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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
FCDiscriminator_low | # 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_7/inductor_cache/xd/cxdqslrdqajmcxikxhvxi7lkzd2yepfzcwkkltrpstapeq35h632.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=[256, 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 = 256
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_7/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_7/inductor_cache/ra/crarmf7s2qf36jg27hprl42qtwcxcnnoyrgzgevtstzj4qgsdzwl.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=[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_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 = 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_7/inductor_cache/yd/cydvmxsmzwizyj5fbgjnjeeo27as6zdlft5s5uj57ovvcxtlbfhh.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=[32768, 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 = 32768
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 % 128
y1 = (yindex // 128)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (128*x2) + (1152*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/nm/cnmd2w5kksypgz2xxaozdqzofis5qg3ogbaepglpjzcvgi3ahraq.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
# 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, [2, 2], [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=[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_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 = 1024
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')
# kernel path: runs/run_shard_7/inductor_cache/3b/c3bly7xeq5u7vks53rlhytqc2grxfdsyjh6dg3tzxdfhvbyu75vn.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# x_2 => convolution_1
# x_3 => 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], [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=[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_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 = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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')
# kernel path: runs/run_shard_7/inductor_cache/kr/ckrovzxohsppjdjbwlm7vydnedcmhlptuiex3ipurrmeka35tzjh.py
# Topologically Sorted Source Nodes: [x_4, x_5], Original ATen: [aten.convolution, aten.leaky_relu]
# Source node to ATen node mapping:
# x_4 => convolution_2
# x_5 => gt, mul, where
# Graph fragment:
# %convolution_2 : [num_users=3] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_1, %primals_6, %primals_7, [2, 2], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_2, 0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_2, 0.2), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %convolution_2, %mul), kwargs = {})
triton_poi_fused_convolution_leaky_relu_6 = async_compile.triton('triton_poi_fused_convolution_leaky_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=[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_leaky_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_leaky_relu_6(in_out_ptr0, in_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 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.2
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(in_out_ptr0 + (x2), tmp7, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/5w/c5w67nvirr37nwzbi4mdydtra4kmtblibc6vuklnxpyr5bk4tipo.py
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_6 => convolution_3
# Graph fragment:
# %convolution_3 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%where, %primals_8, %primals_9, [1, 1], [0, 0], [1, 1], False, [0, 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=[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_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 = 4
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 = tl.load(in_ptr0 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tl.store(in_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, primals_6, primals_7, primals_8, primals_9 = args
args.clear()
assert_size_stride(primals_1, (64, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (64, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (128, ), (1, ))
assert_size_stride(primals_6, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (256, ), (1, ))
assert_size_stride(primals_8, (1, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_9, (1, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 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, 256, 9, grid=grid(256, 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((128, 64, 3, 3), (576, 1, 192, 64), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(primals_4, buf2, 8192, 9, grid=grid(8192, 9), stream=stream0)
del primals_4
buf3 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(primals_6, buf3, 32768, 9, grid=grid(32768, 9), stream=stream0)
del primals_6
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf4 = extern_kernels.convolution(buf1, buf0, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 2, 2), (256, 1, 128, 64))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_4.run(buf5, primals_2, 1024, grid=grid(1024), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf6 = extern_kernels.convolution(buf5, buf2, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 128, 1, 1), (128, 1, 128, 128))
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_5.run(buf7, primals_5, 512, grid=grid(512), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf7, buf3, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 256, 1, 1), (256, 1, 256, 256))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [x_4, x_5], Original ATen: [aten.convolution, aten.leaky_relu]
triton_poi_fused_convolution_leaky_relu_6.run(buf9, primals_7, 1024, grid=grid(1024), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [x_6], 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, 1, 1, 1), (1, 1, 1, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.convolution]
triton_poi_fused_convolution_7.run(buf11, primals_9, 4, grid=grid(4), stream=stream0)
del primals_9
return (buf11, buf0, buf1, buf2, buf3, primals_8, buf5, buf7, 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((64, 4, 3, 3), (36, 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, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((128, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((256, 128, 3, 3), (1152, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((1, 256, 1, 1), (256, 1, 1, 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
class FCDiscriminator_low(nn.Module):
def __init__(self, inplanes, planes=64):
super(FCDiscriminator_low, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 4, 1, kernel_size=1)
def forward(self, x):
x = self.conv1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.relu(x)
x = self.conv3(x)
x = self.leaky_relu(x)
x = self.classifier(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'inplanes': 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 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_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 256
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 % 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_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 % 128
y1 = yindex // 128
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 128 * x2 + 1152 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_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 % 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)
@triton.jit
def triton_poi_fused_convolution_relu_5(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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)
@triton.jit
def triton_poi_fused_convolution_leaky_relu_6(in_out_ptr0, in_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 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.2
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(in_out_ptr0 + x2, tmp7, xmask)
@triton.jit
def triton_poi_fused_convolution_7(in_out_ptr0, in_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_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tl.store(in_out_ptr0 + x0, tmp3, 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, (64, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (64,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (128,), (1,))
assert_size_stride(primals_6, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (256,), (1,))
assert_size_stride(primals_8, (1, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_9, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4, 3, 3), (36, 1, 12, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(256, 9)](primals_1, buf0, 256, 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((128, 64, 3, 3), (576, 1, 192, 64), torch
.float32)
triton_poi_fused_2[grid(8192, 9)](primals_4, buf2, 8192, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(32768, 9)](primals_6, buf3, 32768, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_6
buf4 = extern_kernels.convolution(buf1, buf0, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 2, 2), (256, 1, 128, 64))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_4[grid(1024)](buf5, primals_2,
1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf6 = extern_kernels.convolution(buf5, buf2, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 128, 1, 1), (128, 1, 128, 128))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_5[grid(512)](buf7, primals_5, 512,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf8 = extern_kernels.convolution(buf7, buf3, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 256, 1, 1), (256, 1, 256, 256))
buf9 = buf8
del buf8
triton_poi_fused_convolution_leaky_relu_6[grid(1024)](buf9,
primals_7, 1024, 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, 1, 1, 1), (1, 1, 1, 1))
buf11 = buf10
del buf10
triton_poi_fused_convolution_7[grid(4)](buf11, primals_9, 4, XBLOCK
=4, num_warps=1, num_stages=1)
del primals_9
return buf11, buf0, buf1, buf2, buf3, primals_8, buf5, buf7, buf9
class FCDiscriminator_lowNew(nn.Module):
def __init__(self, inplanes, planes=64):
super(FCDiscriminator_lowNew, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 4, 1, kernel_size=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.classifier.weight
primals_9 = self.classifier.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]
| seabearlmx/PA-DAN | FCDiscriminator_low | false | 4,298 | [
"MIT"
] | 0 | bdd1200396d102e68acdd265db9d22ddb83b6404 | https://github.com/seabearlmx/PA-DAN/tree/bdd1200396d102e68acdd265db9d22ddb83b6404 | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, inplanes, planes=64):
super().__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 4, 1, kernel_size=1)
def forward(self, x):
x = self.conv1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.relu(x)
x = self.conv3(x)
x = self.leaky_relu(x)
x = self.classifier(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
ParallelPolarizedSelfAttention | # 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_7/inductor_cache/u3/cu3litezfpnwhpnfnfuj6dtimz6ml42wmcwnwxlnovd4p5lvyin4.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=[2048, 4096], 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_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 = 2048
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = (yindex // 512)
tmp0 = tl.load(in_ptr0 + (x2 + (4096*y3)), None, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (512*x2) + (2097152*y1)), tmp0, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/iq/ciqt7v5dvn4ukgyyy6maxezu54tu2k6htitfxylgvyziqck3gwix.py
# Topologically Sorted Source Nodes: [channel_wq_2], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# channel_wq_2 => amax, div, exp, sub, sum_1
# 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 = {})
# %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_red_fused__softmax_1 = async_compile.triton('triton_red_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.reduction(
size_hints=[4, 4096],
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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, '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_red_fused__softmax_1(in_ptr0, in_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 4
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
_tmp5 = tl.full([XBLOCK, RBLOCK], float("-inf"), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_last', other=0.0)
tmp3 = tmp0 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = triton_helpers.maximum(_tmp5, tmp4)
_tmp5 = tl.where(rmask & xmask, tmp6, _tmp5)
tmp5 = triton_helpers.max2(_tmp5, 1)[:, None]
tmp8 = tl.load(in_ptr1 + (0))
tmp9 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
_tmp14 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp7 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tmp7 + tmp9
tmp11 = tmp10 - tmp5
tmp12 = tl_math.exp(tmp11)
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = _tmp14 + tmp13
_tmp14 = tl.where(rmask & xmask, tmp15, _tmp14)
tmp14 = tl.sum(_tmp14, 1)[:, None]
tmp17 = tl.load(in_ptr1 + (0))
tmp18 = tl.broadcast_to(tmp17, [XBLOCK, RBLOCK])
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp16 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp19 = tmp16 + tmp18
tmp20 = tmp19 - tmp5
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp21 / tmp14
tl.store(out_ptr2 + (r1 + (4096*x0)), tmp22, rmask & xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ap/capazybbmitbuviukrgvztks4tyvg3dpptakn7q4vjslonrba25g.py
# Topologically Sorted Source Nodes: [channel_wv], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# channel_wv => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_2, %primals_3, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_2 = async_compile.triton('triton_poi_fused_convolution_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, 4096], 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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_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_2(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 1024
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y0 = yindex % 256
y1 = (yindex // 256)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (256*x2) + (1048576*y1)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + (4096*y3)), tmp2, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rw/crwczp7wd3om4xrznty4rti6kiafa2tzqyvn6po6scyvmhmtcnmw.py
# Topologically Sorted Source Nodes: [conv2d_2, layer_norm, sigmoid], Original ATen: [aten.convolution, aten.native_layer_norm, aten.sigmoid]
# Source node to ATen node mapping:
# conv2d_2 => convolution_2
# layer_norm => add, add_1, mul, mul_1, rsqrt, sub_1, var_mean
# sigmoid => sigmoid
# Graph fragment:
# %convolution_2 : [num_users=2] = call_function[target=torch.ops.aten.convolution.default](args = (%unsqueeze, %primals_6, %primals_7, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%permute, [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=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%permute, %getitem_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %rsqrt), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %primals_8), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_9), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%add_1,), kwargs = {})
triton_per_fused_convolution_native_layer_norm_sigmoid_3 = async_compile.triton('triton_per_fused_convolution_native_layer_norm_sigmoid_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=[4, 512],
