_add_diag_embed_12 = async_compile.triton('triton_poi_fused_add_diag_embed_12', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_12', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_12(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp5 = tl.load(in_ptr0 + (32 + x2), xmask)
tmp7 = tl.load(in_ptr1 + (5*x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (x2), xmask)
tmp17 = tl.load(in_ptr2 + (32 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (32 + (5*x0)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 2, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + (x2), tmp25, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/me/cmej2qnos2ywrczshe25nqkszxvcm53ics5lgh4v3txzow4adwk6.py
# Topologically Sorted Source Nodes: [exp_10, mul_6, setitem_11, exp_11, mul_7, setitem_12, sub_2, diag_14, add_6], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_6 => add_6
# diag_14 => eq_8, full_default_20, iota_16, where_11
# exp_10 => exp_10
# exp_11 => exp_11
# mul_6 => mul_6
# mul_7 => mul_7
# setitem_11 => copy_11, full_default_18
# setitem_12 => copy_12, full_default_19
# sub_2 => sub_2
# Graph fragment:
# %exp_10 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_46,), kwargs = {})
# %mul_6 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_10, %permute_11), kwargs = {})
# %full_default_18 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_11 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_48, %full_default_18), kwargs = {})
# %select_scatter_default_11 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_6, %copy_11, 1, 0), kwargs = {})
# %exp_11 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_47,), kwargs = {})
# %mul_7 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_11, %permute_12), kwargs = {})
# %full_default_19 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_12 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_50, %full_default_19), kwargs = {})
# %select_scatter_default_12 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_7, %copy_12, 0, 0), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%select_scatter_default_11, %select_scatter_default_12), kwargs = {})
# %select_scatter_default_13 : [num_users=3] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_10, %sub_2, 0, 2), kwargs = {})
# %iota_16 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_8 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_16, %unsqueeze_17), kwargs = {})
# %full_default_20 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_11 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_8, %permute_14, %full_default_20), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_58, %where_11), kwargs = {})
# %select_scatter_default_15 : [num_users=2] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_13, %add_6, 0, 2), kwargs = {})
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13 = async_compile.triton('triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 16)
x3 = xindex % 16
x0 = xindex % 4
x1 = (xindex // 4) % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (32 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + (5*x0), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + (x5), xmask)
tmp0 = x2
tmp1 = tl.full([1], 2, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + (x5), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/7m/c7mcscnhipdy5wvhd42yefmq35fu4oqgmcy7p53evfyoamhb6fsp.py
# Topologically Sorted Source Nodes: [diag_19, add_8], Original ATen: [aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_8 => add_8
# diag_19 => eq_11, full_default_27, iota_22, where_15
# Graph fragment:
# %iota_22 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_11 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_22, %unsqueeze_23), kwargs = {})
# %full_default_27 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_15 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_11, %permute_19, %full_default_27), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_79, %where_15), kwargs = {})
triton_poi_fused_add_diag_embed_14 = async_compile.triton('triton_poi_fused_add_diag_embed_14', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[16],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_14', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_14(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = (xindex // 4)
tmp5 = tl.load(in_ptr0 + (48 + x2), xmask)
tmp7 = tl.load(in_ptr1 + (5*x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (x2), xmask)
tmp17 = tl.load(in_ptr2 + (48 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (48 + (5*x0)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 3, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + (x2), tmp25, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/xd/cxdxskm7n54mdwp6h2mbtokxdjclp2po3usor7fw3tqm473pzik6.py
# Topologically Sorted Source Nodes: [exp_14, mul_9, setitem_16, exp_15, mul_10, setitem_17, sub_3, diag_19, add_8], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
# Source node to ATen node mapping:
# add_8 => add_8
# diag_19 => eq_11, full_default_27, iota_22, where_15
# exp_14 => exp_14
# exp_15 => exp_15
# mul_10 => mul_10
# mul_9 => mul_9
# setitem_16 => copy_16, full_default_25
# setitem_17 => copy_17, full_default_26
# sub_3 => sub_3
# Graph fragment:
# %exp_14 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_67,), kwargs = {})
# %mul_9 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_14, %permute_16), kwargs = {})
# %full_default_25 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_16 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_69, %full_default_25), kwargs = {})
# %select_scatter_default_16 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_9, %copy_16, 1, 0), kwargs = {})
# %exp_15 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%select_68,), kwargs = {})
# %mul_10 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%exp_15, %permute_17), kwargs = {})
# %full_default_26 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %copy_17 : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%select_71, %full_default_26), kwargs = {})
# %select_scatter_default_17 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%mul_10, %copy_17, 0, 0), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%select_scatter_default_16, %select_scatter_default_17), kwargs = {})
# %select_scatter_default_18 : [num_users=3] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_15, %sub_3, 0, 3), kwargs = {})
# %iota_22 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (4,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %eq_11 : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%iota_22, %unsqueeze_23), kwargs = {})
# %full_default_27 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_15 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%eq_11, %permute_19, %full_default_27), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%select_79, %where_15), kwargs = {})
# %select_scatter_default_19 : [num_users=1] = call_function[target=torch.ops.aten.select_scatter.default](args = (%select_scatter_default_18, %add_8, 0, 3), kwargs = {})
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15 = async_compile.triton('triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 5, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 16)
x3 = xindex % 16
x0 = xindex % 4
x1 = (xindex // 4) % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (48 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + (5*x0), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + (x5), xmask)
tmp0 = x2
tmp1 = tl.full([1], 3, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + (x5), tmp21, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [eye, ne, lap, sum_1], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
stream0 = get_raw_stream(0)
triton_poi_fused_eye_masked_fill_ne_sum_0.run(arg0_1, buf0, 4, grid=grid(4), stream=stream0)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [eye, ne, lap, neg, diag, lap_1, diag_1, exp_1], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1.run(arg0_1, buf0, buf1, 16, grid=grid(16), stream=stream0)
# Topologically Sorted Source Nodes: [eye, ne, lap, neg, diag, lap_1, diag_1, exp_1, inv_laplacian], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp, aten.linalg_inv_ex]
buf2 = torch.ops.aten.linalg_inv_ex.default(buf1)
buf3 = buf2[0]
del buf2
buf5 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [eye_1, ne_1, lap_2, sum_2], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
triton_poi_fused_eye_masked_fill_ne_sum_2.run(arg0_1, buf5, 4, grid=grid(4), stream=stream0)
buf6 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [eye_1, ne_1, lap_2, neg_1, diag_5, lap_3, diag_6, exp_5], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3.run(arg0_1, buf5, buf6, 16, grid=grid(16), stream=stream0)
# Topologically Sorted Source Nodes: [eye_1, ne_1, lap_2, neg_1, diag_5, lap_3, diag_6, exp_5, inv_laplacian_1], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp, aten.linalg_inv_ex]
buf7 = torch.ops.aten.linalg_inv_ex.default(buf6)
buf8 = buf7[0]
del buf7
buf10 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [diag_4, add_2], Original ATen: [aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_4.run(arg0_1, buf3, buf10, 16, grid=grid(16), stream=stream0)
buf11 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [exp_2, mul, setitem_1, exp_3, mul_1, setitem_2, sub, diag_4, add_2], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5.run(buf10, arg0_1, buf3, buf11, 64, grid=grid(64), stream=stream0)
del buf10
buf12 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [eye_2, ne_2, lap_4, sum_3], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
triton_poi_fused_eye_masked_fill_ne_sum_6.run(arg0_1, buf12, 4, grid=grid(4), stream=stream0)
buf13 = reinterpret_tensor(buf3, (4, 4), (4, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [eye_2, ne_2, lap_4, neg_2, diag_10, lap_5, diag_11, exp_9], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7.run(arg0_1, buf12, buf13, 16, grid=grid(16), stream=stream0)
# Topologically Sorted Source Nodes: [eye_2, ne_2, lap_4, neg_2, diag_10, lap_5, diag_11, exp_9, inv_laplacian_2], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp, aten.linalg_inv_ex]
buf14 = torch.ops.aten.linalg_inv_ex.default(buf13)
buf15 = buf14[0]
del buf14
buf17 = buf13; del buf13 # reuse
# Topologically Sorted Source Nodes: [diag_9, add_4], Original ATen: [aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_8.run(arg0_1, buf8, buf11, buf17, 16, grid=grid(16), stream=stream0)
buf18 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [exp_6, mul_3, setitem_6, exp_7, mul_4, setitem_7, sub_1, diag_9, add_4], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9.run(buf18, buf17, arg0_1, buf8, 64, grid=grid(64), stream=stream0)
del buf17
buf19 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [eye_3, ne_3, lap_6, sum_4], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.sum]
triton_poi_fused_eye_masked_fill_ne_sum_10.run(arg0_1, buf19, 4, grid=grid(4), stream=stream0)
buf20 = reinterpret_tensor(buf8, (4, 4), (4, 1), 0); del buf8 # reuse
# Topologically Sorted Source Nodes: [eye_3, ne_3, lap_6, neg_3, diag_15, lap_7, diag_16, exp_13], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp]
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11.run(arg0_1, buf19, buf20, 16, grid=grid(16), stream=stream0)
del buf19
# Topologically Sorted Source Nodes: [eye_3, ne_3, lap_6, neg_3, diag_15, lap_7, diag_16, exp_13, inv_laplacian_3], Original ATen: [aten.eye, aten.ne, aten.masked_fill, aten.neg, aten.diag_embed, aten.add, aten.diagonal_copy, aten.exp, aten.linalg_inv_ex]
buf21 = torch.ops.aten.linalg_inv_ex.default(buf20)
buf22 = buf21[0]
del buf21
buf24 = buf20; del buf20 # reuse
# Topologically Sorted Source Nodes: [diag_14, add_6], Original ATen: [aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_12.run(arg0_1, buf15, buf18, buf24, 16, grid=grid(16), stream=stream0)
buf25 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [exp_10, mul_6, setitem_11, exp_11, mul_7, setitem_12, sub_2, diag_14, add_6], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13.run(buf25, buf24, arg0_1, buf15, 64, grid=grid(64), stream=stream0)
del buf15
buf26 = buf24; del buf24 # reuse
# Topologically Sorted Source Nodes: [diag_19, add_8], Original ATen: [aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_14.run(arg0_1, buf22, buf25, buf26, 16, grid=grid(16), stream=stream0)
buf27 = buf25; del buf25 # reuse
# Topologically Sorted Source Nodes: [exp_14, mul_9, setitem_16, exp_15, mul_10, setitem_17, sub_3, diag_19, add_8], Original ATen: [aten.exp, aten.mul, aten.lift_fresh, aten.fill, aten.sub, aten.diag_embed, aten.add]
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15.run(buf27, buf26, arg0_1, buf22, 64, grid=grid(64), stream=stream0)
del arg0_1
del buf22
del buf26
return (buf27, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4), (16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.cuda
import torch.distributed
class MatrixTree(nn.Module):
"""Implementation of the matrix-tree theorem for computing marginals
of non-projective dependency parsing. This attention layer is used
in the paper "Learning Structured Text Representations"
:cite:`DBLP:journals/corr/LiuL17d`.