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, 8), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_convolution_native_layer_norm_sigmoid_3', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1'], 'no_x_dim': True, '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_layer_norm_sigmoid_3(in_out_ptr0, in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel):
xnumel = 4
XBLOCK: tl.constexpr = 1
rnumel = 512
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 = tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (r1 + (512*x0)), None)
tmp1 = tl.load(in_ptr0 + (r1), None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr2 + (r1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = tl.broadcast_to(tmp3, [RBLOCK])
tmp7 = triton_helpers.promote_to_tensor(tl.sum(tmp5, 0))
tmp8 = tl.full([1], 512, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp3 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 512.0
tmp17 = tmp15 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tmp21 = tmp2 - tmp10
tmp22 = tmp21 * tmp20
tmp24 = tmp22 * tmp23
tmp26 = tmp24 + tmp25
tmp27 = tl.sigmoid(tmp26)
tl.store(in_out_ptr0 + (r1 + (512*x0)), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + (x0), tmp20, None)
tl.store(out_ptr1 + (r1 + (512*x0)), tmp27, None)
tl.store(out_ptr0 + (x0), tmp10, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/4q/c4qd62sdhyd4fglxwxeivg7nkbd43odbdwdvxiybb2xu7tqa2uxs.py
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
# Source node to ATen node mapping:
# spatial_wq => convolution_4
# spatial_wq_1 => mean
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_12, %primals_13, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%convolution_4, [-1, -2], True), kwargs = {})
triton_red_fused_convolution_mean_4 = async_compile.triton('triton_red_fused_convolution_mean_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.reduction(
size_hints=[32768, 128],
reduction_hint=ReductionHint.OUTER,
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_red_fused_convolution_mean_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_red_fused_convolution_mean_4(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 32768
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 256
x1 = (xindex // 256)
tmp1 = tl.load(in_ptr1 + (x0), None, eviction_policy='evict_last')
_tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + (256*r2) + (32768*x1)), rmask, eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = _tmp4 + tmp3
_tmp4 = tl.where(rmask, tmp5, _tmp4)
tmp4 = tl.sum(_tmp4, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/mi/cminjfxbzpwu3wpqzd7rt3ll256xfebkzux3hh5qykvjrgsawjdj.py
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
# Source node to ATen node mapping:
# spatial_wq => convolution_4
# spatial_wq_1 => mean
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_12, %primals_13, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%convolution_4, [-1, -2], True), kwargs = {})
triton_per_fused_convolution_mean_5 = async_compile.triton('triton_per_fused_convolution_mean_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=[1024, 32],
reduction_hint=ReductionHint.OUTER,
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_convolution_mean_5', '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_convolution_mean_5(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1024
rnumel = 32
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 = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 256
x1 = (xindex // 256)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (256*r2) + (8192*x1)), 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 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ji/cjiocwcpglmk3orhjwyrchueogqsp53hxepr45cwdplqbplk7tvb.py
# Topologically Sorted Source Nodes: [spatial_wq_3], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# spatial_wq_3 => amax_1, div_1, exp_1, sub_2, sum_2
# Graph fragment:
# %amax_1 : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_8, [-1], True), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_8, %amax_1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_2,), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_1, [-1], True), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_1, %sum_2), kwargs = {})
triton_per_fused__softmax_6 = async_compile.triton('triton_per_fused__softmax_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=[4, 256],
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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_6', 'mutated_arg_names': [], 'no_x_dim': True, '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_6(in_ptr0, out_ptr2, xnumel, rnumel):
xnumel = 4
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)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (256*x0)), None)
tmp1 = 4096.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(triton_helpers.max2(tmp3, 0))
tmp6 = tmp2 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = tmp7 / tmp10
tl.store(out_ptr2 + (r1 + (256*x0)), tmp11, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/4q/c4qcgqgap2qx55qlbkmr6i7aulzzwfidlkqpvvrhhbpq2j4mbsfz.py
# Topologically Sorted Source Nodes: [channel_out, spatial_weight, spatial_out, out], Original ATen: [aten.mul, aten.sigmoid, aten.add]
# Source node to ATen node mapping:
# channel_out => mul_2
# out => add_2
# spatial_out => mul_3
# spatial_weight => sigmoid_1
# Graph fragment:
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_6, %primals_1), kwargs = {})
# %sigmoid_1 : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_12,), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid_1, %primals_1), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_3, %mul_2), kwargs = {})
triton_poi_fused_add_mul_sigmoid_7 = async_compile.triton('triton_poi_fused_add_mul_sigmoid_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=[16384, 512], tile_hint=TileHint.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, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_mul_sigmoid_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_mul_sigmoid_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16384
xnumel = 512
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
y3 = yindex
x2 = xindex
y1 = (yindex // 4096)
y0 = yindex % 4096
tmp0 = tl.load(in_ptr0 + (y3), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr1 + (x2 + (512*y3)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr2 + (x2 + (512*y1)), xmask, eviction_policy='evict_last')
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp5 = tmp4 * tmp2
tmp6 = tmp3 + tmp5
tl.store(out_ptr0 + (y0 + (4096*x2) + (2097152*y1)), 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 = args
args.clear()
assert_size_stride(primals_1, (4, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_3, (256, ), (1, ))
assert_size_stride(primals_4, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_5, (1, ), (1, ))
assert_size_stride(primals_6, (512, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_7, (512, ), (1, ))
assert_size_stride(primals_8, (512, ), (1, ))
assert_size_stride(primals_9, (512, ), (1, ))
assert_size_stride(primals_10, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (256, ), (1, ))
assert_size_stride(primals_12, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (256, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(primals_1, buf0, 2048, 4096, grid=grid(2048, 4096), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [channel_wv], 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, 256, 64, 64), (1048576, 1, 16384, 256))
# Topologically Sorted Source Nodes: [channel_wq], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf0, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 64, 64), (4096, 1, 64, 1))
buf5 = empty_strided_cuda((4, 4096, 1), (4096, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [channel_wq_2], Original ATen: [aten._softmax]
triton_red_fused__softmax_1.run(buf2, primals_5, buf5, 4, 4096, grid=grid(4), stream=stream0)
del primals_5
buf6 = empty_strided_cuda((4, 256, 64, 64), (1048576, 4096, 64, 1), torch.float32)
# Topologically Sorted Source Nodes: [channel_wv], Original ATen: [aten.convolution]
triton_poi_fused_convolution_2.run(buf1, primals_3, buf6, 1024, 4096, grid=grid(1024, 4096), stream=stream0)
del buf1
del primals_3
buf7 = empty_strided_cuda((4, 256, 1), (256, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 256, 4096), (1048576, 4096, 1), 0), buf5, out=buf7)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(reinterpret_tensor(buf7, (4, 256, 1, 1), (256, 1, 1, 1), 0), primals_6, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 512, 1, 1), (512, 1, 1, 1))
buf9 = reinterpret_tensor(buf8, (4, 512, 1, 1), (512, 1, 512, 512), 0); del buf8 # reuse
buf10 = empty_strided_cuda((4, 1, 1), (1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((4, 1, 1), (1, 4, 4), torch.float32)
buf13 = reinterpret_tensor(buf11, (4, 1, 1), (1, 1, 1), 0); del buf11 # reuse
buf14 = empty_strided_cuda((4, 1, 512), (512, 2048, 1), torch.float32)
# Topologically Sorted Source Nodes: [conv2d_2, layer_norm, sigmoid], Original ATen: [aten.convolution, aten.native_layer_norm, aten.sigmoid]
triton_per_fused_convolution_native_layer_norm_sigmoid_3.run(buf9, buf13, primals_7, primals_8, primals_9, buf10, buf14, 4, 512, grid=grid(4), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [spatial_wv], Original ATen: [aten.convolution]
buf15 = extern_kernels.convolution(buf0, primals_10, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf15, (4, 256, 64, 64), (1048576, 1, 16384, 256))
# Topologically Sorted Source Nodes: [spatial_wq], Original ATen: [aten.convolution]
buf16 = extern_kernels.convolution(buf0, primals_12, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf17 = empty_strided_cuda((4, 256, 1, 1, 32), (8192, 1, 32768, 32768, 256), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
triton_red_fused_convolution_mean_4.run(buf16, primals_13, buf17, 32768, 128, grid=grid(32768), stream=stream0)
del primals_13
buf18 = empty_strided_cuda((4, 256, 1, 1), (256, 1, 1024, 1024), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
triton_per_fused_convolution_mean_5.run(buf17, buf18, 1024, 32, grid=grid(1024), stream=stream0)
del buf17
buf21 = empty_strided_cuda((4, 1, 256), (256, 256, 1), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq_3], Original ATen: [aten._softmax]
triton_per_fused__softmax_6.run(buf18, buf21, 4, 256, grid=grid(4), stream=stream0)
del buf18
buf22 = reinterpret_tensor(buf16, (4, 256, 64, 64), (1048576, 4096, 64, 1), 0); del buf16 # reuse
# Topologically Sorted Source Nodes: [spatial_wv], Original ATen: [aten.convolution]
triton_poi_fused_convolution_2.run(buf15, primals_11, buf22, 1024, 4096, grid=grid(1024, 4096), stream=stream0)
del buf15
del primals_11
buf23 = reinterpret_tensor(buf2, (4, 1, 4096), (4096, 4096, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [spatial_wz], Original ATen: [aten.bmm]
extern_kernels.bmm(buf21, reinterpret_tensor(buf22, (4, 256, 4096), (1048576, 4096, 1), 0), out=buf23)
buf24 = empty_strided_cuda((4, 512, 64, 64), (2097152, 4096, 64, 1), torch.float32)
# Topologically Sorted Source Nodes: [channel_out, spatial_weight, spatial_out, out], Original ATen: [aten.mul, aten.sigmoid, aten.add]
triton_poi_fused_add_mul_sigmoid_7.run(buf23, buf0, buf14, buf24, 16384, 512, grid=grid(16384, 512), stream=stream0)
del buf14
return (buf24, buf0, primals_2, primals_4, primals_6, primals_8, primals_9, primals_10, primals_12, buf5, reinterpret_tensor(buf7, (4, 256, 1, 1), (256, 1, 1, 1), 0), buf9, buf10, buf13, buf21, buf23, reinterpret_tensor(buf22, (4, 4096, 256), (1048576, 1, 4096), 0), reinterpret_tensor(buf6, (4, 4096, 256), (1048576, 1, 4096), 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, 512, 64, 64), (2097152, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((1, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((1, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((512, 256, 1, 1), (256, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((256, ), (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 torch
from torch import nn
class ParallelPolarizedSelfAttention(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, x):
b, c, h, w = x.size()
channel_wv = self.ch_wv(x)
channel_wq = self.ch_wq(x)
channel_wv = channel_wv.reshape(b, c // 2, -1)
channel_wq = channel_wq.reshape(b, -1, 1)
channel_wq = self.softmax_channel(channel_wq)
channel_wz = torch.matmul(channel_wv, channel_wq).unsqueeze(-1)
channel_weight = self.sigmoid(self.ln(self.ch_wz(channel_wz).
reshape(b, c, 1).permute(0, 2, 1))).permute(0, 2, 1).reshape(b,
c, 1, 1)
channel_out = channel_weight * x
spatial_wv = self.sp_wv(x)
spatial_wq = self.sp_wq(x)
spatial_wq = self.agp(spatial_wq)
spatial_wv = spatial_wv.reshape(b, c // 2, -1)
spatial_wq = spatial_wq.permute(0, 2, 3, 1).reshape(b, 1, c // 2)
spatial_wq = self.softmax_spatial(spatial_wq)
spatial_wz = torch.matmul(spatial_wq, spatial_wv)
spatial_weight = self.sigmoid(spatial_wz.reshape(b, 1, h, w))
spatial_out = spatial_weight * x
out = spatial_out + channel_out
return out
def get_inputs():
return [torch.rand([4, 512, 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 libdevice, math as tl_math
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_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = yindex // 512
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y3), None, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 512 * x2 + 2097152 * y1), tmp0, None)
@triton.jit
def triton_red_fused__softmax_1(in_ptr0, in_ptr1, out_ptr2, xnumel, rnumel,
XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 4
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
_tmp5 = tl.full([XBLOCK, RBLOCK], float('-inf'), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_last', other=0.0)
tmp3 = tmp0 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = triton_helpers.maximum(_tmp5, tmp4)
_tmp5 = tl.where(rmask & xmask, tmp6, _tmp5)
tmp5 = triton_helpers.max2(_tmp5, 1)[:, None]
tmp8 = tl.load(in_ptr1 + 0)
tmp9 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
_tmp14 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp7 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tmp7 + tmp9
tmp11 = tmp10 - tmp5
tmp12 = tl_math.exp(tmp11)
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = _tmp14 + tmp13
_tmp14 = tl.where(rmask & xmask, tmp15, _tmp14)
tmp14 = tl.sum(_tmp14, 1)[:, None]
tmp17 = tl.load(in_ptr1 + 0)
tmp18 = tl.broadcast_to(tmp17, [XBLOCK, RBLOCK])
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp16 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp19 = tmp16 + tmp18
tmp20 = tmp19 - tmp5
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp21 / tmp14
tl.store(out_ptr2 + (r1 + 4096 * x0), tmp22, rmask & xmask)
@triton.jit
def triton_poi_fused_convolution_2(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y0 = yindex % 256
y1 = yindex // 256
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 256 * x2 + 1048576 * y1), None,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 4096 * y3), tmp2, None)
@triton.jit
def triton_per_fused_convolution_native_layer_norm_sigmoid_3(in_out_ptr0,
in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel
):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (r1 + 512 * x0), None)
tmp1 = tl.load(in_ptr0 + r1, None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr2 + r1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = tl.broadcast_to(tmp3, [RBLOCK])
tmp7 = triton_helpers.promote_to_tensor(tl.sum(tmp5, 0))
tmp8 = tl.full([1], 512, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp3 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 512.0
tmp17 = tmp15 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tmp21 = tmp2 - tmp10
tmp22 = tmp21 * tmp20
tmp24 = tmp22 * tmp23
tmp26 = tmp24 + tmp25
tmp27 = tl.sigmoid(tmp26)
tl.store(in_out_ptr0 + (r1 + 512 * x0), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x0, tmp20, None)
tl.store(out_ptr1 + (r1 + 512 * x0), tmp27, None)
tl.store(out_ptr0 + x0, tmp10, None)
@triton.jit
def triton_red_fused_convolution_mean_4(in_ptr0, in_ptr1, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 256
x1 = xindex // 256
tmp1 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
_tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + 256 * r2 + 32768 * x1), rmask,
eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = _tmp4 + tmp3
_tmp4 = tl.where(rmask, tmp5, _tmp4)
tmp4 = tl.sum(_tmp4, 1)[:, None]
tl.store(out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_per_fused_convolution_mean_5(in_ptr0, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
xnumel = 1024
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, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 256
x1 = xindex // 256
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 256 * r2 + 8192 * x1), 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 + x3, tmp4, xmask)
@triton.jit
def triton_per_fused__softmax_6(in_ptr0, out_ptr2, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 256 * x0), None)
tmp1 = 4096.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(triton_helpers.max2(tmp3, 0))
tmp6 = tmp2 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = tmp7 / tmp10
tl.store(out_ptr2 + (r1 + 256 * x0), tmp11, None)
@triton.jit
def triton_poi_fused_add_mul_sigmoid_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
xnumel = 512
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
y3 = yindex
x2 = xindex
y1 = yindex // 4096
y0 = yindex % 4096
tmp0 = tl.load(in_ptr0 + y3, None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr1 + (x2 + 512 * y3), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr2 + (x2 + 512 * y1), xmask, eviction_policy=
'evict_last')
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tmp5 = tmp4 * tmp2
tmp6 = tmp3 + tmp5
tl.store(out_ptr0 + (y0 + 4096 * x2 + 2097152 * y1), 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) = args
args.clear()
assert_size_stride(primals_1, (4, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_3, (256,), (1,))
assert_size_stride(primals_4, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_5, (1,), (1,))
assert_size_stride(primals_6, (512, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_7, (512,), (1,))
assert_size_stride(primals_8, (512,), (1,))
assert_size_stride(primals_9, (512,), (1,))
assert_size_stride(primals_10, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (256,), (1,))
assert_size_stride(primals_12, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (256,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512
), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(2048, 4096)](primals_1, buf0, 2048, 4096,
XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
del primals_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, 256, 64, 64), (1048576, 1, 16384, 256))
buf2 = extern_kernels.convolution(buf0, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 64, 64), (4096, 1, 64, 1))
buf5 = empty_strided_cuda((4, 4096, 1), (4096, 1, 1), torch.float32)
triton_red_fused__softmax_1[grid(4)](buf2, primals_5, buf5, 4, 4096,
XBLOCK=1, RBLOCK=2048, num_warps=16, num_stages=1)
del primals_5
buf6 = empty_strided_cuda((4, 256, 64, 64), (1048576, 4096, 64, 1),
torch.float32)
triton_poi_fused_convolution_2[grid(1024, 4096)](buf1, primals_3,
buf6, 1024, 4096, XBLOCK=64, YBLOCK=64, num_warps=8, num_stages=1)
del buf1
del primals_3
buf7 = empty_strided_cuda((4, 256, 1), (256, 1, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 256, 4096), (
1048576, 4096, 1), 0), buf5, out=buf7)
buf8 = extern_kernels.convolution(reinterpret_tensor(buf7, (4, 256,
1, 1), (256, 1, 1, 1), 0), primals_6, stride=(1, 1), padding=(0,
0), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf8, (4, 512, 1, 1), (512, 1, 1, 1))
buf9 = reinterpret_tensor(buf8, (4, 512, 1, 1), (512, 1, 512, 512), 0)
del buf8
buf10 = empty_strided_cuda((4, 1, 1), (1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((4, 1, 1), (1, 4, 4), torch.float32)
buf13 = reinterpret_tensor(buf11, (4, 1, 1), (1, 1, 1), 0)
del buf11
buf14 = empty_strided_cuda((4, 1, 512), (512, 2048, 1), torch.float32)
triton_per_fused_convolution_native_layer_norm_sigmoid_3[grid(4)](buf9,
buf13, primals_7, primals_8, primals_9, buf10, buf14, 4, 512,
num_warps=4, num_stages=1)
del primals_7
buf15 = extern_kernels.convolution(buf0, primals_10, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf15, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf16 = extern_kernels.convolution(buf0, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf17 = empty_strided_cuda((4, 256, 1, 1, 32), (8192, 1, 32768,
32768, 256), torch.float32)
triton_red_fused_convolution_mean_4[grid(32768)](buf16, primals_13,
buf17, 32768, 128, XBLOCK=64, RBLOCK=8, num_warps=4, num_stages=1)
del primals_13
buf18 = empty_strided_cuda((4, 256, 1, 1), (256, 1, 1024, 1024),
torch.float32)
triton_per_fused_convolution_mean_5[grid(1024)](buf17, buf18, 1024,
32, XBLOCK=128, num_warps=8, num_stages=1)
del buf17
buf21 = empty_strided_cuda((4, 1, 256), (256, 256, 1), torch.float32)
triton_per_fused__softmax_6[grid(4)](buf18, buf21, 4, 256,
num_warps=2, num_stages=1)
del buf18
buf22 = reinterpret_tensor(buf16, (4, 256, 64, 64), (1048576, 4096,
64, 1), 0)
del buf16
triton_poi_fused_convolution_2[grid(1024, 4096)](buf15, primals_11,
buf22, 1024, 4096, XBLOCK=64, YBLOCK=64, num_warps=8, num_stages=1)
del buf15
del primals_11
buf23 = reinterpret_tensor(buf2, (4, 1, 4096), (4096, 4096, 1), 0)
del buf2
extern_kernels.bmm(buf21, reinterpret_tensor(buf22, (4, 256, 4096),
(1048576, 4096, 1), 0), out=buf23)
buf24 = empty_strided_cuda((4, 512, 64, 64), (2097152, 4096, 64, 1),
torch.float32)
triton_poi_fused_add_mul_sigmoid_7[grid(16384, 512)](buf23, buf0,
buf14, buf24, 16384, 512, XBLOCK=32, YBLOCK=32, num_warps=4,
num_stages=1)
del buf14
return (buf24, buf0, primals_2, primals_4, primals_6, primals_8,
primals_9, primals_10, primals_12, buf5, reinterpret_tensor(buf7, (
4, 256, 1, 1), (256, 1, 1, 1), 0), buf9, buf10, buf13, buf21, buf23,
reinterpret_tensor(buf22, (4, 4096, 256), (1048576, 1, 4096), 0),
reinterpret_tensor(buf6, (4, 4096, 256), (1048576, 1, 4096), 0))
class ParallelPolarizedSelfAttentionNew(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, input_0):
primals_2 = self.ch_wv.weight
primals_3 = self.ch_wv.bias
primals_4 = self.ch_wq.weight
primals_5 = self.ch_wq.bias
primals_6 = self.ch_wz.weight
primals_7 = self.ch_wz.bias
primals_8 = self.ln.weight
primals_9 = self.ln.bias
primals_10 = self.sp_wv.weight
primals_11 = self.sp_wv.bias
primals_12 = self.sp_wq.weight
primals_13 = self.sp_wq.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,
primals_10, primals_11, primals_12, primals_13])
return output[0]
| rushirajsherlocked/External-Attention-pytorch | ParallelPolarizedSelfAttention | false | 4,299 | [
"MIT"
] | 0 | 7d6814b2d90909adf81c62f3f8a89e30a59d6481 | https://github.com/rushirajsherlocked/External-Attention-pytorch/tree/7d6814b2d90909adf81c62f3f8a89e30a59d6481 | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, x):
b, c, h, w = x.size()
channel_wv = self.ch_wv(x)
channel_wq = self.ch_wq(x)
channel_wv = channel_wv.reshape(b, c // 2, -1)
channel_wq = channel_wq.reshape(b, -1, 1)
channel_wq = self.softmax_channel(channel_wq)
channel_wz = torch.matmul(channel_wv, channel_wq).unsqueeze(-1)
channel_weight = self.sigmoid(self.ln(self.ch_wz(channel_wz).
reshape(b, c, 1).permute(0, 2, 1))).permute(0, 2, 1).reshape(b,
c, 1, 1)
channel_out = channel_weight * x
spatial_wv = self.sp_wv(x)
spatial_wq = self.sp_wq(x)
spatial_wq = self.agp(spatial_wq)
spatial_wv = spatial_wv.reshape(b, c // 2, -1)
spatial_wq = spatial_wq.permute(0, 2, 3, 1).reshape(b, 1, c // 2)
spatial_wq = self.softmax_spatial(spatial_wq)
spatial_wz = torch.matmul(spatial_wq, spatial_wv)
spatial_weight = self.sigmoid(spatial_wz.reshape(b, 1, h, w))
spatial_out = spatial_weight * x
out = spatial_out + channel_out
return out
def get_inputs():
return [torch.rand([4, 512, 64, 64])]
def get_init_inputs():
return []
|
BertSelfAttention | # 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_7/inductor_cache/dk/cdk4odz276xorciau5ehgl7f3s2mgkf3hrye6xep6kzubczdeqqy.py
# Topologically Sorted Source Nodes: [attention_scores], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# attention_scores => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand,), 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=[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_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_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
tl.store(out_ptr0 + (x2 + (4*y3)), tmp2, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/fh/cfhydrwaejluqurspx2dvattjq4qiiexqog4gfsdmfob43r5rnk5.py
# Topologically Sorted Source Nodes: [attention_scores_1, attention_scores_2, attention_probs], Original ATen: [aten.div, aten.add, aten._softmax]
# Source node to ATen node mapping:
# attention_probs => amax, exp, sub, sum_1
# attention_scores_1 => div
# attention_scores_2 => add
# Graph fragment:
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%view_11, 1.0), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, %primals_8), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %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_add_div_1 = async_compile.triton('triton_poi_fused__softmax_add_div_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__softmax_add_div_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_add_div_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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + (4*x2), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + (4*x2)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (2 + (4*x2)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (3 + (4*x2)), xmask, eviction_policy='evict_last')
tmp17 = tl.load(in_ptr1 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
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_7/inductor_cache/g6/cg65jkqmchbdahuikqawep5p32pz7fmy6cfvgzohspjy5l4iqhrs.py
# Topologically Sorted Source Nodes: [attention_scores_1, attention_scores_2, attention_probs], Original ATen: [aten.div, aten.add, aten._softmax]
# Source node to ATen node mapping:
# attention_probs => amax, div_1, exp, sub
# attention_scores_1 => div
# attention_scores_2 => add
# Graph fragment:
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%view_11, 1.0), kwargs = {})
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, %primals_8), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add, [-1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_add_div_2 = async_compile.triton('triton_poi_fused__softmax_add_div_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__softmax_add_div_2', '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__softmax_add_div_2(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
x3 = xindex
x4 = xindex % 64
x5 = (xindex // 4)
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp3 = tl.load(in_ptr0 + (x4), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (x5), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr2 + (x5), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tl.store(in_out_ptr0 + (x3), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/xt/cxtkkmujo4ytg6ycpz5lk5livtstr63pg5nsf5ijewjbtrfrqx6k.py
# Topologically Sorted Source Nodes: [context_layer_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# context_layer_1 => clone_4
# Graph fragment:
# %clone_4 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_6,), 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.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_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 = 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 = 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))
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), (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: [], 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 = 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_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_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [attention_scores], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(buf0, primals_2, buf3, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [attention_scores], Original ATen: [aten.clone]
triton_poi_fused_clone_0.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: [attention_scores], 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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0); del buf1 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [attention_scores_1, attention_scores_2, attention_probs], Original ATen: [aten.div, aten.add, aten._softmax]
triton_poi_fused__softmax_add_div_1.run(buf5, primals_8, buf6, buf7, 64, grid=grid(64), stream=stream0)
buf8 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [attention_scores_1, attention_scores_2, attention_probs], Original ATen: [aten.div, aten.add, aten._softmax]
triton_poi_fused__softmax_add_div_2.run(buf8, primals_8, buf6, buf7, 256, grid=grid(256), stream=stream0)
del primals_8
buf9 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [context_layer], Original ATen: [aten.clone]
triton_poi_fused_clone_0.run(buf2, primals_7, buf9, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_7
buf10 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [context_layer], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf9, (16, 4, 1), (4, 1, 0), 0), out=buf10)
buf11 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [context_layer_1], Original ATen: [aten.clone]
triton_poi_fused_clone_3.run(buf10, buf11, 16, 4, grid=grid(16, 4), stream=stream0)
del buf10
return (reinterpret_tensor(buf11, (4, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), buf8, reinterpret_tensor(buf9, (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, 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), (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])
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 math
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
import torch.utils.data.distributed
import torch.nn.functional as F
class BertSelfAttention(nn.Module):
def __init__(self, config):
super(BertSelfAttention, self).__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 1, 3)
def transpose_key_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 3, 1)
def forward(self, hidden_states, attention_mask):
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(hidden_states)
mixed_value_layer = self.value(hidden_states)
query_layer = self.transpose_for_scores(mixed_query_layer)
key_layer = self.transpose_key_for_scores(mixed_key_layer)
value_layer = self.transpose_for_scores(mixed_value_layer)
attention_scores = torch.matmul(query_layer, key_layer)
attention_scores = attention_scores / math.sqrt(self.
attention_head_size)
attention_scores = attention_scores + attention_mask
attention_probs = F.softmax(attention_scores, dim=-1)
attention_probs = self.dropout(attention_probs)
context_layer = torch.matmul(attention_probs, value_layer)
context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
new_context_layer_shape = context_layer.size()[:-2] + (self.