"""
def __init__(self, eps=1e-05):
self.eps = eps
super(MatrixTree, self).__init__()
def forward(self, input):
laplacian = input.exp() + self.eps
output = input.clone()
for b in range(input.size(0)):
lap = laplacian[b].masked_fill(torch.eye(input.size(1), device=
input.device).ne(0), 0)
lap = -lap + torch.diag(lap.sum(0))
lap[0] = input[b].diag().exp()
inv_laplacian = lap.inverse()
factor = inv_laplacian.diag().unsqueeze(1).expand_as(input[b]
).transpose(0, 1)
term1 = input[b].exp().mul(factor).clone()
term2 = input[b].exp().mul(inv_laplacian.transpose(0, 1)).clone()
term1[:, 0] = 0
term2[0] = 0
output[b] = term1 - term2
roots_output = input[b].diag().exp().mul(inv_laplacian.
transpose(0, 1)[0])
output[b] = output[b] + torch.diag(roots_output)
return output
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {}]
| import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
import torch.cuda
import torch.distributed
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + x0, xmask)
tmp16 = tl.load(in_ptr0 + (4 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (8 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (12 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + x0, tmp38, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1(
in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + 5 * x0, xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + x2, xmask)
tmp18 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + x2, tmp21, xmask)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_2(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (16 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (20 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (24 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (28 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + x0, tmp38, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3(
in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (16 + 5 * x0), xmask, eviction_policy='evict_last'
)
tmp11 = tl.load(in_ptr0 + (16 + x2), xmask)
tmp18 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + x2, tmp21, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_4(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp4 = tl.load(in_ptr0 + x2, xmask)
tmp6 = tl.load(in_ptr1 + 5 * x0, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + x2, xmask)
tmp18 = tl.load(in_ptr0 + 5 * x0, xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 0, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tmp2 == tmp0
tmp5 = tl_math.exp(tmp4)
tmp7 = tmp5 * tmp6
tmp8 = 0.0
tmp9 = tl.where(tmp3, tmp8, tmp7)
tmp10 = x1
tmp11 = tmp10 == tmp0
tmp13 = tmp5 * tmp12
tmp14 = tl.where(tmp11, tmp8, tmp13)
tmp15 = tmp9 - tmp14
tmp16 = tl.where(tmp1, tmp15, tmp4)
tmp17 = tmp2 == tmp10
tmp19 = tl_math.exp(tmp18)
tmp21 = tmp19 * tmp20
tmp22 = tl.where(tmp17, tmp21, tmp8)
tmp23 = tmp16 + tmp22
tl.store(out_ptr0 + x2, tmp23, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5(in_ptr0,
in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 16
x3 = xindex % 16
x0 = xindex % 4
x1 = xindex // 4 % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr1 + x3, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr2 + 5 * x0, xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr1 + x5, xmask)
tmp0 = x2
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tmp4 == tmp1
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 * tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp5, tmp10, tmp9)
tmp12 = x1
tmp13 = tmp12 == tmp1
tmp15 = tmp7 * tmp14
tmp16 = tl.where(tmp13, tmp10, tmp15)
tmp17 = tmp11 - tmp16
tmp19 = tl.where(tmp2, tmp17, tmp18)
tmp20 = tl.where(tmp2, tmp3, tmp19)
tl.store(out_ptr0 + x5, tmp20, xmask)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_6(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (32 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (36 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (40 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (44 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + x0, tmp38, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7(
in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (32 + 5 * x0), xmask, eviction_policy='evict_last'
)
tmp11 = tl.load(in_ptr0 + (32 + x2), xmask)
tmp18 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + x2, tmp21, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_8(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp5 = tl.load(in_ptr0 + (16 + x2), xmask)
tmp7 = tl.load(in_ptr1 + 5 * x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + x2, xmask)
tmp17 = tl.load(in_ptr2 + (16 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (16 + 5 * x0), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 1, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + x2, tmp25, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 16
x3 = xindex % 16
x0 = xindex % 4
x1 = xindex // 4 % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (16 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + 5 * x0, xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + x5, xmask)
tmp0 = x2
tmp1 = tl.full([1], 1, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + x5, tmp21, xmask)
@triton.jit
def triton_poi_fused_eye_masked_fill_ne_sum_10(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp7 = tl.load(in_ptr0 + (48 + x0), xmask)
tmp16 = tl.load(in_ptr0 + (52 + x0), xmask)
tmp25 = tl.load(in_ptr0 + (56 + x0), xmask)
tmp34 = tl.load(in_ptr0 + (60 + x0), xmask)
tmp0 = tl.full([1], 0, tl.int64)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = 1.0
tmp4 = 0.0
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = tmp5 != tmp4
tmp8 = tl_math.exp(tmp7)
tmp9 = 1e-05
tmp10 = tmp8 + tmp9
tmp11 = tl.where(tmp6, tmp4, tmp10)
tmp12 = tl.full([1], 1, tl.int64)
tmp13 = tmp12 == tmp1
tmp14 = tl.where(tmp13, tmp3, tmp4)
tmp15 = tmp14 != tmp4
tmp17 = tl_math.exp(tmp16)
tmp18 = tmp17 + tmp9
tmp19 = tl.where(tmp15, tmp4, tmp18)
tmp20 = tmp11 + tmp19
tmp21 = tl.full([1], 2, tl.int64)
tmp22 = tmp21 == tmp1
tmp23 = tl.where(tmp22, tmp3, tmp4)
tmp24 = tmp23 != tmp4
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp26 + tmp9
tmp28 = tl.where(tmp24, tmp4, tmp27)
tmp29 = tmp20 + tmp28
tmp30 = tl.full([1], 3, tl.int64)
tmp31 = tmp30 == tmp1
tmp32 = tl.where(tmp31, tmp3, tmp4)
tmp33 = tmp32 != tmp4
tmp35 = tl_math.exp(tmp34)
tmp36 = tmp35 + tmp9
tmp37 = tl.where(tmp33, tmp4, tmp36)
tmp38 = tmp29 + tmp37
tl.store(out_ptr0 + x0, tmp38, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11(
in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp3 = tl.load(in_ptr0 + (48 + 5 * x0), xmask, eviction_policy='evict_last'
)
tmp11 = tl.load(in_ptr0 + (48 + x2), xmask)
tmp18 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = tl_math.exp(tmp3)
tmp5 = x0
tmp6 = tmp0 == tmp5
tmp7 = 1.0
tmp8 = 0.0
tmp9 = tl.where(tmp6, tmp7, tmp8)
tmp10 = tmp9 != tmp8
tmp12 = tl_math.exp(tmp11)
tmp13 = 1e-05
tmp14 = tmp12 + tmp13
tmp15 = tl.where(tmp10, tmp8, tmp14)
tmp16 = -tmp15
tmp17 = tmp5 == tmp0
tmp19 = tl.where(tmp17, tmp18, tmp8)
tmp20 = tmp16 + tmp19
tmp21 = tl.where(tmp2, tmp4, tmp20)
tl.store(out_ptr0 + x2, tmp21, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_12(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp5 = tl.load(in_ptr0 + (32 + x2), xmask)
tmp7 = tl.load(in_ptr1 + 5 * x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + x2, xmask)
tmp17 = tl.load(in_ptr2 + (32 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (32 + 5 * x0), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 2, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + x2, tmp25, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 16
x3 = xindex % 16
x0 = xindex % 4
x1 = xindex // 4 % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (32 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + 5 * x0, xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + x5, xmask)
tmp0 = x2
tmp1 = tl.full([1], 2, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + x5, tmp21, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_14(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp5 = tl.load(in_ptr0 + (48 + x2), xmask)
tmp7 = tl.load(in_ptr1 + 5 * x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + x2, xmask)
tmp17 = tl.load(in_ptr2 + (48 + x2), xmask)
tmp20 = tl.load(in_ptr0 + (48 + 5 * x0), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp0 = tl.full([1], 3, tl.int32)
tmp1 = tmp0 == tmp0
tmp2 = x0
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = tmp2 == tmp3
tmp6 = tl_math.exp(tmp5)
tmp8 = tmp6 * tmp7
tmp9 = 0.0
tmp10 = tl.where(tmp4, tmp9, tmp8)
tmp11 = x1
tmp12 = tmp11 == tmp3
tmp14 = tmp6 * tmp13
tmp15 = tl.where(tmp12, tmp9, tmp14)
tmp16 = tmp10 - tmp15
tmp18 = tl.where(tmp1, tmp16, tmp17)
tmp19 = tmp2 == tmp11
tmp21 = tl_math.exp(tmp20)
tmp23 = tmp21 * tmp22
tmp24 = tl.where(tmp19, tmp23, tmp9)
tmp25 = tmp18 + tmp24
tl.store(out_ptr0 + x2, tmp25, xmask)
@triton.jit
def triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 16
x3 = xindex % 16
x0 = xindex % 4
x1 = xindex // 4 % 4
x5 = xindex
tmp3 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (48 + x3), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr2 + 5 * x0, xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_out_ptr0 + x5, xmask)
tmp0 = x2
tmp1 = tl.full([1], 3, tl.int32)
tmp2 = tmp0 == tmp1
tmp4 = x0
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp4 == tmp5
tmp8 = tl_math.exp(tmp7)
tmp10 = tmp8 * tmp9
tmp11 = 0.0
tmp12 = tl.where(tmp6, tmp11, tmp10)
tmp13 = x1
tmp14 = tmp13 == tmp5
tmp16 = tmp8 * tmp15
tmp17 = tl.where(tmp14, tmp11, tmp16)
tmp18 = tmp12 - tmp17
tmp20 = tl.where(tmp2, tmp18, tmp19)
tmp21 = tl.where(tmp2, tmp3, tmp20)
tl.store(in_out_ptr0 + x5, tmp21, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4,), (1,), torch.float32)
get_raw_stream(0)
triton_poi_fused_eye_masked_fill_ne_sum_0[grid(4)](arg0_1, buf0, 4,
XBLOCK=4, num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_1[
grid(16)](arg0_1, buf0, buf1, 16, XBLOCK=16, num_warps=1,
num_stages=1)
buf2 = torch.ops.aten.linalg_inv_ex.default(buf1)
buf3 = buf2[0]
del buf2
buf5 = buf0
del buf0
triton_poi_fused_eye_masked_fill_ne_sum_2[grid(4)](arg0_1, buf5, 4,
XBLOCK=4, num_warps=1, num_stages=1)
buf6 = buf1
del buf1
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_3[
grid(16)](arg0_1, buf5, buf6, 16, XBLOCK=16, num_warps=1,
num_stages=1)
buf7 = torch.ops.aten.linalg_inv_ex.default(buf6)
buf8 = buf7[0]
del buf7
buf10 = buf6
del buf6
triton_poi_fused_add_diag_embed_4[grid(16)](arg0_1, buf3, buf10, 16,
XBLOCK=16, num_warps=1, num_stages=1)
buf11 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_5[grid(64)
](buf10, arg0_1, buf3, buf11, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del buf10
buf12 = buf5
del buf5
triton_poi_fused_eye_masked_fill_ne_sum_6[grid(4)](arg0_1, buf12, 4,
XBLOCK=4, num_warps=1, num_stages=1)
buf13 = reinterpret_tensor(buf3, (4, 4), (4, 1), 0)
del buf3
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_7[
grid(16)](arg0_1, buf12, buf13, 16, XBLOCK=16, num_warps=1,
num_stages=1)
buf14 = torch.ops.aten.linalg_inv_ex.default(buf13)
buf15 = buf14[0]
del buf14
buf17 = buf13
del buf13
triton_poi_fused_add_diag_embed_8[grid(16)](arg0_1, buf8, buf11,
buf17, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf18 = buf11
del buf11
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_9[grid(64)
](buf18, buf17, arg0_1, buf8, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del buf17
buf19 = buf12
del buf12
triton_poi_fused_eye_masked_fill_ne_sum_10[grid(4)](arg0_1, buf19,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf20 = reinterpret_tensor(buf8, (4, 4), (4, 1), 0)
del buf8
triton_poi_fused_add_diag_embed_diagonal_copy_exp_eye_masked_fill_ne_neg_11[
grid(16)](arg0_1, buf19, buf20, 16, XBLOCK=16, num_warps=1,
num_stages=1)
del buf19
buf21 = torch.ops.aten.linalg_inv_ex.default(buf20)
buf22 = buf21[0]
del buf21
buf24 = buf20
del buf20
triton_poi_fused_add_diag_embed_12[grid(16)](arg0_1, buf15, buf18,
buf24, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf25 = buf18
del buf18
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_13[grid(64)
](buf25, buf24, arg0_1, buf15, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del buf15
buf26 = buf24
del buf24
triton_poi_fused_add_diag_embed_14[grid(16)](arg0_1, buf22, buf25,
buf26, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf27 = buf25
del buf25
triton_poi_fused_add_diag_embed_exp_fill_lift_fresh_mul_sub_15[grid(64)
](buf27, buf26, arg0_1, buf22, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del arg0_1
del buf22
del buf26
return buf27,
class MatrixTreeNew(nn.Module):
"""Implementation of the matrix-tree theorem for computing marginals
of non-projective dependency parsing. This attention layer is used
in the paper "Learning Structured Text Representations"
:cite:`DBLP:journals/corr/LiuL17d`.