all_head_size,)
context_layer = torch.reshape(context_layer, new_context_layer_shape)
return context_layer
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'config': _mock_config(hidden_size=4, num_attention_heads=
4, attention_probs_dropout_prob=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 import triton_helpers
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
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
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_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
tl.store(out_ptr0 + (x2 + 4 * y3), tmp2, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_add_div_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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + 4 * x2, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (1 + 4 * x2), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (2 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr0 + (3 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp17 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp1 = 1.0
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_add_div_2(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
x3 = xindex
x4 = xindex % 64
x5 = xindex // 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp3 = tl.load(in_ptr0 + x4, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + x5, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr2 + x5, xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tl.store(in_out_ptr0 + x3, tmp9, xmask)
@triton.jit
def triton_poi_fused_clone_3(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) = 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))
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), (16, 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 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (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_3, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(16, 4)](buf0, primals_2, buf3, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_clone_0[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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0)
del buf1
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
triton_poi_fused__softmax_add_div_1[grid(64)](buf5, primals_8, buf6,
buf7, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf8 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused__softmax_add_div_2[grid(256)](buf8, primals_8,
buf6, buf7, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_8
buf9 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf7
triton_poi_fused_clone_0[grid(16, 4)](buf2, primals_7, buf9, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf10 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf9, (16, 4, 1), (4, 1, 0), 0), out=buf10)
buf11 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf6
triton_poi_fused_clone_3[grid(16, 4)](buf10, buf11, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
del buf10
return reinterpret_tensor(buf11, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0
), buf8, reinterpret_tensor(buf9, (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 BertSelfAttentionNew(nn.Module):
def __init__(self, config):
super(BertSelfAttentionNew, self).__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 1, 3)
def transpose_key_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 3, 1)
def forward(self, input_0, input_1):
primals_1 = self.query.weight
primals_2 = self.query.bias
primals_4 = self.key.weight
primals_5 = self.key.bias
primals_6 = self.value.weight
primals_7 = self.value.bias
primals_3 = input_0
primals_8 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8])
return output[0]
| sermolin/amazon-sagemaker-examples | BertSelfAttention | false | 4,300 | [
"Apache-2.0"
] | 0 | 3e6083d1b53cb718893a04c46513a9482a17bd6b | https://github.com/sermolin/amazon-sagemaker-examples/tree/3e6083d1b53cb718893a04c46513a9482a17bd6b | from _paritybench_helpers import _mock_config
import math
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
import torch.utils.data.distributed
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, config):
super().__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 1, 3)
def transpose_key_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = torch.reshape(x, new_x_shape)
return x.permute(0, 2, 3, 1)
def forward(self, hidden_states, attention_mask):
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(hidden_states)
mixed_value_layer = self.value(hidden_states)
query_layer = self.transpose_for_scores(mixed_query_layer)
key_layer = self.transpose_key_for_scores(mixed_key_layer)
value_layer = self.transpose_for_scores(mixed_value_layer)
attention_scores = torch.matmul(query_layer, key_layer)
attention_scores = attention_scores / math.sqrt(self.
attention_head_size)
attention_scores = attention_scores + attention_mask
attention_probs = F.softmax(attention_scores, dim=-1)
attention_probs = self.dropout(attention_probs)
context_layer = torch.matmul(attention_probs, value_layer)
context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
new_context_layer_shape = context_layer.size()[:-2] + (self.
all_head_size,)
context_layer = torch.reshape(context_layer, new_context_layer_shape)
return context_layer
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'config': _mock_config(hidden_size=4, num_attention_heads=
4, attention_probs_dropout_prob=0.5)}]
|
BaselineNN | # 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_7/inductor_cache/ky/cky64l574tkwxzjewzevqyhty73x4t3q4p6d2tu2humfvstjwiaa.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
# %le_3 : [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_7/inductor_cache/q5/cq52p2qap7uob2ddnn4qeh67r3muutkp3yhbkqpu4eqaemol3idl.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.sigmoid]
# Source node to ATen node mapping:
# x_4 => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_9,), kwargs = {})
triton_poi_fused_sigmoid_1 = async_compile.triton('triton_poi_fused_sigmoid_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_sigmoid_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_sigmoid_1(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 = tl.sigmoid(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, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11 = 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, (32, 32), (32, 1))
assert_size_stride(primals_5, (32, ), (1, ))
assert_size_stride(primals_6, (32, 32), (32, 1))
assert_size_stride(primals_7, (32, ), (1, ))
assert_size_stride(primals_8, (32, 32), (32, 1))
assert_size_stride(primals_9, (32, ), (1, ))
assert_size_stride(primals_10, (4, 32), (32, 1))
assert_size_stride(primals_11, (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
buf13 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf13, 2048, grid=grid(2048), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 32), (32, 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, 32), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 32), (512, 128, 32, 1), 0); del buf2 # reuse
buf12 = 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]
triton_poi_fused_relu_threshold_backward_0.run(buf3, primals_5, buf12, 2048, grid=grid(2048), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf3, (64, 32), (32, 1), 0), reinterpret_tensor(primals_6, (32, 32), (1, 32), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 32), (512, 128, 32, 1), 0); del buf4 # reuse
buf11 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_0.run(buf5, primals_7, buf11, 2048, grid=grid(2048), stream=stream0)
del primals_7
buf6 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf5, (64, 32), (32, 1), 0), reinterpret_tensor(primals_8, (32, 32), (1, 32), 0), out=buf6)
buf7 = reinterpret_tensor(buf6, (4, 4, 4, 32), (512, 128, 32, 1), 0); del buf6 # reuse
buf10 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_0.run(buf7, primals_9, buf10, 2048, grid=grid(2048), stream=stream0)
del primals_9
buf8 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf7, (64, 32), (32, 1), 0), reinterpret_tensor(primals_10, (32, 4), (1, 32), 0), out=buf8)
buf9 = reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf8 # reuse
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.sigmoid]
triton_poi_fused_sigmoid_1.run(buf9, primals_11, 256, grid=grid(256), stream=stream0)
del primals_11
return (buf9, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(buf3, (64, 32), (32, 1), 0), reinterpret_tensor(buf5, (64, 32), (32, 1), 0), reinterpret_tensor(buf7, (64, 32), (32, 1), 0), buf9, primals_10, buf10, primals_8, buf11, primals_6, buf12, primals_4, 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((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((32, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((32, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((32, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, 32), (32, 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
from torch import nn
import torch.nn.functional as F
class BaselineNN(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(4, 32)
self.fc2 = nn.Linear(32, 32)
self.fc3 = nn.Linear(32, 32)
self.fc4 = nn.Linear(32, 32)
self.fc5 = nn.Linear(32, 4)
def forward(self, x):
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = F.relu(self.fc3(x))
x = F.relu(self.fc4(x))
x = torch.sigmoid(self.fc5(x))
return x
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 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_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_sigmoid_1(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 = tl.sigmoid(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, 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, (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, (32, 32), (32, 1))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (32, 32), (32, 1))
assert_size_stride(primals_7, (32,), (1,))
assert_size_stride(primals_8, (32, 32), (32, 1))
assert_size_stride(primals_9, (32,), (1,))
assert_size_stride(primals_10, (4, 32), (32, 1))
assert_size_stride(primals_11, (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
buf13 = 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, buf13, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_4, (32, 32), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf2
buf12 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool
)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf3,
primals_5, buf12, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_6, (32, 32), (1, 32), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf4
buf11 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool
)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf5,
primals_7, buf11, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf5, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_8, (32, 32), (1, 32), 0), out=buf6)
buf7 = reinterpret_tensor(buf6, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf6
buf10 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool
)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf7,
primals_9, buf10, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_10, (32, 4), (1, 32), 0), out=buf8)
buf9 = reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf8
triton_poi_fused_sigmoid_1[grid(256)](buf9, primals_11, 256, XBLOCK
=256, num_warps=4, num_stages=1)
del primals_11
return (buf9, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(
buf3, (64, 32), (32, 1), 0), reinterpret_tensor(buf5, (64, 32), (32,
1), 0), reinterpret_tensor(buf7, (64, 32), (32, 1), 0), buf9,
primals_10, buf10, primals_8, buf11, primals_6, buf12, primals_4, buf13
)
class BaselineNNNew(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(4, 32)
self.fc2 = nn.Linear(32, 32)
self.fc3 = nn.Linear(32, 32)
self.fc4 = nn.Linear(32, 32)
self.fc5 = nn.Linear(32, 4)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.bias
primals_8 = self.fc4.weight
primals_9 = self.fc4.bias
primals_10 = self.fc5.weight
primals_11 = self.fc5.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])
return output[0]
| severilov/master-thesis | BaselineNN | false | 4,301 | [
"MIT"
] | 0 | 145382d5d551761fcdbd2b77d7b96fabcc8f78ec | https://github.com/severilov/master-thesis/tree/145382d5d551761fcdbd2b77d7b96fabcc8f78ec | import torch
from torch import nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self):
super().__init__()
self.fc1 = nn.Linear(4, 32)
self.fc2 = nn.Linear(32, 32)
self.fc3 = nn.Linear(32, 32)
self.fc4 = nn.Linear(32, 32)
self.fc5 = nn.Linear(32, 4)
def forward(self, x):
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = F.relu(self.fc3(x))
x = F.relu(self.fc4(x))
x = torch.sigmoid(self.fc5(x))
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Maxout | # 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_7/inductor_cache/7x/c7xn44wdrtri4v2aoqtyt5ru7o3rxafk4evdmpdrhtxp3gme35rl.py
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.maximum, aten.eq, aten.gt, aten.lt]
# Source node to ATen node mapping:
# max_1 => maximum
# Graph fragment:
# %maximum : [num_users=1] = call_function[target=torch.ops.aten.maximum.default](args = (%view_1, %view_3), kwargs = {})
# %eq : [num_users=1] = call_function[target=torch.ops.aten.eq.Tensor](args = (%view_1, %view_3), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Tensor](args = (%view_1, %view_3), kwargs = {})
# %lt : [num_users=1] = call_function[target=torch.ops.aten.lt.Tensor](args = (%view_1, %view_3), kwargs = {})
triton_poi_fused_eq_gt_lt_maximum_0 = async_compile.triton('triton_poi_fused_eq_gt_lt_maximum_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: '*fp32', 5: '*i1', 6: '*i1', 7: '*i1', 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_poi_fused_eq_gt_lt_maximum_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_eq_gt_lt_maximum_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, out_ptr1, out_ptr2, out_ptr3, 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')