"""
def __init__(self, eps=1e-05):
self.eps = eps
super(MatrixTreeNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MaxatTezekbayev/OpenNMT-py-lexical | MatrixTree | false | 5,613 | [
"MIT"
] | 1 | 44182999b863fc4074d67e0281c5bdab19abddfe | https://github.com/MaxatTezekbayev/OpenNMT-py-lexical/tree/44182999b863fc4074d67e0281c5bdab19abddfe | import torch
import torch.nn as nn
import torch.cuda
import torch.distributed
class Model(nn.Module):
"""Implementation of the matrix-tree theorem for computing marginals
of non-projective dependency parsing. This attention layer is used
in the paper "Learning Structured Text Representations"
:cite:`DBLP:journals/corr/LiuL17d`.
"""
def __init__(self, eps=1e-05):
self.eps = eps
super().__init__()
def forward(self, input):
laplacian = input.exp() + self.eps
output = input.clone()
for b in range(input.size(0)):
lap = laplacian[b].masked_fill(torch.eye(input.size(1), device=
input.device).ne(0), 0)
lap = -lap + torch.diag(lap.sum(0))
lap[0] = input[b].diag().exp()
inv_laplacian = lap.inverse()
factor = inv_laplacian.diag().unsqueeze(1).expand_as(input[b]
).transpose(0, 1)
term1 = input[b].exp().mul(factor).clone()
term2 = input[b].exp().mul(inv_laplacian.transpose(0, 1)).clone()
term1[:, 0] = 0
term2[0] = 0
output[b] = term1 - term2
roots_output = input[b].diag().exp().mul(inv_laplacian.
transpose(0, 1)[0])
output[b] = output[b] + torch.diag(roots_output)
return output
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return []
|
RandomShiftsAug | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/sq/csqzbjgigod7iwdpblwlrtqegrwxplnexkan6oxv2b3voplnbg6n.py
# Topologically Sorted Source Nodes: [base_grid], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# base_grid => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%unsqueeze_1, %permute], 2), kwargs = {})
triton_poi_fused_cat_0 = async_compile.triton('triton_poi_fused_cat_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_cat_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_cat_0(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 2
x1 = (xindex // 2) % 4
x2 = (xindex // 8)
x4 = xindex
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = x1
tmp6 = tmp5.to(tl.float32)
tmp7 = 6.0
tmp8 = tmp6 < tmp7
tmp9 = 0.16666666666666666
tmp10 = tmp6 * tmp9
tmp11 = -0.9166666666666666
tmp12 = tmp10 + tmp11
tmp13 = 11 + ((-1)*x1)
tmp14 = tmp13.to(tl.float32)
tmp15 = tmp14 * tmp9
tmp16 = 0.9166666666666666
tmp17 = tmp16 - tmp15
tmp18 = tl.where(tmp8, tmp12, tmp17)
tmp19 = tl.full(tmp18.shape, 0.0, tmp18.dtype)
tmp20 = tl.where(tmp4, tmp18, tmp19)
tmp21 = tmp0 >= tmp3
tmp22 = tl.full([1], 2, tl.int64)
tmp23 = tmp0 < tmp22
tmp24 = x2
tmp25 = tmp24.to(tl.float32)
tmp26 = tmp25 < tmp7
tmp27 = tmp25 * tmp9
tmp28 = tmp27 + tmp11
tmp29 = 11 + ((-1)*x2)
tmp30 = tmp29.to(tl.float32)
tmp31 = tmp30 * tmp9
tmp32 = tmp16 - tmp31
tmp33 = tl.where(tmp26, tmp28, tmp32)
tmp34 = tl.full(tmp33.shape, 0.0, tmp33.dtype)
tmp35 = tl.where(tmp21, tmp33, tmp34)
tmp36 = tl.where(tmp4, tmp20, tmp35)
tl.store(out_ptr0 + (x4), tmp36, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/ec/cec7fujvvbe4hi4zs7gswfcyerr6hgw7amw2cirx7s2wg224nm7z.py
# Topologically Sorted Source Nodes: [x_1, grid_sample], Original ATen: [aten.replication_pad2d, aten.grid_sampler_2d]
# Source node to ATen node mapping:
# grid_sample => add_2, add_3, add_4, add_5, add_6, add_7, add_8, convert_element_type_4, convert_element_type_5, convert_element_type_6, convert_element_type_7, convert_element_type_8, convert_element_type_9, floor, floor_1, full_default_10, full_default_11, full_default_2, full_default_3, full_default_4, full_default_5, full_default_6, full_default_7, full_default_8, full_default_9, ge, ge_1, ge_2, ge_3, ge_4, ge_5, ge_6, ge_7, index, index_1, index_2, index_3, logical_and, logical_and_1, logical_and_10, logical_and_11, logical_and_2, logical_and_3, logical_and_4, logical_and_5, logical_and_6, logical_and_7, logical_and_8, logical_and_9, lt_1, lt_2, lt_3, lt_4, lt_5, lt_6, lt_7, lt_8, mul_10, mul_11, mul_12, mul_3, mul_4, mul_5, mul_6, mul_7, mul_8, mul_9, sub_2, sub_3, sub_4, sub_5, sub_6, sub_7, sub_8, sub_9, where_10, where_11, where_12, where_3, where_4, where_5, where_6, where_7, where_8, where_9
# x_1 => _unsafe_index, _unsafe_index_1
# Graph fragment:
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%arg0_1, [None, None, %clamp_max, None]), kwargs = {})
# %_unsafe_index_1 : [num_users=4] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index, [None, None, None, %clamp_max_1]), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%select, 6.0), kwargs = {})
# %add_2 : [num_users=5] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_3, 5.5), kwargs = {})
# %floor : [num_users=9] = call_function[target=torch.ops.aten.floor.default](args = (%add_2,), kwargs = {})
# %ge : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%floor, 0), kwargs = {})
# %lt_1 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%floor, 12), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%select_1, 6.0), kwargs = {})
# %add_3 : [num_users=5] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, 5.5), kwargs = {})
# %floor_1 : [num_users=9] = call_function[target=torch.ops.aten.floor.default](args = (%add_3,), kwargs = {})
# %ge_1 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%floor_1, 0), kwargs = {})
# %lt_2 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%floor_1, 12), kwargs = {})
# %logical_and : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_1, %lt_2), kwargs = {})
# %logical_and_1 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%lt_1, %logical_and), kwargs = {})
# %logical_and_2 : [num_users=3] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge, %logical_and_1), kwargs = {})
# %index : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%_unsafe_index_1, [%view_1, %view_2, %where_2, %where_1]), kwargs = {})
# %add_4 : [num_users=8] = call_function[target=torch.ops.aten.add.Tensor](args = (%floor, 1), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_4, %add_2), kwargs = {})
# %add_5 : [num_users=8] = call_function[target=torch.ops.aten.add.Tensor](args = (%floor_1, 1), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_5, %add_3), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_2, %sub_3), kwargs = {})
# %full_default_2 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_3 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_2, %mul_5, %full_default_2), kwargs = {})
# %mul_9 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index, %where_3), kwargs = {})
# %ge_2 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%add_4, 0), kwargs = {})
# %lt_3 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%add_4, 12), kwargs = {})
# %ge_3 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%floor_1, 0), kwargs = {})
# %lt_4 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%floor_1, 12), kwargs = {})
# %logical_and_3 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_3, %lt_4), kwargs = {})
# %logical_and_4 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%lt_3, %logical_and_3), kwargs = {})
# %logical_and_5 : [num_users=3] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_2, %logical_and_4), kwargs = {})
# %convert_element_type_5 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%floor_1, torch.int64), kwargs = {})
# %full_default_4 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_5 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_5, %convert_element_type_5, %full_default_4), kwargs = {})
# %convert_element_type_4 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%add_4, torch.int64), kwargs = {})
# %full_default_3 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_4 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_5, %convert_element_type_4, %full_default_3), kwargs = {})
# %index_1 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%_unsafe_index_1, [%view_1, %view_2, %where_5, %where_4]), kwargs = {})
# %sub_4 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_2, %floor), kwargs = {})
# %sub_5 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_5, %add_3), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_4, %sub_5), kwargs = {})
# %full_default_5 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_6 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_5, %mul_6, %full_default_5), kwargs = {})
# %mul_10 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index_1, %where_6), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_9, %mul_10), kwargs = {})
# %ge_4 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%floor, 0), kwargs = {})
# %lt_5 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%floor, 12), kwargs = {})
# %ge_5 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%add_5, 0), kwargs = {})
# %lt_6 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%add_5, 12), kwargs = {})
# %logical_and_6 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_5, %lt_6), kwargs = {})
# %logical_and_7 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%lt_5, %logical_and_6), kwargs = {})
# %logical_and_8 : [num_users=3] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_4, %logical_and_7), kwargs = {})
# %convert_element_type_7 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%add_5, torch.int64), kwargs = {})
# %full_default_7 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_8 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_8, %convert_element_type_7, %full_default_7), kwargs = {})
# %convert_element_type_6 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%floor, torch.int64), kwargs = {})
# %full_default_6 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_7 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_8, %convert_element_type_6, %full_default_6), kwargs = {})
# %index_2 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%_unsafe_index_1, [%view_1, %view_2, %where_8, %where_7]), kwargs = {})
# %sub_6 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_4, %add_2), kwargs = {})
# %sub_7 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_3, %floor_1), kwargs = {})
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_6, %sub_7), kwargs = {})