tmp3 = tl.load(in_ptr2 + (x2), xmask)
tmp4 = tl.load(in_ptr3 + (x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = triton_helpers.maximum(tmp2, tmp5)
tmp7 = tmp2 == tmp5
tmp8 = tmp2 > tmp5
tmp9 = tmp2 < tmp5
tl.store(out_ptr0 + (x2), tmp6, xmask)
tl.store(out_ptr1 + (x2), tmp7, xmask)
tl.store(out_ptr2 + (x2), tmp8, xmask)
tl.store(out_ptr3 + (x2), 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 = 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 = 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_4, (4, 4), (1, 4), 0), out=buf1)
del primals_4
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.maximum, aten.eq, aten.gt, aten.lt]
stream0 = get_raw_stream(0)
triton_poi_fused_eq_gt_lt_maximum_0.run(buf0, primals_2, buf1, primals_5, buf2, buf3, buf4, buf5, 256, grid=grid(256), stream=stream0)
del buf0
del buf1
del primals_2
del primals_5
return (buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf3, 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), (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 torch import nn
class Maxout(nn.Module):
def __init__(self, in_features, out_features):
super(Maxout, self).__init__()
self.layer1 = nn.Linear(in_features, out_features)
self.layer2 = nn.Linear(in_features, out_features)
def forward(self, x):
output1 = self.layer1(x)
output2 = self.layer2(x)
return torch.max(output1, output2)
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 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_eq_gt_lt_maximum_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3,
out_ptr0, out_ptr1, out_ptr2, out_ptr3, 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')
tmp3 = tl.load(in_ptr2 + x2, xmask)
tmp4 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = triton_helpers.maximum(tmp2, tmp5)
tmp7 = tmp2 == tmp5
tmp8 = tmp2 > tmp5
tmp9 = tmp2 < tmp5
tl.store(out_ptr0 + x2, tmp6, xmask)
tl.store(out_ptr1 + x2, tmp7, xmask)
tl.store(out_ptr2 + x2, tmp8, xmask)
tl.store(out_ptr3 + x2, tmp9, 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 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf1)
del primals_4
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_eq_gt_lt_maximum_0[grid(256)](buf0, primals_2,
buf1, primals_5, buf2, buf3, buf4, buf5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del buf0
del buf1
del primals_2
del primals_5
return buf2, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf3, buf4, buf5
class MaxoutNew(nn.Module):
def __init__(self, in_features, out_features):
super(MaxoutNew, self).__init__()
self.layer1 = nn.Linear(in_features, out_features)
self.layer2 = nn.Linear(in_features, out_features)
def forward(self, input_0):
primals_1 = self.layer1.weight
primals_2 = self.layer1.bias
primals_4 = self.layer2.weight
primals_5 = self.layer2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| shadow2496/KAIST_2019_Deep-Learning_HW4 | Maxout | false | 4,302 | [
"MIT"
] | 0 | f837ee23816c7486952733925b1f338b54d7086f | https://github.com/shadow2496/KAIST_2019_Deep-Learning_HW4/tree/f837ee23816c7486952733925b1f338b54d7086f | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, in_features, out_features):
super().__init__()
self.layer1 = nn.Linear(in_features, out_features)
self.layer2 = nn.Linear(in_features, out_features)
def forward(self, x):
output1 = self.layer1(x)
output2 = self.layer2(x)
return torch.max(output1, output2)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
ResidualAttention | # 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_7/inductor_cache/u3/cu3litezfpnwhpnfnfuj6dtimz6ml42wmcwnwxlnovd4p5lvyin4.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=[2048, 4096], 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_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 = 2048
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = (yindex // 512)
tmp0 = tl.load(in_ptr0 + (x2 + (4096*y3)), None, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (512*x2) + (2097152*y1)), tmp0, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/x3/cx3zatlapixnvnybklv4hr3jf2bzbsxdhp4jqm5sowz7qy425ovw.py
# Topologically Sorted Source Nodes: [y_avg, max_1], Original ATen: [aten.mean, aten.max]
# Source node to ATen node mapping:
# max_1 => max_1
# y_avg => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view, [2]), kwargs = {})
# %max_1 : [num_users=2] = call_function[target=torch.ops.aten.max.dim](args = (%view, 2), kwargs = {})
triton_red_fused_max_mean_1 = async_compile.triton('triton_red_fused_max_mean_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.reduction(
size_hints=[131072, 128],
reduction_hint=ReductionHint.OUTER,
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_red_fused_max_mean_1', '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_red_fused_max_mean_1(in_ptr0, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 128000
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 1000
x1 = (xindex // 1000)
_tmp2 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
_tmp4 = tl.full([XBLOCK, RBLOCK], float("-inf"), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + (1000*r2) + (128000*x1)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = _tmp2 + tmp1
_tmp2 = tl.where(rmask & xmask, tmp3, _tmp2)
tmp5 = triton_helpers.maximum(_tmp4, tmp1)
_tmp4 = tl.where(rmask & xmask, tmp5, _tmp4)
tmp2 = tl.sum(_tmp2, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp2, xmask)
tmp4 = triton_helpers.max2(_tmp4, 1)[:, None]
tl.store(out_ptr1 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/pr/cpraieink233y5r33a6msz5ocnfafpb4445uks3glrkdxl5pbont.py
# Topologically Sorted Source Nodes: [y_avg, max_1, mul, score], Original ATen: [aten.mean, aten.max, aten.mul, aten.add]
# Source node to ATen node mapping:
# max_1 => max_1
# mul => mul
# score => add
# y_avg => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view, [2]), kwargs = {})
# %max_1 : [num_users=2] = call_function[target=torch.ops.aten.max.dim](args = (%view, 2), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%getitem, 0.2), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mean, %mul), kwargs = {})
triton_per_fused_add_max_mean_mul_2 = async_compile.triton('triton_per_fused_add_max_mean_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.persistent_reduction(
size_hints=[4096, 32],
reduction_hint=ReductionHint.OUTER,
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_add_max_mean_mul_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 2, '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_max_mean_mul_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4000
rnumel = 32
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 = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 1000
x1 = (xindex // 1000)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (1000*r2) + (32000*x1)), xmask, other=0.0)
tmp5 = tl.load(in_ptr1 + (x0 + (1000*r2) + (32000*x1)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, float("-inf"))
tmp9 = triton_helpers.max2(tmp8, 1)[:, None]
tmp10 = 4096.0
tmp11 = tmp4 / tmp10
tmp12 = 0.2
tmp13 = tmp9 * tmp12
tmp14 = tmp11 + tmp13
tl.debug_barrier()
tl.store(in_out_ptr0 + (x3), tmp14, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/7d/c7dc3ovmujmnb2r4l3y6koo326v5usfc6mthj2l6e4jfcvgqvcqe.py
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.max]
# Source node to ATen node mapping:
# max_1 => max_1
# Graph fragment:
# %max_1 : [num_users=2] = call_function[target=torch.ops.aten.max.dim](args = (%view, 2), kwargs = {})
triton_red_fused_max_3 = async_compile.triton('triton_red_fused_max_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.reduction(
size_hints=[4096, 4096],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 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_red_fused_max_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_red_fused_max_3(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 4000
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 1000
x1 = (xindex // 1000)
_tmp2 = tl.full([XBLOCK, RBLOCK], float("-inf"), tl.float32)
_tmp2_index = tl.full([XBLOCK, RBLOCK], 9223372036854775807, tl.int64)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + (1000*r2) + (4096000*x1)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
_tmp2_next, _tmp2_index_next = triton_helpers.maximum_with_index(
_tmp2, _tmp2_index, tmp1, rindex
)
_tmp2 = tl.where(rmask & xmask, _tmp2_next, _tmp2)
_tmp2_index = tl.where(rmask & xmask, _tmp2_index_next, _tmp2_index)
_, tmp2_tmp = triton_helpers.max_with_index(_tmp2, _tmp2_index, 1)
tmp2 = tmp2_tmp[:, None]
tl.store(out_ptr0 + (x3), tmp2, 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, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (1000, 512, 1, 1), (512, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(primals_1, buf0, 2048, 4096, grid=grid(2048, 4096), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [conv2d], 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, 1000, 64, 64), (4096000, 1, 64000, 1000))
buf2 = empty_strided_cuda((4, 1000, 32), (32000, 1, 1000), torch.float32)
buf4 = empty_strided_cuda((4, 1000, 32), (32000, 1, 1000), torch.float32)
# Topologically Sorted Source Nodes: [y_avg, max_1], Original ATen: [aten.mean, aten.max]
triton_red_fused_max_mean_1.run(buf1, buf2, buf4, 128000, 128, grid=grid(128000), stream=stream0)
buf3 = empty_strided_cuda((4, 1000), (1000, 1), torch.float32)
buf7 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [y_avg, max_1, mul, score], Original ATen: [aten.mean, aten.max, aten.mul, aten.add]
triton_per_fused_add_max_mean_mul_2.run(buf7, buf2, buf4, 4000, 32, grid=grid(4000), stream=stream0)
del buf2
del buf4
buf6 = empty_strided_cuda((4, 1000), (1000, 1), torch.int64)
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.max]
triton_red_fused_max_3.run(buf1, buf6, 4000, 4096, grid=grid(4000), stream=stream0)
del buf1
return (buf7, buf0, primals_2, reinterpret_tensor(buf6, (4, 1000, 1), (1000, 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, 512, 64, 64), (2097152, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((1000, 512, 1, 1), (512, 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
from torch import nn
class ResidualAttention(nn.Module):
def __init__(self, channel=512, num_class=1000, la=0.2):
super().__init__()
self.la = la
self.fc = nn.Conv2d(in_channels=channel, out_channels=num_class,
kernel_size=1, stride=1, bias=False)
def forward(self, x):
_b, _c, _h, _w = x.shape
y_raw = self.fc(x).flatten(2)
y_avg = torch.mean(y_raw, dim=2)
y_max = torch.max(y_raw, dim=2)[0]
score = y_avg + self.la * y_max
return score
def get_inputs():
return [torch.rand([4, 512, 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
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):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = yindex // 512
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y3), None, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 512 * x2 + 2097152 * y1), tmp0, None)
@triton.jit
def triton_red_fused_max_mean_1(in_ptr0, out_ptr0, out_ptr1, xnumel, rnumel,
XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 128000
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 1000
x1 = xindex // 1000
_tmp2 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
_tmp4 = tl.full([XBLOCK, RBLOCK], float('-inf'), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + 1000 * r2 + 128000 * x1), rmask &
xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = _tmp2 + tmp1
_tmp2 = tl.where(rmask & xmask, tmp3, _tmp2)
tmp5 = triton_helpers.maximum(_tmp4, tmp1)
_tmp4 = tl.where(rmask & xmask, tmp5, _tmp4)
tmp2 = tl.sum(_tmp2, 1)[:, None]
tl.store(out_ptr0 + x3, tmp2, xmask)
tmp4 = triton_helpers.max2(_tmp4, 1)[:, None]
tl.store(out_ptr1 + x3, tmp4, xmask)
@triton.jit
def triton_per_fused_add_max_mean_mul_2(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 4000
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, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 1000
x1 = xindex // 1000
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 1000 * r2 + 32000 * x1), xmask, other=0.0)
tmp5 = tl.load(in_ptr1 + (x0 + 1000 * r2 + 32000 * x1), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(xmask, tmp1, 0)
tmp4 = tl.sum(tmp3, 1)[:, None]
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp8 = tl.where(xmask, tmp6, float('-inf'))
tmp9 = triton_helpers.max2(tmp8, 1)[:, None]
tmp10 = 4096.0
tmp11 = tmp4 / tmp10
tmp12 = 0.2
tmp13 = tmp9 * tmp12
tmp14 = tmp11 + tmp13
tl.debug_barrier()
tl.store(in_out_ptr0 + x3, tmp14, xmask)
@triton.jit
def triton_red_fused_max_3(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK: tl.
constexpr, RBLOCK: tl.constexpr):
xnumel = 4000
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 1000
x1 = xindex // 1000
_tmp2 = tl.full([XBLOCK, RBLOCK], float('-inf'), tl.float32)
_tmp2_index = tl.full([XBLOCK, RBLOCK], 9223372036854775807, tl.int64)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + 1000 * r2 + 4096000 * x1), rmask &
xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
_tmp2_next, _tmp2_index_next = triton_helpers.maximum_with_index(_tmp2,
_tmp2_index, tmp1, rindex)
_tmp2 = tl.where(rmask & xmask, _tmp2_next, _tmp2)
_tmp2_index = tl.where(rmask & xmask, _tmp2_index_next, _tmp2_index)
_, tmp2_tmp = triton_helpers.max_with_index(_tmp2, _tmp2_index, 1)
tmp2 = tmp2_tmp[:, None]
tl.store(out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (1000, 512, 1, 1), (512, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512
), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(2048, 4096)](primals_1, buf0, 2048, 4096,
XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
del primals_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, 1000, 64, 64), (4096000, 1, 64000, 1000))
buf2 = empty_strided_cuda((4, 1000, 32), (32000, 1, 1000), torch.
float32)
buf4 = empty_strided_cuda((4, 1000, 32), (32000, 1, 1000), torch.