# %full_default_8 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_9 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_8, %mul_7, %full_default_8), kwargs = {})
# %mul_11 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index_2, %where_9), kwargs = {})
# %add_7 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_6, %mul_11), kwargs = {})
# %ge_6 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%add_4, 0), kwargs = {})
# %lt_7 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%add_4, 12), kwargs = {})
# %ge_7 : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%add_5, 0), kwargs = {})
# %lt_8 : [num_users=1] = call_function[target=torch.ops.aten.lt.Scalar](args = (%add_5, 12), kwargs = {})
# %logical_and_9 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_7, %lt_8), kwargs = {})
# %logical_and_10 : [num_users=1] = call_function[target=torch.ops.aten.logical_and.default](args = (%lt_7, %logical_and_9), kwargs = {})
# %logical_and_11 : [num_users=3] = call_function[target=torch.ops.aten.logical_and.default](args = (%ge_6, %logical_and_10), kwargs = {})
# %convert_element_type_9 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%add_5, torch.int64), kwargs = {})
# %full_default_10 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_11 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_11, %convert_element_type_9, %full_default_10), kwargs = {})
# %convert_element_type_8 : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%add_4, torch.int64), kwargs = {})
# %full_default_9 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0), kwargs = {dtype: torch.int64, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_10 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_11, %convert_element_type_8, %full_default_9), kwargs = {})
# %index_3 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%_unsafe_index_1, [%view_1, %view_2, %where_11, %where_10]), kwargs = {})
# %sub_8 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_2, %floor), kwargs = {})
# %sub_9 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_3, %floor_1), kwargs = {})
# %mul_8 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_8, %sub_9), kwargs = {})
# %full_default_11 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], 0.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_12 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%logical_and_11, %mul_8, %full_default_11), kwargs = {})
# %mul_12 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%index_3, %where_12), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_7, %mul_12), kwargs = {})
triton_poi_fused_grid_sampler_2d_replication_pad2d_1 = async_compile.triton('triton_poi_fused_grid_sampler_2d_replication_pad2d_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_grid_sampler_2d_replication_pad2d_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 4, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_grid_sampler_2d_replication_pad2d_1(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x2 = (xindex // 64)
x5 = xindex
x3 = (xindex // 16)
tmp0 = tl.load(in_ptr0 + (2*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (2*x2), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr0 + (1 + (2*x0)), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr1 + (1 + (2*x2)), xmask, eviction_policy='evict_last')
tmp2 = 0.16666666666666666
tmp3 = tmp1 * tmp2
tmp4 = tmp0 + tmp3
tmp5 = 6.0
tmp6 = tmp4 * tmp5
tmp7 = 5.5
tmp8 = tmp6 + tmp7
tmp9 = libdevice.floor(tmp8)
tmp10 = 0.0
tmp11 = tmp9 >= tmp10
tmp12 = 12.0
tmp13 = tmp9 < tmp12
tmp16 = tmp15 * tmp2
tmp17 = tmp14 + tmp16
tmp18 = tmp17 * tmp5
tmp19 = tmp18 + tmp7
tmp20 = libdevice.floor(tmp19)
tmp21 = tmp20 >= tmp10
tmp22 = tmp20 < tmp12
tmp23 = tmp21 & tmp22
tmp24 = tmp13 & tmp23
tmp25 = tmp11 & tmp24
tmp26 = 1.0
tmp27 = tmp9 + tmp26
tmp28 = tmp27 - tmp8
tmp29 = tmp20 + tmp26
tmp30 = tmp29 - tmp19
tmp31 = tmp28 * tmp30
tmp32 = tl.where(tmp25, tmp31, tmp10)
tmp33 = tmp27 >= tmp10
tmp34 = tmp27 < tmp12
tmp35 = tmp34 & tmp23
tmp36 = tmp33 & tmp35
tmp37 = tmp8 - tmp9
tmp38 = tmp37 * tmp30
tmp39 = tl.where(tmp36, tmp38, tmp10)
tmp40 = tmp29 >= tmp10
tmp41 = tmp29 < tmp12
tmp42 = tmp40 & tmp41
tmp43 = tmp34 & tmp42
tmp44 = tmp33 & tmp43
tmp45 = tmp19 - tmp20
tmp46 = tmp37 * tmp45
tmp47 = tl.where(tmp44, tmp46, tmp10)
tmp48 = tmp13 & tmp42
tmp49 = tmp11 & tmp48
tmp50 = tmp28 * tmp45
tmp51 = tl.where(tmp49, tmp50, tmp10)
tmp52 = tmp20.to(tl.int64)
tmp53 = tl.full([1], 0, tl.int64)
tmp54 = tl.where(tmp25, tmp52, tmp53)
tmp55 = tl.full([XBLOCK], 12, tl.int32)
tmp56 = tmp54 + tmp55
tmp57 = tmp54 < 0
tmp58 = tl.where(tmp57, tmp56, tmp54)
tl.device_assert(((0 <= tmp58) & (tmp58 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp58 < 12")
tmp60 = tmp9.to(tl.int64)
tmp61 = tl.where(tmp25, tmp60, tmp53)
tmp62 = tmp61 + tmp55
tmp63 = tmp61 < 0
tmp64 = tl.where(tmp63, tmp62, tmp61)
tl.device_assert(((0 <= tmp64) & (tmp64 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp64 < 12")
tmp66 = tl.load(in_ptr2 + ((4*((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp58)) + ((-4) + tmp58) * (((-4) + tmp58) > (0))))) + (((0) * ((0) >= ((-4) + tmp58)) + ((-4) + tmp58) * (((-4) + tmp58) > (0)))) * ((((0) * ((0) >= ((-4) + tmp58)) + ((-4) + tmp58) * (((-4) + tmp58) > (0)))) < (3)))) + (16*x3) + ((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp64)) + ((-4) + tmp64) * (((-4) + tmp64) > (0))))) + (((0) * ((0) >= ((-4) + tmp64)) + ((-4) + tmp64) * (((-4) + tmp64) > (0)))) * ((((0) * ((0) >= ((-4) + tmp64)) + ((-4) + tmp64) * (((-4) + tmp64) > (0)))) < (3)))), xmask, eviction_policy='evict_last')
tmp67 = tl.where(tmp36, tmp52, tmp53)
tmp68 = tmp27.to(tl.int64)
tmp69 = tl.where(tmp36, tmp68, tmp53)
tmp70 = tmp29.to(tl.int64)
tmp71 = tl.where(tmp49, tmp70, tmp53)
tmp72 = tl.where(tmp49, tmp60, tmp53)
tmp73 = tmp66 * tmp32
tmp74 = tmp67 + tmp55
tmp75 = tmp67 < 0
tmp76 = tl.where(tmp75, tmp74, tmp67)
tl.device_assert(((0 <= tmp76) & (tmp76 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp76 < 12")
tmp78 = tmp69 + tmp55
tmp79 = tmp69 < 0
tmp80 = tl.where(tmp79, tmp78, tmp69)
tl.device_assert(((0 <= tmp80) & (tmp80 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp80 < 12")
tmp82 = tl.load(in_ptr2 + ((4*((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp76)) + ((-4) + tmp76) * (((-4) + tmp76) > (0))))) + (((0) * ((0) >= ((-4) + tmp76)) + ((-4) + tmp76) * (((-4) + tmp76) > (0)))) * ((((0) * ((0) >= ((-4) + tmp76)) + ((-4) + tmp76) * (((-4) + tmp76) > (0)))) < (3)))) + (16*x3) + ((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp80)) + ((-4) + tmp80) * (((-4) + tmp80) > (0))))) + (((0) * ((0) >= ((-4) + tmp80)) + ((-4) + tmp80) * (((-4) + tmp80) > (0)))) * ((((0) * ((0) >= ((-4) + tmp80)) + ((-4) + tmp80) * (((-4) + tmp80) > (0)))) < (3)))), xmask, eviction_policy='evict_last')
tmp83 = tmp82 * tmp39
tmp84 = tmp73 + tmp83
tmp85 = tmp71 + tmp55
tmp86 = tmp71 < 0
tmp87 = tl.where(tmp86, tmp85, tmp71)
tl.device_assert(((0 <= tmp87) & (tmp87 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp87 < 12")
tmp89 = tmp72 + tmp55
tmp90 = tmp72 < 0
tmp91 = tl.where(tmp90, tmp89, tmp72)
tl.device_assert(((0 <= tmp91) & (tmp91 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp91 < 12")
tmp93 = tl.load(in_ptr2 + ((4*((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp87)) + ((-4) + tmp87) * (((-4) + tmp87) > (0))))) + (((0) * ((0) >= ((-4) + tmp87)) + ((-4) + tmp87) * (((-4) + tmp87) > (0)))) * ((((0) * ((0) >= ((-4) + tmp87)) + ((-4) + tmp87) * (((-4) + tmp87) > (0)))) < (3)))) + (16*x3) + ((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp91)) + ((-4) + tmp91) * (((-4) + tmp91) > (0))))) + (((0) * ((0) >= ((-4) + tmp91)) + ((-4) + tmp91) * (((-4) + tmp91) > (0)))) * ((((0) * ((0) >= ((-4) + tmp91)) + ((-4) + tmp91) * (((-4) + tmp91) > (0)))) < (3)))), xmask, eviction_policy='evict_last')
tmp94 = tmp93 * tmp51
tmp95 = tmp84 + tmp94
tmp96 = tl.where(tmp44, tmp70, tmp53)
tmp97 = tl.where(tmp44, tmp68, tmp53)
tmp98 = tmp96 + tmp55
tmp99 = tmp96 < 0
tmp100 = tl.where(tmp99, tmp98, tmp96)
tl.device_assert(((0 <= tmp100) & (tmp100 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp100 < 12")
tmp102 = tmp97 + tmp55
tmp103 = tmp97 < 0
tmp104 = tl.where(tmp103, tmp102, tmp97)
tl.device_assert(((0 <= tmp104) & (tmp104 < 12)) | ~(xmask), "index out of bounds: 0 <= tmp104 < 12")
tmp106 = tl.load(in_ptr2 + ((4*((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp100)) + ((-4) + tmp100) * (((-4) + tmp100) > (0))))) + (((0) * ((0) >= ((-4) + tmp100)) + ((-4) + tmp100) * (((-4) + tmp100) > (0)))) * ((((0) * ((0) >= ((-4) + tmp100)) + ((-4) + tmp100) * (((-4) + tmp100) > (0)))) < (3)))) + (16*x3) + ((3) * ((3) <= (((0) * ((0) >= ((-4) + tmp104)) + ((-4) + tmp104) * (((-4) + tmp104) > (0))))) + (((0) * ((0) >= ((-4) + tmp104)) + ((-4) + tmp104) * (((-4) + tmp104) > (0)))) * ((((0) * ((0) >= ((-4) + tmp104)) + ((-4) + tmp104) * (((-4) + tmp104) > (0)))) < (3)))), xmask, eviction_policy='evict_last')