float32)
triton_red_fused_max_mean_1[grid(128000)](buf1, buf2, buf4, 128000,
128, XBLOCK=64, RBLOCK=8, num_warps=4, num_stages=1)
buf3 = empty_strided_cuda((4, 1000), (1000, 1), torch.float32)
buf7 = buf3
del buf3
triton_per_fused_add_max_mean_mul_2[grid(4000)](buf7, buf2, buf4,
4000, 32, XBLOCK=128, num_warps=8, num_stages=1)
del buf2
del buf4
buf6 = empty_strided_cuda((4, 1000), (1000, 1), torch.int64)
triton_red_fused_max_3[grid(4000)](buf1, buf6, 4000, 4096, XBLOCK=8,
RBLOCK=512, num_warps=16, num_stages=1)
del buf1
return buf7, buf0, primals_2, reinterpret_tensor(buf6, (4, 1000, 1), (
1000, 1, 1), 0)
class ResidualAttentionNew(nn.Module):
def __init__(self, channel=512, num_class=1000, la=0.2):
super().__init__()
self.la = la
self.fc = nn.Conv2d(in_channels=channel, out_channels=num_class,
kernel_size=1, stride=1, bias=False)
def forward(self, input_0):
primals_2 = self.fc.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| rushirajsherlocked/External-Attention-pytorch | ResidualAttention | false | 4,303 | [
"MIT"
] | 0 | 7d6814b2d90909adf81c62f3f8a89e30a59d6481 | https://github.com/rushirajsherlocked/External-Attention-pytorch/tree/7d6814b2d90909adf81c62f3f8a89e30a59d6481 | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, channel=512, num_class=1000, la=0.2):
super().__init__()
self.la = la
self.fc = nn.Conv2d(in_channels=channel, out_channels=num_class,
kernel_size=1, stride=1, bias=False)
def forward(self, x):
_b, _c, _h, _w = x.shape
y_raw = self.fc(x).flatten(2)
y_avg = torch.mean(y_raw, dim=2)
y_max = torch.max(y_raw, dim=2)[0]
score = y_avg + self.la * y_max
return score
def get_inputs():
return [torch.rand([4, 512, 64, 64])]
def get_init_inputs():
return []
|
LipSwish | # 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_7/inductor_cache/ou/cou4j63iugr7lynt26fs2xku6nxdvfa42t7nsy6ivzmygxmvbaab.py
# Topologically Sorted Source Nodes: [silu, mul], Original ATen: [aten.silu, aten.mul]
# Source node to ATen node mapping:
# mul => mul_1
# silu => mul, sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%arg0_1,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, %sigmoid), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, 0.909), kwargs = {})
triton_poi_fused_mul_silu_0 = async_compile.triton('triton_poi_fused_mul_silu_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_mul_silu_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_mul_silu_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.sigmoid(tmp0)
tmp2 = tmp0 * tmp1
tmp3 = 0.909
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x0), tmp4, 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: [silu, mul], Original ATen: [aten.silu, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_silu_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
class LipSwish(torch.nn.Module):
def forward(self, x):
return 0.909 * torch.nn.functional.silu(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
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_silu_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.sigmoid(tmp0)
tmp2 = tmp0 * tmp1
tmp3 = 0.909
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x0, tmp4, 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_mul_silu_0[grid(256)](arg0_1, buf0, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class LipSwishNew(torch.nn.Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| shi27feng/torchsde | LipSwish | false | 4,304 | [
"Apache-2.0"
] | 0 | 58105bb6b839766c1d27b73c4fe3f949869d7394 | https://github.com/shi27feng/torchsde/tree/58105bb6b839766c1d27b73c4fe3f949869d7394 | import torch
class Model(torch.nn.Module):
def forward(self, x):
return 0.909 * torch.nn.functional.silu(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
Actor | # 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_7/inductor_cache/nc/cncwsucylpsg2zmlivjfxu6vbd64ztxjndlsix2ysjtby3xohgk4.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# x => 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')
# kernel path: runs/run_shard_7/inductor_cache/e6/ce67gf2cewx4xm5ylnmnwaictyxryblw7teby2qpwzhpjtkbwdba.py
# Topologically Sorted Source Nodes: [std], Original ATen: [aten.exp]
# Source node to ATen node mapping:
# std => full_default
# Graph fragment:
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4, 4, 4], 1.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
triton_poi_fused_exp_1 = async_compile.triton('triton_poi_fused_exp_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: 'i32'}, 'device': 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_exp_1', '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_exp_1(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 = 1.0
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 = 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: [], 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], 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: [x_1], Original ATen: [aten.tanh]
triton_poi_fused_tanh_0.run(buf3, primals_5, 256, grid=grid(256), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mu], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 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: [std], Original ATen: [aten.exp]
triton_poi_fused_exp_1.run(buf5, 256, grid=grid(256), stream=stream0)
return (reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0), buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), 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, ), (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)
| from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
class Actor(nn.Module):
def __init__(self, num_inputs, num_outputs, args):
super(Actor, self).__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, num_outputs)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, x):
x = torch.tanh(self.fc1(x))
x = torch.tanh(self.fc2(x))
mu = self.fc3(x)
logstd = torch.zeros_like(mu)
std = torch.exp(logstd)
return mu, std
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_inputs': 4, 'num_outputs': 4, 'args': _mock_config(
hidden_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_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)
@triton.jit
def triton_poi_fused_exp_1(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 = 1.0
tl.store(out_ptr0 + x0, tmp0, 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,), (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.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=256,
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=256,
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, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 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_exp_1[grid(256)](buf5, 256, XBLOCK=256, num_warps=
4, num_stages=1)
return reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0
), buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf1, buf3, primals_6, primals_4
class ActorNew(nn.Module):
def __init__(self, num_inputs, num_outputs, args):
super(ActorNew, self).__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, num_outputs)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.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]
| sgrimbly/lets-do-irl | Actor | false | 4,305 | [
"MIT"
] | 0 | 4233e238342394feef6a7bd495cc6b700d435b00 | https://github.com/sgrimbly/lets-do-irl/tree/4233e238342394feef6a7bd495cc6b700d435b00 | from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, num_inputs, num_outputs, args):
super().__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, num_outputs)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, x):
x = torch.tanh(self.fc1(x))
x = torch.tanh(self.fc2(x))
mu = self.fc3(x)
logstd = torch.zeros_like(mu)
std = torch.exp(logstd)
return mu, std
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_inputs': 4, 'num_outputs': 4, 'args': _mock_config(
hidden_size=4)}]
|
Encoder1 | # 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_7/inductor_cache/i3/ci3txaej5ebehjq4jgfqzwc2gfqiohduzmbk6fccpefhj2kwy6bv.py
# Topologically Sorted Source Nodes: [conv2d, out], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# out => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [2, 2], [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=[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 = 131072
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 1024) % 32
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_7/inductor_cache/cy/ccyojkpx43k5pgtwa2byecil57lweiyjjkqrsv23rii3oqnm6fk4.py
# Topologically Sorted Source Nodes: [conv2d_1, out_1], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_1 => convolution_1
# out_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], [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_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 = 32768
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 256) % 32
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_7/inductor_cache/sa/csau44x3bjeb576lxxwb7f3ekmdgv2qxtvkceluute2s36ivj6kc.py
# Topologically Sorted Source Nodes: [conv2d_2, out_2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_2 => convolution_2
# out_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], [1, 1], [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 = 16384
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 64) % 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_7/inductor_cache/rn/crnor7cn4o43gsx7osh4pwgjrowsclmfxirhbe7qayoobmyw5syx.py
# Topologically Sorted Source Nodes: [conv2d_3, out_3], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_3 => convolution_3
# out_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], [1, 1], [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_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=[4096],
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 = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = (xindex // 16) % 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)
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_7/inductor_cache/3d/c3daizw6k7n3mdqdhvpifdxm2baxmazxb7opdgejzyeaavnbkn3d.py
# Topologically Sorted Source Nodes: [out_5], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# out_5 => relu_4
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_11), kwargs = {})
# %relu_4 : [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=[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_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 = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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')
# kernel path: runs/run_shard_7/inductor_cache/cg/ccgc5tmknbetkxsi5ql5eooiqvexxjdrldgfzy34k27kp2mtzh4z.py
# Topologically Sorted Source Nodes: [exp, std], Original ATen: [aten.exp, aten.sqrt]
# Source node to ATen node mapping:
# exp => exp
# std => sqrt
# Graph fragment:
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%slice_4,), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%exp,), kwargs = {})
triton_poi_fused_exp_sqrt_5 = async_compile.triton('triton_poi_fused_exp_sqrt_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: 'i32'}, 'device': 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_exp_sqrt_5', '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_exp_sqrt_5(in_ptr0, 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 + (10 + x0 + (20*x1)), xmask)
tmp1 = tl_math.exp(tmp0)
tmp2 = libdevice.sqrt(tmp1)
tl.store(out_ptr0 + (x2), 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, primals_7, primals_8, primals_9, primals_10, primals_11, primals_12, primals_13 = args
args.clear()
assert_size_stride(primals_1, (32, 1, 4, 4), (16, 16, 4, 1))
assert_size_stride(primals_2, (32, ), (1, ))
assert_size_stride(primals_3, (4, 1, 64, 64), (4096, 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, (128, 1024), (1024, 1))
assert_size_stride(primals_11, (128, ), (1, ))
assert_size_stride(primals_12, (20, 128), (128, 1))
assert_size_stride(primals_13, (20, ), (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=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 32, 32, 32), (32768, 1024, 32, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d, out], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 131072, grid=grid(131072), 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=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 32, 16, 16), (8192, 256, 16, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, out_1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_1.run(buf3, primals_5, 32768, grid=grid(32768), 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=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 8, 8), (4096, 64, 8, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [conv2d_2, out_2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_2.run(buf5, primals_7, 16384, grid=grid(16384), 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=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 64, 4, 4), (1024, 16, 4, 1))
buf7 = buf6; del buf6 # reuse
buf12 = empty_strided_cuda((4, 64, 4, 4), (1024, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_3, out_3], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_3.run(buf7, primals_9, buf12, 4096, grid=grid(4096), stream=stream0)
del primals_9
buf8 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf7, (4, 1024), (1024, 1), 0), reinterpret_tensor(primals_10, (1024, 128), (1, 1024), 0), out=buf8)
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [out_5], Original ATen: [aten.relu]
triton_poi_fused_relu_4.run(buf9, primals_11, 512, grid=grid(512), stream=stream0)
del primals_11
buf10 = empty_strided_cuda((4, 20), (20, 1), torch.float32)
# Topologically Sorted Source Nodes: [stats], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_13, buf9, reinterpret_tensor(primals_12, (128, 20), (1, 128), 0), alpha=1, beta=1, out=buf10)
del primals_13
buf11 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
# Topologically Sorted Source Nodes: [exp, std], Original ATen: [aten.exp, aten.sqrt]
triton_poi_fused_exp_sqrt_5.run(buf10, buf11, 40, grid=grid(40), stream=stream0)
return (reinterpret_tensor(buf10, (4, 10), (20, 1), 0), buf11, reinterpret_tensor(buf10, (4, 10), (20, 1), 10), buf10, primals_1, primals_3, primals_4, primals_6, primals_8, buf1, buf3, buf5, reinterpret_tensor(buf7, (4, 1024), (1024, 1), 0), buf9, reinterpret_tensor(buf10, (4, 10), (20, 1), 10), buf11, primals_12, primals_10, buf12, )
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, 1, 4, 4), (16, 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, 1, 64, 64), (4096, 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((128, 1024), (1024, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((20, 128), (128, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((20, ), (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 torch
import torch.nn as nn
import torch.nn.init as init
import torch.nn.functional as F
def kaiming_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.kaiming_normal_(m.weight)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
def normal_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.normal_(m.weight, 0, 0.02)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
class Encoder1(nn.Module):
"""
encoder architecture for the "dsprites" data
"""
def __init__(self, z_dim=10):
super(Encoder1, self).__init__()
self.z_dim = z_dim
self.conv1 = nn.Conv2d(1, 32, 4, 2, 1)
self.conv2 = nn.Conv2d(32, 32, 4, 2, 1)
self.conv3 = nn.Conv2d(32, 64, 4, 2, 1)
self.conv4 = nn.Conv2d(64, 64, 4, 2, 1)
self.fc5 = nn.Linear(64 * 4 * 4, 128)
self.fc6 = nn.Linear(128, 2 * z_dim)
self.weight_init()
def weight_init(self, mode='normal'):
if mode == 'kaiming':
initializer = kaiming_init
elif mode == 'normal':
initializer = normal_init
for m in self._modules:
initializer(self._modules[m])
def forward(self, x):
out = F.relu(self.conv1(x))
out = F.relu(self.conv2(out))
out = F.relu(self.conv3(out))
out = F.relu(self.conv4(out))
out = out.view(out.size(0), -1)
out = F.relu(self.fc5(out))
stats = self.fc6(out)
mu = stats[:, :self.z_dim]
logvar = stats[:, self.z_dim:]
std = torch.sqrt(torch.exp(logvar))
return mu, std, logvar
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._inductor.runtime.triton_helpers import libdevice, math as tl_math
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_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 // 1024 % 32
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_convolution_relu_1(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 // 256 % 32
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_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 // 64 % 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_convolution_relu_threshold_backward_3(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 // 16 % 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)
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_relu_4(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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)
@triton.jit
def triton_poi_fused_exp_sqrt_5(in_ptr0, 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 + (10 + x0 + 20 * x1), xmask)
tmp1 = tl_math.exp(tmp0)
tmp2 = libdevice.sqrt(tmp1)
tl.store(out_ptr0 + x2, tmp2, 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, (32, 1, 4, 4), (16, 16, 4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 1, 64, 64), (4096, 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, (128, 1024), (1024, 1))
assert_size_stride(primals_11, (128,), (1,))
assert_size_stride(primals_12, (20, 128), (128, 1))
assert_size_stride(primals_13, (20,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(2,
2), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 32, 32, 32), (32768, 1024, 32, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(131072)](buf1, primals_2,
131072, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 32, 16, 16), (8192, 256, 16, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_1[grid(32768)](buf3, primals_5,
32768, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 8, 8), (4096, 64, 8, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_2[grid(16384)](buf5, primals_7,
16384, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(2, 2),
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, 16, 4, 1))
buf7 = buf6
del buf6
buf12 = empty_strided_cuda((4, 64, 4, 4), (1024, 16, 4, 1), torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_3[grid(4096)](buf7
, primals_9, buf12, 4096, XBLOCK=128, num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (4, 1024), (1024, 1), 0),
reinterpret_tensor(primals_10, (1024, 128), (1, 1024), 0), out=buf8
)
buf9 = buf8
del buf8
triton_poi_fused_relu_4[grid(512)](buf9, primals_11, 512, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_11
buf10 = empty_strided_cuda((4, 20), (20, 1), torch.float32)
extern_kernels.addmm(primals_13, buf9, reinterpret_tensor(
primals_12, (128, 20), (1, 128), 0), alpha=1, beta=1, out=buf10)
del primals_13
buf11 = empty_strided_cuda((4, 10), (10, 1), torch.float32)
triton_poi_fused_exp_sqrt_5[grid(40)](buf10, buf11, 40, XBLOCK=64,
num_warps=1, num_stages=1)
return (reinterpret_tensor(buf10, (4, 10), (20, 1), 0), buf11,
reinterpret_tensor(buf10, (4, 10), (20, 1), 10), buf10, primals_1,
primals_3, primals_4, primals_6, primals_8, buf1, buf3, buf5,
reinterpret_tensor(buf7, (4, 1024), (1024, 1), 0), buf9,
reinterpret_tensor(buf10, (4, 10), (20, 1), 10), buf11, primals_12,
primals_10, buf12)
def kaiming_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.kaiming_normal_(m.weight)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
def normal_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.normal_(m.weight, 0, 0.02)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
class Encoder1New(nn.Module):
"""
encoder architecture for the "dsprites" data
"""
def __init__(self, z_dim=10):
super(Encoder1New, self).__init__()
self.z_dim = z_dim
self.conv1 = nn.Conv2d(1, 32, 4, 2, 1)
self.conv2 = nn.Conv2d(32, 32, 4, 2, 1)
self.conv3 = nn.Conv2d(32, 64, 4, 2, 1)
self.conv4 = nn.Conv2d(64, 64, 4, 2, 1)
self.fc5 = nn.Linear(64 * 4 * 4, 128)
self.fc6 = nn.Linear(128, 2 * z_dim)
self.weight_init()
def weight_init(self, mode='normal'):
if mode == 'kaiming':
initializer = kaiming_init
elif mode == 'normal':
initializer = normal_init
for m in self._modules:
initializer(self._modules[m])
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.fc5.weight
primals_11 = self.fc5.bias
primals_12 = self.fc6.weight
primals_13 = self.fc6.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])
return output[0], output[1], output[2]
| seqam-lab/rfvae | Encoder1 | false | 4,306 | [
"MIT"
] | 0 | 07089e2cca6d51f305731750c2c67b83a42df12a | https://github.com/seqam-lab/rfvae/tree/07089e2cca6d51f305731750c2c67b83a42df12a | import torch
import torch.nn as nn
import torch.nn.init as init
import torch.nn.functional as F
def kaiming_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.kaiming_normal_(m.weight)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
def normal_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.normal_(m.weight, 0, 0.02)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
class Model(nn.Module):
"""
encoder architecture for the "dsprites" data
"""
def __init__(self, z_dim=10):
super().__init__()
self.z_dim = z_dim
self.conv1 = nn.Conv2d(1, 32, 4, 2, 1)
self.conv2 = nn.Conv2d(32, 32, 4, 2, 1)
self.conv3 = nn.Conv2d(32, 64, 4, 2, 1)
self.conv4 = nn.Conv2d(64, 64, 4, 2, 1)
self.fc5 = nn.Linear(64 * 4 * 4, 128)
self.fc6 = nn.Linear(128, 2 * z_dim)
self.weight_init()
def weight_init(self, mode='normal'):
if mode == 'kaiming':
initializer = kaiming_init
elif mode == 'normal':
initializer = normal_init
for m in self._modules:
initializer(self._modules[m])
def forward(self, x):
out = F.relu(self.conv1(x))
out = F.relu(self.conv2(out))
out = F.relu(self.conv3(out))
out = F.relu(self.conv4(out))
out = out.view(out.size(0), -1)
out = F.relu(self.fc5(out))
stats = self.fc6(out)
mu = stats[:, :self.z_dim]
logvar = stats[:, self.z_dim:]
std = torch.sqrt(torch.exp(logvar))
return mu, std, logvar
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return []
|
SelfAttention | # 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_7/inductor_cache/7e/c7eul75w5s2nxzr6mevxwrjuv2vslc5jmikyp4moxhc5kqk225qk.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=[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_tanh_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_tanh_0(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 = libdevice.tanh(tmp0)
tl.store(in_out_ptr0 + (x0), tmp1, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/dm/cdmkcxuzpnailvibeivaikqdr4zvashgzwju7qijhq5aizlo3aor.py
# Topologically Sorted Source Nodes: [out_4], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# out_4 => 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, [2], 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_7/inductor_cache/ot/cotghubvxq6i3kdxijgdhem4ntfmdc5cn3al7gv2kuqi2pbqiyxe.py
# Topologically Sorted Source Nodes: [out_4], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# out_4 => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [2], True), kwargs = {})
# %div : [num_users=1] = 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 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (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: [out], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_2, (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: [out_1], Original ATen: [aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_tanh_0.run(buf1, 64, grid=grid(64), stream=stream0)
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_2], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4, 4), (16, 1, 4), torch.float32)
# Topologically Sorted Source Nodes: [out_4], 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: [out_4], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf3, buf4, 16, 4, grid=grid(16, 4), stream=stream0)
del buf3
return (buf4, reinterpret_tensor(primals_2, (16, 4), (4, 1), 0), buf1, buf4, 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, 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, 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 torch.nn import functional as F
class SelfAttention(nn.Module):
"""
Implementation of the attention block
"""
def __init__(self, input_size, hidden_size, output_size):
super(SelfAttention, self).__init__()
self.layer1 = nn.Linear(input_size, hidden_size, bias=False)
self.layer2 = nn.Linear(hidden_size, output_size, bias=False)
def forward(self, attention_input):
out = self.layer1(attention_input)
out = torch.tanh(out)
out = self.layer2(out)
out = out.permute(0, 2, 1)
out = F.softmax(out, dim=2)
return out
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'input_size': 4, 'hidden_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
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, 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 = libdevice.tanh(tmp0)
tl.store(in_out_ptr0 + x0, tmp1, 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 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (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_2, (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, 64, XBLOCK=64, num_warps=1,
num_stages=1)
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_3, (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)
del buf3
return buf4, reinterpret_tensor(primals_2, (16, 4), (4, 1), 0
), buf1, buf4, primals_3
class SelfAttentionNew(nn.Module):
"""
Implementation of the attention block
"""
def __init__(self, input_size, hidden_size, output_size):
super(SelfAttentionNew, self).__init__()
self.layer1 = nn.Linear(input_size, hidden_size, bias=False)
self.layer2 = nn.Linear(hidden_size, output_size, bias=False)
def forward(self, input_0):
primals_1 = self.layer1.weight
primals_3 = self.layer2.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| shahrukhx01/model_serve_pytorch | SelfAttention | false | 4,307 | [
"MIT"
] | 0 | c97ab45264b41ce349828e8b230ed85a51d6b213 | https://github.com/shahrukhx01/model_serve_pytorch/tree/c97ab45264b41ce349828e8b230ed85a51d6b213 | import torch
import torch.nn as nn
from torch.nn import functional as F
class Model(nn.Module):
"""
Implementation of the attention block
"""
def __init__(self, input_size, hidden_size, output_size):
super().__init__()
self.layer1 = nn.Linear(input_size, hidden_size, bias=False)
self.layer2 = nn.Linear(hidden_size, output_size, bias=False)
def forward(self, attention_input):
out = self.layer1(attention_input)
out = torch.tanh(out)
out = self.layer2(out)
out = out.permute(0, 2, 1)
out = F.softmax(out, dim=2)
return out
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
Discriminator | # 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_7/inductor_cache/nc/cncwsucylpsg2zmlivjfxu6vbd64ztxjndlsix2ysjtby3xohgk4.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# x => 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')
# kernel path: runs/run_shard_7/inductor_cache/7z/c7zsuucunqdovb2xa6tywxjxwmolzjzdk72ratro7fi3qvgyqb7c.py
# Topologically Sorted Source Nodes: [prob], Original ATen: [aten.sigmoid]
# Source node to ATen node mapping:
# prob => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_5,), kwargs = {})
triton_poi_fused_sigmoid_1 = async_compile.triton('triton_poi_fused_sigmoid_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_sigmoid_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_sigmoid_1(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
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr0 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.sigmoid(tmp3)
tl.store(in_out_ptr0 + (x0), 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, ), (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, (1, 4), (4, 1))
assert_size_stride(primals_7, (1, ), (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], 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: [x_1], Original ATen: [aten.tanh]
triton_poi_fused_tanh_0.run(buf3, primals_5, 256, grid=grid(256), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 1), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [prob], Original ATen: [aten.sigmoid]
triton_poi_fused_sigmoid_1.run(buf5, primals_7, 64, grid=grid(64), stream=stream0)
del primals_7
return (buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, buf3, buf5, 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, ), (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((1, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = 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])
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 torch.nn as nn
class Discriminator(nn.Module):
def __init__(self, num_inputs, args):
super(Discriminator, self).__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, 1)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, x):
x = torch.tanh(self.fc1(x))
x = torch.tanh(self.fc2(x))
prob = torch.sigmoid(self.fc3(x))
return prob
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_inputs': 4, 'args': _mock_config(hidden_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_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)
@triton.jit
def triton_poi_fused_sigmoid_1(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
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.sigmoid(tmp3)
tl.store(in_out_ptr0 + x0, 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,), (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, (1, 4), (4, 1))
assert_size_stride(primals_7, (1,), (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=256,
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=256,
num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 1), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf4
triton_poi_fused_sigmoid_1[grid(64)](buf5, primals_7, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_7
return buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf1, buf3, buf5, primals_6, primals_4
class DiscriminatorNew(nn.Module):
def __init__(self, num_inputs, args):
super(DiscriminatorNew, self).__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, 1)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| sgrimbly/lets-do-irl | Discriminator | false | 4,308 | [
"MIT"
] | 0 | 4233e238342394feef6a7bd495cc6b700d435b00 | https://github.com/sgrimbly/lets-do-irl/tree/4233e238342394feef6a7bd495cc6b700d435b00 | from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, num_inputs, args):
super().__init__()
self.fc1 = nn.Linear(num_inputs, args.hidden_size)
self.fc2 = nn.Linear(args.hidden_size, args.hidden_size)
self.fc3 = nn.Linear(args.hidden_size, 1)
self.fc3.weight.data.mul_(0.1)
self.fc3.bias.data.mul_(0.0)
def forward(self, x):
x = torch.tanh(self.fc1(x))
x = torch.tanh(self.fc2(x))
prob = torch.sigmoid(self.fc3(x))
return prob
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
Attention | # 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_7/inductor_cache/ch/cchblwxibhbejgpezaajsszmynwayrtalecnd5ksg22infzh2v3o.py
# Topologically Sorted Source Nodes: [dot, dot_1], Original ATen: [aten.add, aten.tanh]
# Source node to ATen node mapping:
# dot => add
# dot_1 => tanh
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view, %expand), kwargs = {})
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%add,), kwargs = {})
triton_poi_fused_add_tanh_0 = async_compile.triton('triton_poi_fused_add_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: '*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_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_tanh_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
x3 = xindex
x0 = xindex % 4
x2 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr1 + (x0 + (4*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tmp5 = libdevice.tanh(tmp4)
tl.store(out_ptr0 + (x3), tmp5, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/pw/cpw7p22sygsv72j67klg2k3xeokdqrvalgmkjpkzpqpj5pv4xmkc.py
# Topologically Sorted Source Nodes: [weight], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# weight => amax, div, exp, sub, sum_1
# Graph fragment:
# %amax : [num_users=2] = call_function[target=torch.ops.aten.amax.default](args = (%view_2, [1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_2, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub,), kwargs = {})
# %sum_1 : [num_users=2] = 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_per_fused__softmax_1 = async_compile.triton('triton_per_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.persistent_reduction(
size_hints=[4, 16],
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__softmax_1', '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_1(in_ptr0, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
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, 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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tmp6 / tmp10
tl.store(out_ptr2 + (r1 + (16*x0)), tmp11, xmask)