tmp107 = tmp106 * tmp47
tmp108 = tmp95 + tmp107
tl.store(in_out_ptr0 + (x5), tmp108, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [base_grid], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(buf0, 32, grid=grid(32), stream=stream0)
# Topologically Sorted Source Nodes: [shift], Original ATen: [aten.randint]
buf1 = torch.ops.aten.randint.low(0, 9, [4, 1, 1, 2], dtype=torch.float32, device=device(type='cuda', index=0), pin_memory=False)
buf2 = buf1
del buf1
buf10 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf11 = buf10; del buf10 # reuse
buf15 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [x_1, grid_sample], Original ATen: [aten.replication_pad2d, aten.grid_sampler_2d]
triton_poi_fused_grid_sampler_2d_replication_pad2d_1.run(buf15, buf0, buf2, arg0_1, 256, grid=grid(256), stream=stream0)
del arg0_1
del buf0
del buf2
return (buf15, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
import torch.nn.functional as F
class RandomShiftsAug(nn.Module):
def __init__(self, pad):
super().__init__()
self.pad = pad
def forward(self, x):
x = x.float()
n, _c, h, w = x.size()
assert h == w
padding = tuple([self.pad] * 4)
x = F.pad(x, padding, 'replicate')
eps = 1.0 / (h + 2 * self.pad)
arange = torch.linspace(-1.0 + eps, 1.0 - eps, h + 2 * self.pad,
device=x.device, dtype=x.dtype)[:h]
arange = arange.unsqueeze(0).repeat(h, 1).unsqueeze(2)
base_grid = torch.cat([arange, arange.transpose(1, 0)], dim=2)
base_grid = base_grid.unsqueeze(0).repeat(n, 1, 1, 1)
shift = torch.randint(0, 2 * self.pad + 1, size=(n, 1, 1, 2),
device=x.device, dtype=x.dtype)
shift *= 2.0 / (h + 2 * self.pad)
grid = base_grid + shift
return F.grid_sample(x, grid, padding_mode='zeros', align_corners=False
)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'pad': 4}]
| import torch
from torch import device
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_cat_0(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 2
x1 = xindex // 2 % 4
x2 = xindex // 8
x4 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = x1
tmp6 = tmp5.to(tl.float32)
tmp7 = 6.0
tmp8 = tmp6 < tmp7
tmp9 = 0.16666666666666666
tmp10 = tmp6 * tmp9
tmp11 = -0.9166666666666666
tmp12 = tmp10 + tmp11
tmp13 = 11 + -1 * x1
tmp14 = tmp13.to(tl.float32)
tmp15 = tmp14 * tmp9
tmp16 = 0.9166666666666666
tmp17 = tmp16 - tmp15
tmp18 = tl.where(tmp8, tmp12, tmp17)
tmp19 = tl.full(tmp18.shape, 0.0, tmp18.dtype)
tmp20 = tl.where(tmp4, tmp18, tmp19)
tmp21 = tmp0 >= tmp3
tl.full([1], 2, tl.int64)
tmp24 = x2
tmp25 = tmp24.to(tl.float32)
tmp26 = tmp25 < tmp7
tmp27 = tmp25 * tmp9
tmp28 = tmp27 + tmp11
tmp29 = 11 + -1 * x2
tmp30 = tmp29.to(tl.float32)
tmp31 = tmp30 * tmp9
tmp32 = tmp16 - tmp31
tmp33 = tl.where(tmp26, tmp28, tmp32)
tmp34 = tl.full(tmp33.shape, 0.0, tmp33.dtype)
tmp35 = tl.where(tmp21, tmp33, tmp34)
tmp36 = tl.where(tmp4, tmp20, tmp35)
tl.store(out_ptr0 + x4, tmp36, xmask)
@triton.jit
def triton_poi_fused_grid_sampler_2d_replication_pad2d_1(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x2 = xindex // 64
x5 = xindex
x3 = xindex // 16
tmp0 = tl.load(in_ptr0 + 2 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 2 * x2, xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr0 + (1 + 2 * x0), xmask, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr1 + (1 + 2 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = 0.16666666666666666
tmp3 = tmp1 * tmp2
tmp4 = tmp0 + tmp3
tmp5 = 6.0
tmp6 = tmp4 * tmp5
tmp7 = 5.5
tmp8 = tmp6 + tmp7
tmp9 = libdevice.floor(tmp8)
tmp10 = 0.0
tmp11 = tmp9 >= tmp10
tmp12 = 12.0
tmp13 = tmp9 < tmp12
tmp16 = tmp15 * tmp2
tmp17 = tmp14 + tmp16
tmp18 = tmp17 * tmp5
tmp19 = tmp18 + tmp7
tmp20 = libdevice.floor(tmp19)
tmp21 = tmp20 >= tmp10
tmp22 = tmp20 < tmp12
tmp23 = tmp21 & tmp22
tmp24 = tmp13 & tmp23
tmp25 = tmp11 & tmp24
tmp26 = 1.0
tmp27 = tmp9 + tmp26
tmp28 = tmp27 - tmp8
tmp29 = tmp20 + tmp26
tmp30 = tmp29 - tmp19
tmp31 = tmp28 * tmp30
tmp32 = tl.where(tmp25, tmp31, tmp10)
tmp33 = tmp27 >= tmp10
tmp34 = tmp27 < tmp12
tmp35 = tmp34 & tmp23
tmp36 = tmp33 & tmp35
tmp37 = tmp8 - tmp9
tmp38 = tmp37 * tmp30
tmp39 = tl.where(tmp36, tmp38, tmp10)
tmp40 = tmp29 >= tmp10
tmp41 = tmp29 < tmp12
tmp42 = tmp40 & tmp41
tmp43 = tmp34 & tmp42
tmp44 = tmp33 & tmp43
tmp45 = tmp19 - tmp20
tmp46 = tmp37 * tmp45
tmp47 = tl.where(tmp44, tmp46, tmp10)
tmp48 = tmp13 & tmp42
tmp49 = tmp11 & tmp48
tmp50 = tmp28 * tmp45
tmp51 = tl.where(tmp49, tmp50, tmp10)
tmp52 = tmp20.to(tl.int64)
tmp53 = tl.full([1], 0, tl.int64)
tmp54 = tl.where(tmp25, tmp52, tmp53)
tmp55 = tl.full([XBLOCK], 12, tl.int32)
tmp56 = tmp54 + tmp55
tmp57 = tmp54 < 0
tmp58 = tl.where(tmp57, tmp56, tmp54)
tl.device_assert((0 <= tmp58) & (tmp58 < 12) | ~xmask,
'index out of bounds: 0 <= tmp58 < 12')
tmp60 = tmp9.to(tl.int64)
tmp61 = tl.where(tmp25, tmp60, tmp53)
tmp62 = tmp61 + tmp55
tmp63 = tmp61 < 0
tmp64 = tl.where(tmp63, tmp62, tmp61)
tl.device_assert((0 <= tmp64) & (tmp64 < 12) | ~xmask,
'index out of bounds: 0 <= tmp64 < 12')
tmp66 = tl.load(in_ptr2 + (4 * (3 * (3 <= 0 * (0 >= -4 + tmp58) + (-4 +
tmp58) * (-4 + tmp58 > 0)) + (0 * (0 >= -4 + tmp58) + (-4 + tmp58) *
(-4 + tmp58 > 0)) * (0 * (0 >= -4 + tmp58) + (-4 + tmp58) * (-4 +
tmp58 > 0) < 3)) + 16 * x3 + (3 * (3 <= 0 * (0 >= -4 + tmp64) + (-4 +
tmp64) * (-4 + tmp64 > 0)) + (0 * (0 >= -4 + tmp64) + (-4 + tmp64) *
(-4 + tmp64 > 0)) * (0 * (0 >= -4 + tmp64) + (-4 + tmp64) * (-4 +
tmp64 > 0) < 3))), xmask, eviction_policy='evict_last')
tmp67 = tl.where(tmp36, tmp52, tmp53)
tmp68 = tmp27.to(tl.int64)
tmp69 = tl.where(tmp36, tmp68, tmp53)
tmp70 = tmp29.to(tl.int64)
tmp71 = tl.where(tmp49, tmp70, tmp53)
tmp72 = tl.where(tmp49, tmp60, tmp53)
tmp73 = tmp66 * tmp32
tmp74 = tmp67 + tmp55
tmp75 = tmp67 < 0
tmp76 = tl.where(tmp75, tmp74, tmp67)
tl.device_assert((0 <= tmp76) & (tmp76 < 12) | ~xmask,
'index out of bounds: 0 <= tmp76 < 12')
tmp78 = tmp69 + tmp55
tmp79 = tmp69 < 0
tmp80 = tl.where(tmp79, tmp78, tmp69)
tl.device_assert((0 <= tmp80) & (tmp80 < 12) | ~xmask,
'index out of bounds: 0 <= tmp80 < 12')
tmp82 = tl.load(in_ptr2 + (4 * (3 * (3 <= 0 * (0 >= -4 + tmp76) + (-4 +
tmp76) * (-4 + tmp76 > 0)) + (0 * (0 >= -4 + tmp76) + (-4 + tmp76) *
(-4 + tmp76 > 0)) * (0 * (0 >= -4 + tmp76) + (-4 + tmp76) * (-4 +
tmp76 > 0) < 3)) + 16 * x3 + (3 * (3 <= 0 * (0 >= -4 + tmp80) + (-4 +
tmp80) * (-4 + tmp80 > 0)) + (0 * (0 >= -4 + tmp80) + (-4 + tmp80) *
(-4 + tmp80 > 0)) * (0 * (0 >= -4 + tmp80) + (-4 + tmp80) * (-4 +
tmp80 > 0) < 3))), xmask, eviction_policy='evict_last')
tmp83 = tmp82 * tmp39
tmp84 = tmp73 + tmp83
tmp85 = tmp71 + tmp55
tmp86 = tmp71 < 0
tmp87 = tl.where(tmp86, tmp85, tmp71)
tl.device_assert((0 <= tmp87) & (tmp87 < 12) | ~xmask,
'index out of bounds: 0 <= tmp87 < 12')
tmp89 = tmp72 + tmp55
tmp90 = tmp72 < 0
tmp91 = tl.where(tmp90, tmp89, tmp72)
tl.device_assert((0 <= tmp91) & (tmp91 < 12) | ~xmask,
'index out of bounds: 0 <= tmp91 < 12')
tmp93 = tl.load(in_ptr2 + (4 * (3 * (3 <= 0 * (0 >= -4 + tmp87) + (-4 +
tmp87) * (-4 + tmp87 > 0)) + (0 * (0 >= -4 + tmp87) + (-4 + tmp87) *
(-4 + tmp87 > 0)) * (0 * (0 >= -4 + tmp87) + (-4 + tmp87) * (-4 +
tmp87 > 0) < 3)) + 16 * x3 + (3 * (3 <= 0 * (0 >= -4 + tmp91) + (-4 +
tmp91) * (-4 + tmp91 > 0)) + (0 * (0 >= -4 + tmp91) + (-4 + tmp91) *
(-4 + tmp91 > 0)) * (0 * (0 >= -4 + tmp91) + (-4 + tmp91) * (-4 +
tmp91 > 0) < 3))), xmask, eviction_policy='evict_last')
tmp94 = tmp93 * tmp51
tmp95 = tmp84 + tmp94
tmp96 = tl.where(tmp44, tmp70, tmp53)
tmp97 = tl.where(tmp44, tmp68, tmp53)
tmp98 = tmp96 + tmp55
tmp99 = tmp96 < 0
tmp100 = tl.where(tmp99, tmp98, tmp96)
tl.device_assert((0 <= tmp100) & (tmp100 < 12) | ~xmask,
'index out of bounds: 0 <= tmp100 < 12')
tmp102 = tmp97 + tmp55
tmp103 = tmp97 < 0
tmp104 = tl.where(tmp103, tmp102, tmp97)
tl.device_assert((0 <= tmp104) & (tmp104 < 12) | ~xmask,
'index out of bounds: 0 <= tmp104 < 12')
tmp106 = tl.load(in_ptr2 + (4 * (3 * (3 <= 0 * (0 >= -4 + tmp100) + (-4 +
tmp100) * (-4 + tmp100 > 0)) + (0 * (0 >= -4 + tmp100) + (-4 +
tmp100) * (-4 + tmp100 > 0)) * (0 * (0 >= -4 + tmp100) + (-4 +
tmp100) * (-4 + tmp100 > 0) < 3)) + 16 * x3 + (3 * (3 <= 0 * (0 >=
-4 + tmp104) + (-4 + tmp104) * (-4 + tmp104 > 0)) + (0 * (0 >= -4 +
tmp104) + (-4 + tmp104) * (-4 + tmp104 > 0)) * (0 * (0 >= -4 +
tmp104) + (-4 + tmp104) * (-4 + tmp104 > 0) < 3))), xmask,
eviction_policy='evict_last')
tmp107 = tmp106 * tmp47
tmp108 = tmp95 + tmp107
tl.store(in_out_ptr0 + x5, tmp108, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(32)](buf0, 32, XBLOCK=32, num_warps=1,
num_stages=1)
buf1 = torch.ops.aten.randint.low(0, 9, [4, 1, 1, 2], dtype=torch.
float32, device=device(type='cuda', index=0), pin_memory=False)
buf2 = buf1
del buf1
buf10 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf11 = buf10
del buf10
buf15 = buf11
del buf11
triton_poi_fused_grid_sampler_2d_replication_pad2d_1[grid(256)](buf15,
buf0, buf2, arg0_1, 256, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
del buf0
del buf2
return buf15,
class RandomShiftsAugNew(nn.Module):
def __init__(self, pad):
super().__init__()
self.pad = pad
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| MishaLaskin/url_benchmark | RandomShiftsAug | false | 5,614 | [
"MIT"
] | 1 | a81aed0a0aec3a7dad83d930e54d480f97cf535d | https://github.com/MishaLaskin/url_benchmark/tree/a81aed0a0aec3a7dad83d930e54d480f97cf535d | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, pad):
super().__init__()
self.pad = pad
def forward(self, x):
x = x.float()
n, _c, h, w = x.size()
assert h == w
padding = tuple([self.pad] * 4)
x = F.pad(x, padding, 'replicate')
eps = 1.0 / (h + 2 * self.pad)
arange = torch.linspace(-1.0 + eps, 1.0 - eps, h + 2 * self.pad,
device=x.device, dtype=x.dtype)[:h]
arange = arange.unsqueeze(0).repeat(h, 1).unsqueeze(2)
base_grid = torch.cat([arange, arange.transpose(1, 0)], dim=2)
base_grid = base_grid.unsqueeze(0).repeat(n, 1, 1, 1)
shift = torch.randint(0, 2 * self.pad + 1, size=(n, 1, 1, 2),
device=x.device, dtype=x.dtype)
shift *= 2.0 / (h + 2 * self.pad)
grid = base_grid + shift
return F.grid_sample(x, grid, padding_mode='zeros', align_corners=False
)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
FeedForwardNetwork | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/v7/cv7humnywkkqhrumbeetegqlkretdwtkj5pcanrbgxrolupvobzt.py
# Topologically Sorted Source Nodes: [mul, pow_1, mul_1, add, mul_2, tanh, add_1, hid], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# hid => mul_3
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# pow_1 => pow_1
# tanh => tanh
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, 0.5), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_1, 3), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_1, 0.044715), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %mul_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 0.7978845608028654), kwargs = {})
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%mul_2,), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%tanh, 1), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %add_1), kwargs = {})
triton_poi_fused_add_mul_pow_tanh_0 = async_compile.triton('triton_poi_fused_add_mul_pow_tanh_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_pow_tanh_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = tmp0 * tmp0
tmp4 = tmp3 * tmp0
tmp5 = 0.044715
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tmp8 = 0.7978845608028654
tmp9 = tmp7 * tmp8
tmp10 = libdevice.tanh(tmp9)
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp2 * tmp12
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul, pow_1, mul_1, add, mul_2, tanh, add_1, hid], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mul_pow_tanh_0.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del primals_5
return (reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), primals_4, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3, primals_4, primals_5])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import math
import torch
import torch.nn as nn
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class FeedForwardNetwork(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, inputs):
hid = self.gleu(self.lin1(inputs))
return self.lin2(self.dropout(hid))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_dim': 4, 'hid_dim': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = tmp0 * tmp0
tmp4 = tmp3 * tmp0
tmp5 = 0.044715
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tmp8 = 0.7978845608028654
tmp9 = tmp7 * tmp8
tmp10 = libdevice.tanh(tmp9)
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp2 * tmp12
tl.store(out_ptr0 + x0, tmp13, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_pow_tanh_0[grid(256)](buf0, buf1, 256,
XBLOCK=256, num_warps=4, num_stages=1)
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del primals_5
return reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf0, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), primals_4
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class FeedForwardNetworkNew(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, input_0):
primals_1 = self.lin1.weight
primals_2 = self.lin1.bias
primals_4 = self.lin2.weight
primals_5 = self.lin2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| MobtgZhang/MWMLNet | FeedForwardNetwork | false | 5,615 | [
"MIT"
] | 1 | 125bb39935916b6b4be505c51cb6a04eb49b96d0 | https://github.com/MobtgZhang/MWMLNet/tree/125bb39935916b6b4be505c51cb6a04eb49b96d0 | import math
import torch
import torch.nn as nn
class GELU(nn.Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class Model(nn.Module):
def __init__(self, in_dim, hid_dim) ->None:
super().__init__()
self.lin1 = nn.Linear(in_dim, hid_dim)
self.lin2 = nn.Linear(hid_dim, in_dim)
self.gleu = GELU()
self.dropout = nn.Dropout()
def forward(self, inputs):
hid = self.gleu(self.lin1(inputs))
return self.lin2(self.dropout(hid))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
BERTNextSentence | # AOT ID: ['0_forward']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/yy/cyya3js6wt64vdji3sfisvrqyfvqxwkwqq5mzg5bqjl2crzjs4t3.py
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# linear => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%select,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_0 = async_compile.triton('triton_poi_fused_clone_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[64],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_clone_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = (xindex // 16)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (64*x1)), xmask)
tl.store(out_ptr0 + (x2), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/t6/ct6t4t4f6df2rswen66qmskrag4dcnuaoxkebk4pssna63yxl3v3.py
# Topologically Sorted Source Nodes: [linear, log_softmax], Original ATen: [aten.add, aten._log_softmax]
# Source node to ATen node mapping:
# linear => add
# log_softmax => amax, sub
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_3), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add, [-1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %amax), kwargs = {})
triton_poi_fused__log_softmax_add_1 = async_compile.triton('triton_poi_fused__log_softmax_add_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__log_softmax_add_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__log_softmax_add_1(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 2
x1 = (xindex // 2)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2*x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (0))
tmp5 = tl.broadcast_to(tmp4, [XBLOCK])
tmp7 = tl.load(in_ptr0 + (1 + (2*x1)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (1))
tmp9 = tl.broadcast_to(tmp8, [XBLOCK])
tmp2 = tmp0 + tmp1
tmp6 = tmp3 + tmp5
tmp10 = tmp7 + tmp9
tmp11 = triton_helpers.maximum(tmp6, tmp10)
tmp12 = tmp2 - tmp11
tl.store(out_ptr0 + (x2), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/jt/cjtlgs566wh4nqwf6wvcsracaovzta35wbm5vllgfuavp5ebkiq4.py
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# log_softmax => exp, log, sub_1, sum_1
# Graph fragment:
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
# %log : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%sum_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%sub, %log), kwargs = {})
triton_poi_fused__log_softmax_2 = async_compile.triton('triton_poi_fused__log_softmax_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__log_softmax_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__log_softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 2)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (2*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (2*x1)), xmask, eviction_policy='evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp6 = tl_math.log(tmp5)
tmp7 = tmp0 - tmp6
tl.store(out_ptr0 + (x2), tmp7, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (2, 4), (4, 1))
assert_size_stride(primals_3, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(primals_1, buf0, 64, grid=grid(64), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf0, (16, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 2), (1, 4), 0), out=buf1)
del primals_2
buf2 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear, log_softmax], Original ATen: [aten.add, aten._log_softmax]
triton_poi_fused__log_softmax_add_1.run(buf1, primals_3, buf2, 32, grid=grid(32), stream=stream0)
del primals_3
buf3 = reinterpret_tensor(buf1, (4, 4, 2), (8, 2, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
triton_poi_fused__log_softmax_2.run(buf2, buf3, 32, grid=grid(32), stream=stream0)
del buf2
return (buf3, reinterpret_tensor(buf0, (16, 4), (4, 1), 0), buf3, )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
primals_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((2, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
fn = lambda: call([primals_1, primals_2, primals_3])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| import torch
import torch.nn as nn
class BERTNextSentence(nn.Module):
def __init__(self, hidden):
super().__init__()
self.linear = nn.Linear(hidden, 2)
self.softmax = nn.LogSoftmax(dim=-1)
def forward(self, x):
return self.softmax(self.linear(x[:, 0]))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'hidden': 4}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = xindex // 16
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 64 * x1), xmask)
tl.store(out_ptr0 + x2, tmp0, xmask)
@triton.jit
def triton_poi_fused__log_softmax_add_1(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 2
x1 = xindex // 2
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + 2 * x1, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + 0)
tmp5 = tl.broadcast_to(tmp4, [XBLOCK])
tmp7 = tl.load(in_ptr0 + (1 + 2 * x1), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + 1)
tmp9 = tl.broadcast_to(tmp8, [XBLOCK])
tmp2 = tmp0 + tmp1
tmp6 = tmp3 + tmp5
tmp10 = tmp7 + tmp9
tmp11 = triton_helpers.maximum(tmp6, tmp10)
tmp12 = tmp2 - tmp11
tl.store(out_ptr0 + x2, tmp12, xmask)
@triton.jit
def triton_poi_fused__log_softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 2
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 2 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 2 * x1), xmask, eviction_policy='evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp6 = tl_math.log(tmp5)
tmp7 = tmp0 - tmp6
tl.store(out_ptr0 + x2, tmp7, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (2, 4), (4, 1))
assert_size_stride(primals_3, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(64)](primals_1, buf0, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 2), (1, 4), 0), out=buf1)
del primals_2
buf2 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.float32)
triton_poi_fused__log_softmax_add_1[grid(32)](buf1, primals_3, buf2,
32, XBLOCK=32, num_warps=1, num_stages=1)
del primals_3
buf3 = reinterpret_tensor(buf1, (4, 4, 2), (8, 2, 1), 0)
del buf1
triton_poi_fused__log_softmax_2[grid(32)](buf2, buf3, 32, XBLOCK=32,
num_warps=1, num_stages=1)
del buf2
return buf3, reinterpret_tensor(buf0, (16, 4), (4, 1), 0), buf3
class BERTNextSentenceNew(nn.Module):
def __init__(self, hidden):
super().__init__()
self.linear = nn.Linear(hidden, 2)
self.softmax = nn.LogSoftmax(dim=-1)
def forward(self, input_0):
primals_2 = self.linear.weight
primals_3 = self.linear.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| Moymix/BERT-pytorch | BERTNextSentence | false | 5,616 | [
"Apache-2.0"
] | 1 | f0b9c3ae53e05c00adcc761e0422e4222d8b5619 | https://github.com/Moymix/BERT-pytorch/tree/f0b9c3ae53e05c00adcc761e0422e4222d8b5619 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, hidden):
super().__init__()
self.linear = nn.Linear(hidden, 2)
self.softmax = nn.LogSoftmax(dim=-1)
def forward(self, x):
return self.softmax(self.linear(x[:, 0]))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
Upsample2d | # AOT ID: ['0_inference']
from ctypes import c_void_p, c_long, c_int
import torch
import math
import random
import os
import tempfile
from math import inf, nan
from torch._inductor.hooks import run_intermediate_hooks
from torch._inductor.utils import maybe_profile
from torch._inductor.codegen.memory_planning import _align as align
from torch import device, empty_strided
from torch._inductor.async_compile import AsyncCompile
from torch._inductor.select_algorithm import extern_kernels
from torch._inductor.codegen.multi_kernel import MultiKernelCall
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid, split_scan_grid, grid_combo_kernels, start_graph, end_graph
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
aten = torch.ops.aten
inductor_ops = torch.ops.inductor
_quantized = torch.ops._quantized
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cpu = torch._C._dynamo.guards._empty_strided_cpu
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
empty_strided_xpu = torch._C._dynamo.guards._empty_strided_xpu
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
alloc_from_pool = torch.ops.inductor._alloc_from_pool
async_compile = AsyncCompile()
# kernel path: runs/run_shard_4/inductor_cache/s5/cs53ve6ff5smmuu5zxn6uwnvgsi32uuuvewmis3iy7oejdcvopvs.py
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.constant_pad_nd]
# Source node to ATen node mapping:
# y_5 => constant_pad_nd_1
# Graph fragment:
# %constant_pad_nd_1 : [num_users=1] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%view_2, [0, 0, 1, 0, 1, 0, 0, 0], 0.0), kwargs = {})
triton_poi_fused_constant_pad_nd_0 = async_compile.triton('triton_poi_fused_constant_pad_nd_0', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_constant_pad_nd_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = (xindex // 72) % 2
x1 = (xindex // 8) % 9
x0 = xindex % 8
x3 = (xindex // 144)
x7 = xindex
tmp0 = (-1) + x2
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = (-1) + x1
tmp4 = tmp3 >= tmp1
tmp5 = tmp2 & tmp4
tmp6 = (-1) + (((-1) + x1) % 2)
tmp7 = tmp6 >= tmp1
tmp8 = (-1) + (x0 % 2)
tmp9 = tmp8 >= tmp1
tmp10 = tmp7 & tmp9
tmp11 = tmp10 & tmp5
tmp12 = tl.load(in_ptr0 + ((4*((((-1) + x1) // 2) % 4)) + (16*x3) + (x0 // 2)), tmp11 & xmask, eviction_policy='evict_last', other=0.0)
tmp13 = tl.full(tmp12.shape, 0.0, tmp12.dtype)
tmp14 = tl.where(tmp5, tmp12, tmp13)
tl.store(out_ptr0 + (x7), tmp14, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/qb/cqbibvks2cjhauc3nur2g3ih7ql3ix5qhuh7srxpiit2lwdgiiiw.py
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# y_9 => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%view_3, %arg1_1, None, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_1 = async_compile.triton('triton_poi_fused_convolution_1', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[128, 8], tile_hint=TileHint.SQUARE,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32', 3: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused_convolution_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 128
xnumel = 8
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 8
y1 = (yindex // 8)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (8*x2) + (144*y1)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + (8*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_4/inductor_cache/zh/czh33meqtkkfeqw325p6bix3tuaq4yza5hjchatomlwpg3z2e2yu.py
# Topologically Sorted Source Nodes: [y_11], Original ATen: [aten._unsafe_index]
# Source node to ATen node mapping:
# y_11 => _unsafe_index
# Graph fragment:
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%view_4, [None, None, %unsqueeze, %convert_element_type_3]), kwargs = {})
triton_poi_fused__unsafe_index_2 = async_compile.triton('triton_poi_fused__unsafe_index_2', '''
import triton
import triton.language as tl
from triton.compiler.compiler import AttrsDescriptor
from torch._inductor.runtime import triton_helpers, triton_heuristics
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch._inductor.runtime.hints import AutotuneHint, ReductionHint, TileHint, instance_descriptor, DeviceProperties
@triton_heuristics.pointwise(
size_hints=[1024],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: 'i32'}, 'device': DeviceProperties(type='cuda', index=0, cc=80, major=8, regs_per_multiprocessor=65536, max_threads_per_multi_processor=2048, multi_processor_count=108), 'constants': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__unsafe_index_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 0, 'num_reduction': 0, 'backend_hash': 'A9C866B4A14FD3277824029365D703C2427B2E685E54EC9B3EF4ADC8D1EEAC1D', 'are_deterministic_algorithms_enabled': False, 'assert_indirect_indexing': True, 'autotune_local_cache': True, 'autotune_pointwise': True, 'autotune_remote_cache': None, 'force_disable_caches': False, 'dynamic_scale_rblock': True, 'max_autotune': False, 'max_autotune_pointwise': False, 'min_split_scan_rblock': 256, 'spill_threshold': 16, 'store_cubin': False},
min_elem_per_thread=0
)
@triton.jit
def triton_poi_fused__unsafe_index_2(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 8) % 8
x0 = xindex % 8
x2 = (xindex // 64)
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.625
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + (5*tmp4) + (25*x2)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x4), tmp9, xmask)
''', device_str='cuda')
async_compile.wait(globals())
del async_compile
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (1, 1, 4, 4), (16, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 2, 9, 8), (144, 72, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [y_5], Original ATen: [aten.constant_pad_nd]
stream0 = get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0.run(arg0_1, buf0, 2304, grid=grid(2304), stream=stream0)
del arg0_1
buf1 = empty_strided_cuda((16, 1, 8, 8), (64, 1, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.convolution]
triton_poi_fused_convolution_1.run(buf0, buf1, 128, 8, grid=grid(128, 8), stream=stream0)
del buf0
# Topologically Sorted Source Nodes: [y_9], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, arg1_1, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (16, 1, 5, 5), (25, 1, 5, 1))
del arg1_1
buf3 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [y_11], Original ATen: [aten._unsafe_index]
triton_poi_fused__unsafe_index_2.run(buf2, buf3, 1024, grid=grid(1024), stream=stream0)
del buf2
return (reinterpret_tensor(buf3, (4, 4, 8, 8), (256, 64, 8, 1), 0), )
def benchmark_compiled_module(times=10, repeat=10):
from torch._dynamo.testing import rand_strided
from torch._inductor.utils import print_performance
arg0_1 = rand_strided((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((1, 1, 4, 4), (16, 16, 4, 1), device='cuda:0', dtype=torch.float32)
fn = lambda: call([arg0_1, arg1_1])
return print_performance(fn, times=times, repeat=repeat)
if __name__ == "__main__":
from torch._inductor.wrapper_benchmark import compiled_module_main
compiled_module_main('None', benchmark_compiled_module)
| from _paritybench_helpers import _mock_config
import torch
import numpy as np
import torch.nn as nn
import torch.nn.functional as F
def _setup_kernel(k):
k = np.asarray(k, dtype=np.float32)
if k.ndim == 1:
k = np.outer(k, k)
k /= np.sum(k)
assert k.ndim == 2
assert k.shape[0] == k.shape[1]
return k
class Upsample2d(nn.Module):
def __init__(self, opts, k=[1, 3, 3, 1], factor=2, down=1, gain=1):
"""
Upsample2d method in G_synthesis_stylegan2.
:param k: FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
The default is `[1] * factor`, which corresponds to average pooling.
:param factor: Integer downsampling factor (default: 2).
:param gain: Scaling factor for signal magnitude (default: 1.0).
Returns: Tensor of the shape `[N, C, H // factor, W // factor]`
"""
super().__init__()
assert isinstance(factor, int
) and factor >= 1, 'factor must be larger than 1! (default: 2)'
self.gain = gain
self.factor = factor
self.opts = opts
self.k = _setup_kernel(k) * (self.gain * factor ** 2)
self.k = torch.FloatTensor(self.k).unsqueeze(0).unsqueeze(0)
self.k = nn.Parameter(self.k, requires_grad=False)
self.p = self.k.shape[0] - self.factor
self.padx0, self.pady0 = (self.p + 1) // 2 + factor - 1, (self.p + 1
) // 2 + factor - 1
self.padx1, self.pady1 = self.p // 2, self.p // 2
self.kernelH, self.kernelW = self.k.shape[2:]
self.down = down
def forward(self, x):
y = x.clone()
y = y.reshape([-1, x.shape[2], x.shape[3], 1])
inC, inH, inW = x.shape[1:]
y = torch.reshape(y, (-1, inH, 1, inW, 1, 1))
y = F.pad(y, (0, 0, self.factor - 1, 0, 0, 0, self.factor - 1, 0, 0,
0, 0, 0))
y = torch.reshape(y, (-1, 1, inH * self.factor, inW * self.factor))
y = F.pad(y, (0, 0, max(self.pady0, 0), max(self.pady1, 0), max(
self.padx0, 0), max(self.padx1, 0), 0, 0))
y = y[:, max(-self.pady0, 0):y.shape[1] - max(-self.pady1, 0), max(
-self.padx0, 0):y.shape[2] - max(-self.padx1, 0), :]
y = y.permute(0, 3, 1, 2)
y = y.reshape(-1, 1, inH * self.factor + self.pady0 + self.pady1,
inW * self.factor + self.padx0 + self.padx1)
y = F.conv2d(y, self.k)
y = y.view(-1, 1, inH * self.factor + self.pady0 + self.pady1 -
self.kernelH + 1, inW * self.factor + self.padx0 + self.padx1 -
self.kernelW + 1)
if inH * self.factor != y.shape[1]:
y = F.interpolate(y, size=(inH * self.factor, inW * self.factor))
y = y.permute(0, 2, 3, 1)
y = y.reshape(-1, inC, inH * self.factor, inW * self.factor)
return y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'opts': _mock_config()}]
| import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex // 72 % 2
x1 = xindex // 8 % 9
x0 = xindex % 8
x3 = xindex // 144
x7 = xindex
tmp0 = -1 + x2
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = -1 + x1
tmp4 = tmp3 >= tmp1
tmp5 = tmp2 & tmp4
tmp6 = -1 + (-1 + x1) % 2
tmp7 = tmp6 >= tmp1
tmp8 = -1 + x0 % 2
tmp9 = tmp8 >= tmp1
tmp10 = tmp7 & tmp9
tmp11 = tmp10 & tmp5
tmp12 = tl.load(in_ptr0 + (4 * ((-1 + x1) // 2 % 4) + 16 * x3 + x0 // 2
), tmp11 & xmask, eviction_policy='evict_last', other=0.0)
tmp13 = tl.full(tmp12.shape, 0.0, tmp12.dtype)
tmp14 = tl.where(tmp5, tmp12, tmp13)
tl.store(out_ptr0 + x7, tmp14, xmask)
@triton.jit
def triton_poi_fused_convolution_1(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 128
xnumel = 8
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 8
y1 = yindex // 8
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 8 * x2 + 144 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 8 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused__unsafe_index_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 8 % 8
x0 = xindex % 8
x2 = xindex // 64
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.625
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tmp5 = x0
tmp6 = tmp5.to(tl.float32)
tmp7 = tmp6 * tmp2
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (tmp8 + 5 * tmp4 + 25 * x2), xmask,
eviction_policy='evict_last')
tl.store(out_ptr0 + x4, tmp9, xmask)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (1, 1, 4, 4), (16, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 2, 9, 8), (144, 72, 8, 1), torch.float32
)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(2304)](arg0_1, buf0, 2304,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((16, 1, 8, 8), (64, 1, 8, 1), torch.float32)
triton_poi_fused_convolution_1[grid(128, 8)](buf0, buf1, 128, 8,
XBLOCK=8, YBLOCK=128, num_warps=4, num_stages=1)
del buf0
buf2 = extern_kernels.convolution(buf1, arg1_1, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (16, 1, 5, 5), (25, 1, 5, 1))
del arg1_1
buf3 = buf1
del buf1
triton_poi_fused__unsafe_index_2[grid(1024)](buf2, buf3, 1024,
XBLOCK=256, num_warps=4, num_stages=1)
del buf2
return reinterpret_tensor(buf3, (4, 4, 8, 8), (256, 64, 8, 1), 0),
def _setup_kernel(k):
k = np.asarray(k, dtype=np.float32)
if k.ndim == 1:
k = np.outer(k, k)
k /= np.sum(k)
assert k.ndim == 2
assert k.shape[0] == k.shape[1]
return k
class Upsample2dNew(nn.Module):
def __init__(self, opts, k=[1, 3, 3, 1], factor=2, down=1, gain=1):
"""
Upsample2d method in G_synthesis_stylegan2.
:param k: FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
The default is `[1] * factor`, which corresponds to average pooling.
:param factor: Integer downsampling factor (default: 2).
:param gain: Scaling factor for signal magnitude (default: 1.0).
Returns: Tensor of the shape `[N, C, H // factor, W // factor]`
"""
super().__init__()
assert isinstance(factor, int
) and factor >= 1, 'factor must be larger than 1! (default: 2)'
self.gain = gain
self.factor = factor
self.opts = opts
self.k = _setup_kernel(k) * (self.gain * factor ** 2)
self.k = torch.FloatTensor(self.k).unsqueeze(0).unsqueeze(0)
self.k = nn.Parameter(self.k, requires_grad=False)
self.p = self.k.shape[0] - self.factor
self.padx0, self.pady0 = (self.p + 1) // 2 + factor - 1, (self.p + 1
) // 2 + factor - 1
self.padx1, self.pady1 = self.p // 2, self.p // 2
self.kernelH, self.kernelW = self.k.shape[2:]
self.down = down
def forward(self, input_0):
arg1_1 = self.k
arg0_1 = input_0
output = call([arg0_1, arg1_1])
return output[0]
| Iceland-Leo/StyleGAN2_PyTorch | Upsample2d | false | 5,617 | [
"MIT"
] | 1 | 3621f5e4ba1c7fde7e2fae1f4700d050656a0b02 | https://github.com/Iceland-Leo/StyleGAN2_PyTorch/tree/3621f5e4ba1c7fde7e2fae1f4700d050656a0b02 | from _paritybench_helpers import _mock_config
import torch
import numpy as np
import torch.nn as nn
import torch.nn.functional as F
def _setup_kernel(k):
k = np.asarray(k, dtype=np.float32)
if k.ndim == 1:
k = np.outer(k, k)
k /= np.sum(k)
assert k.ndim == 2
assert k.shape[0] == k.shape[1]
return k
class Model(nn.Module):
def __init__(self, opts, k=[1, 3, 3, 1], factor=2, down=1, gain=1):
"""
Upsample2d method in G_synthesis_stylegan2.
:param k: FIR filter of the shape `[firH, firW]` or `[firN]` (separable).
The default is `[1] * factor`, which corresponds to average pooling.
:param factor: Integer downsampling factor (default: 2).
:param gain: Scaling factor for signal magnitude (default: 1.0).
Returns: Tensor of the shape `[N, C, H // factor, W // factor]`
"""
super().__init__()
assert isinstance(factor, int
) and factor >= 1, 'factor must be larger than 1! (default: 2)'
self.gain = gain
self.factor = factor
self.opts = opts
self.k = _setup_kernel(k) * (self.gain * factor ** 2)
self.k = torch.FloatTensor(self.k).unsqueeze(0).unsqueeze(0)
self.k = nn.Parameter(self.k, requires_grad=False)
self.p = self.k.shape[0] - self.factor
self.padx0, self.pady0 = (self.p + 1) // 2 + factor - 1, (self.p + 1
) // 2 + factor - 1
self.padx1, self.pady1 = self.p // 2, self.p // 2
self.kernelH, self.kernelW = self.k.shape[2:]
self.down = down
def forward(self, x):
y = x.clone()
y = y.reshape([-1, x.shape[2], x.shape[3], 1])
inC, inH, inW = x.shape[1:]
y = torch.reshape(y, (-1, inH, 1, inW, 1, 1))
y = F.pad(y, (0, 0, self.factor - 1, 0, 0, 0, self.factor - 1, 0, 0,
0, 0, 0))
y = torch.reshape(y, (-1, 1, inH * self.factor, inW * self.factor))
y = F.pad(y, (0, 0, max(self.pady0, 0), max(self.pady1, 0), max(
self.padx0, 0), max(self.padx1, 0), 0, 0))
y = y[:, max(-self.pady0, 0):y.shape[1] - max(-self.pady1, 0), max(
-self.padx0, 0):y.shape[2] - max(-self.padx1, 0), :]
y = y.permute(0, 3, 1, 2)
y = y.reshape(-1, 1, inH * self.factor + self.pady0 + self.pady1,
inW * self.factor + self.padx0 + self.padx1)
y = F.conv2d(y, self.k)
y = y.view(-1, 1, inH * self.factor + self.pady0 + self.pady1 -
self.kernelH + 1, inW * self.factor + self.padx0 + self.padx1 -
self.kernelW + 1)
if inH * self.factor != y.shape[1]:
y = F.interpolate(y, size=(inH * self.factor, inW * self.factor))
y = y.permute(0, 2, 3, 1)
y = y.reshape(-1, inC, inH * self.factor, inW * self.factor)
return y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.