tl.store(out_ptr0 + (x0), tmp4, xmask)
tl.store(out_ptr1 + (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, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (1, 4), (4, 1))
assert_size_stride(primals_7, (1, ), (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_5, reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), out=buf0)
del primals_3
buf1 = empty_strided_cuda((4, 16, 4), (64, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [dot, dot_1], Original ATen: [aten.add, aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_add_tanh_0.run(primals_2, buf0, primals_4, buf1, 256, grid=grid(256), stream=stream0)
del primals_2
del primals_4
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [dot_3], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 1), (1, 4), 0), alpha=1, beta=1, out=buf3)
del primals_7
buf4 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
buf5 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
buf6 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [weight], Original ATen: [aten._softmax]
triton_per_fused__softmax_1.run(buf3, buf4, buf5, buf6, 4, 16, grid=grid(4), stream=stream0)
buf7 = reinterpret_tensor(buf0, (4, 1, 4), (4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [bmm], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 1, 16), (16, 0, 1), 0), reinterpret_tensor(primals_1, (4, 16, 4), (64, 4, 1), 0), out=buf7)
del buf6
return (reinterpret_tensor(buf7, (4, 4), (4, 1), 0), primals_5, buf1, buf3, buf4, buf5, reinterpret_tensor(primals_1, (4, 4, 16), (64, 1, 4), 0), 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, 4), (64, 16, 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)
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, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((1, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_7 = 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])
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 torch.nn as nn
import torch.nn.functional as F
from torch.autograd import *
class Attention(nn.Module):
def __init__(self, opt):
super(Attention, self).__init__()
self.rnn_size = opt.rnn_size
self.att_hid_size = opt.att_hid_size
self.h2att = nn.Linear(self.rnn_size, self.att_hid_size)
self.alpha_net = nn.Linear(self.att_hid_size, 1)
def forward(self, h, att_feats, p_att_feats, att_masks=None):
att_size = att_feats.numel() // att_feats.size(0) // att_feats.size(-1)
att = p_att_feats.view(-1, att_size, self.att_hid_size)
att_h = self.h2att(h)
att_h = att_h.unsqueeze(1).expand_as(att)
dot = att + att_h
dot = torch.tanh(dot)
dot = dot.view(-1, self.att_hid_size)
dot = self.alpha_net(dot)
dot = dot.view(-1, att_size)
weight = F.softmax(dot, dim=1)
if att_masks is not None:
weight = weight * att_masks.view(-1, att_size)
weight = weight / weight.sum(1, keepdim=True)
att_feats_ = att_feats.view(-1, att_size, att_feats.size(-1))
att_res = torch.bmm(weight.unsqueeze(1), att_feats_).squeeze(1)
return att_res
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4, 4, 4]), torch.rand([4, 4,
4, 4])]
def get_init_inputs():
return [[], {'opt': _mock_config(rnn_size=4, att_hid_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 libdevice, math as tl_math
import torch.nn as nn
from torch.autograd 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_add_tanh_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
x3 = xindex
x0 = xindex % 4
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + (x0 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tmp5 = libdevice.tanh(tmp4)
tl.store(out_ptr0 + x3, tmp5, xmask)
@triton.jit
def triton_per_fused__softmax_1(in_ptr0, out_ptr0, out_ptr1, out_ptr2,
xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 4
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, 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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tmp6 / tmp10
tl.store(out_ptr2 + (r1 + 16 * x0), tmp11, xmask)
tl.store(out_ptr0 + x0, tmp4, xmask)
tl.store(out_ptr1 + 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, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (1, 4), (4, 1))
assert_size_stride(primals_7, (1,), (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_5, reinterpret_tensor(primals_3, (4, 4),
(1, 4), 0), out=buf0)
del primals_3
buf1 = empty_strided_cuda((4, 16, 4), (64, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_tanh_0[grid(256)](primals_2, buf0, primals_4,
buf1, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
del primals_4
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 1), (1, 4), 0),
alpha=1, beta=1, out=buf3)
del primals_7
buf4 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
buf5 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
buf6 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
triton_per_fused__softmax_1[grid(4)](buf3, buf4, buf5, buf6, 4, 16,
XBLOCK=1, num_warps=2, num_stages=1)
buf7 = reinterpret_tensor(buf0, (4, 1, 4), (4, 4, 1), 0)
del buf0
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 1, 16), (16, 0, 1),
0), reinterpret_tensor(primals_1, (4, 16, 4), (64, 4, 1), 0),
out=buf7)
del buf6
return reinterpret_tensor(buf7, (4, 4), (4, 1), 0
), primals_5, buf1, buf3, buf4, buf5, reinterpret_tensor(primals_1,
(4, 4, 16), (64, 1, 4), 0), primals_6
class AttentionNew(nn.Module):
def __init__(self, opt):
super(AttentionNew, self).__init__()
self.rnn_size = opt.rnn_size
self.att_hid_size = opt.att_hid_size
self.h2att = nn.Linear(self.rnn_size, self.att_hid_size)
self.alpha_net = nn.Linear(self.att_hid_size, 1)
def forward(self, input_0, input_1, input_2):
primals_3 = self.h2att.weight
primals_4 = self.h2att.bias
primals_6 = self.alpha_net.weight
primals_7 = self.alpha_net.bias
primals_5 = input_0
primals_1 = input_1
primals_2 = input_2
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| Romero027/ImageCaptioning.pytorch | Attention | false | 4,309 | [
"MIT"
] | 0 | 069c95f5d343fb126afa8b10ec18e472f30b7b35 | https://github.com/Romero027/ImageCaptioning.pytorch/tree/069c95f5d343fb126afa8b10ec18e472f30b7b35 | from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import *
class Model(nn.Module):
def __init__(self, opt):
super().__init__()
self.rnn_size = opt.rnn_size
self.att_hid_size = opt.att_hid_size
self.h2att = nn.Linear(self.rnn_size, self.att_hid_size)
self.alpha_net = nn.Linear(self.att_hid_size, 1)
def forward(self, h, att_feats, p_att_feats, att_masks=None):
att_size = att_feats.numel() // att_feats.size(0) // att_feats.size(-1)
att = p_att_feats.view(-1, att_size, self.att_hid_size)
att_h = self.h2att(h)
att_h = att_h.unsqueeze(1).expand_as(att)
dot = att + att_h
dot = torch.tanh(dot)
dot = dot.view(-1, self.att_hid_size)
dot = self.alpha_net(dot)
dot = dot.view(-1, att_size)
weight = F.softmax(dot, dim=1)
if att_masks is not None:
weight = weight * att_masks.view(-1, att_size)
weight = weight / weight.sum(1, keepdim=True)
att_feats_ = att_feats.view(-1, att_size, att_feats.size(-1))
att_res = torch.bmm(weight.unsqueeze(1), att_feats_).squeeze(1)
return att_res
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4, 4, 4]), torch.rand([4, 4,
4, 4])]
def get_init_inputs():
return []
|
DeltaGFit | # 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_7/inductor_cache/fx/cfxn3ej4g35nq4vgpzy7xhfeoum7tvqqyq2tkeq66js476yni7bi.py
# Topologically Sorted Source Nodes: [mul, truediv, pfact, add, truediv_1], Original ATen: [aten.mul, aten.reciprocal, aten.exp, aten.add, aten.div]
# Source node to ATen node mapping:
# add => add
# mul => mul
# pfact => exp
# truediv => mul_1, reciprocal
# truediv_1 => div
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 8.31446261815324), kwargs = {})
# %reciprocal : [num_users=1] = call_function[target=torch.ops.aten.reciprocal.default](args = (%mul,), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%reciprocal, 4), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%mul_1,), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%exp, 1), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%arg1_1, %add), kwargs = {})
triton_poi_fused_add_div_exp_mul_reciprocal_0 = async_compile.triton('triton_poi_fused_add_div_exp_mul_reciprocal_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_add_div_exp_mul_reciprocal_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_div_exp_mul_reciprocal_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 = 8.31446261815324
tmp3 = tmp1 * tmp2
tmp4 = tl.full([1], 1, tl.int32)
tmp5 = tmp4 / tmp3
tmp6 = 4.0
tmp7 = tmp5 * tmp6
tmp8 = tl_math.exp(tmp7)
tmp9 = 1.0
tmp10 = tmp8 + tmp9
tmp11 = tmp0 / tmp10
tl.store(out_ptr0 + (x0), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/5c/c5cr5fuoorl33qtodhdoqnduxl2pp7cb7he3yazbqd5l2nd6mvh3.py
# Topologically Sorted Source Nodes: [neg, exp_1, uptake], Original ATen: [aten.neg, aten.exp, aten.rsub]
# Source node to ATen node mapping:
# exp_1 => exp_1
# neg => neg
# uptake => sub
# Graph fragment:
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%view_2,), kwargs = {})
# %exp_1 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%neg,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %exp_1), kwargs = {})
triton_poi_fused_exp_neg_rsub_1 = async_compile.triton('triton_poi_fused_exp_neg_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.pointwise(
size_hints=[256],
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_exp_neg_rsub_1', '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_exp_neg_rsub_1(in_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 = -tmp0
tmp2 = tl_math.exp(tmp1)
tmp3 = 1.0
tmp4 = tmp3 - tmp2
tl.store(in_out_ptr0 + (x0), tmp4, xmask)
''', 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)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul, truediv, pfact, add, truediv_1], Original ATen: [aten.mul, aten.reciprocal, aten.exp, aten.add, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_exp_mul_reciprocal_0.run(arg1_1, arg0_1, buf0, 256, grid=grid(256), stream=stream0)
del arg0_1
del arg1_1
buf1 = 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(buf0, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf1)
del arg2_1
buf2 = reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [neg, exp_1, uptake], Original ATen: [aten.neg, aten.exp, aten.rsub]
triton_poi_fused_exp_neg_rsub_1.run(buf2, 256, grid=grid(256), stream=stream0)
buf3 = reinterpret_tensor(buf0, (16, 4, 4), (16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [matmul_1], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(arg3_1, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1), 0), out=buf3)
del arg3_1
del buf2
return (reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 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
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
from scipy import constants
import torch.nn as nn
import torch as t
class DeltaGFit(nn.Module):
def __init__(self, deltaG):
super(DeltaGFit, self).__init__()
self.deltaG = deltaG
def forward(self, temperature, X, k_int, timepoints):
"""
# inputs, list of:
temperatures: scalar (1,)
X (N_peptides, N_residues)
k_int: (N_peptides, 1)
"""
pfact = t.exp(self.deltaG / (constants.R * temperature))
uptake = 1 - t.exp(-t.matmul(k_int / (1 + pfact), timepoints))
return t.matmul(X, uptake)
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 [[], {'deltaG': 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 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_add_div_exp_mul_reciprocal_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 = 8.31446261815324
tmp3 = tmp1 * tmp2
tmp4 = tl.full([1], 1, tl.int32)
tmp5 = tmp4 / tmp3
tmp6 = 4.0
tmp7 = tmp5 * tmp6
tmp8 = tl_math.exp(tmp7)
tmp9 = 1.0
tmp10 = tmp8 + tmp9
tmp11 = tmp0 / tmp10
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused_exp_neg_rsub_1(in_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 = -tmp0
tmp2 = tl_math.exp(tmp1)
tmp3 = 1.0
tmp4 = tmp3 - tmp2
tl.store(in_out_ptr0 + x0, tmp4, xmask)
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)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_exp_mul_reciprocal_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((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf0, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(arg2_1, (16, 4, 4), (16, 4, 1), 0), out=buf1
)
del arg2_1
buf2 = reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf1
triton_poi_fused_exp_neg_rsub_1[grid(256)](buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf3 = reinterpret_tensor(buf0, (16, 4, 4), (16, 4, 1), 0)
del buf0
extern_kernels.bmm(reinterpret_tensor(arg3_1, (16, 4, 4), (16, 4, 1
), 0), reinterpret_tensor(buf2, (16, 4, 4), (16, 4, 1), 0), out
=buf3)
del arg3_1
del buf2
return reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0),
class DeltaGFitNew(nn.Module):
def __init__(self, deltaG):
super(DeltaGFitNew, self).__init__()
self.deltaG = deltaG
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]
| sajetan/PyHDX | DeltaGFit | false | 4,310 | [
"MIT"
] | 0 | f764849e33b2dd1bcae5824795a38c64ef01e13c | https://github.com/sajetan/PyHDX/tree/f764849e33b2dd1bcae5824795a38c64ef01e13c | import torch
from scipy import constants
import torch.nn as nn
import torch as t
class Model(nn.Module):
def __init__(self, deltaG):
super().__init__()
self.deltaG = deltaG
def forward(self, temperature, X, k_int, timepoints):
"""
# inputs, list of:
temperatures: scalar (1,)
X (N_peptides, N_residues)
k_int: (N_peptides, 1)
"""
pfact = t.exp(self.deltaG / (constants.R * temperature))
uptake = 1 - t.exp(-t.matmul(k_int / (1 + pfact), timepoints))
return t.matmul(X, uptake)
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 [4]
|
SequentialPolarizedSelfAttention | # 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_7/inductor_cache/u3/cu3litezfpnwhpnfnfuj6dtimz6ml42wmcwnwxlnovd4p5lvyin4.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=[2048, 4096], 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_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 = 2048
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = (yindex // 512)
tmp0 = tl.load(in_ptr0 + (x2 + (4096*y3)), None, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (512*x2) + (2097152*y1)), tmp0, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/iq/ciqt7v5dvn4ukgyyy6maxezu54tu2k6htitfxylgvyziqck3gwix.py
# Topologically Sorted Source Nodes: [channel_wq_2], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# channel_wq_2 => amax, div, exp, sub, sum_1
# 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 = {})
# %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_red_fused__softmax_1 = async_compile.triton('triton_red_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.reduction(
size_hints=[4, 4096],
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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused__softmax_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, '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_red_fused__softmax_1(in_ptr0, in_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 4
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
_tmp5 = tl.full([XBLOCK, RBLOCK], float("-inf"), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_last', other=0.0)
tmp3 = tmp0 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = triton_helpers.maximum(_tmp5, tmp4)
_tmp5 = tl.where(rmask & xmask, tmp6, _tmp5)
tmp5 = triton_helpers.max2(_tmp5, 1)[:, None]
tmp8 = tl.load(in_ptr1 + (0))
tmp9 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
_tmp14 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp7 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tmp7 + tmp9
tmp11 = tmp10 - tmp5
tmp12 = tl_math.exp(tmp11)
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = _tmp14 + tmp13
_tmp14 = tl.where(rmask & xmask, tmp15, _tmp14)
tmp14 = tl.sum(_tmp14, 1)[:, None]
tmp17 = tl.load(in_ptr1 + (0))
tmp18 = tl.broadcast_to(tmp17, [XBLOCK, RBLOCK])
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp16 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp19 = tmp16 + tmp18
tmp20 = tmp19 - tmp5
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp21 / tmp14
tl.store(out_ptr2 + (r1 + (4096*x0)), tmp22, rmask & xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ap/capazybbmitbuviukrgvztks4tyvg3dpptakn7q4vjslonrba25g.py
# Topologically Sorted Source Nodes: [channel_wv], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# channel_wv => convolution
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_2, %primals_3, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_2 = async_compile.triton('triton_poi_fused_convolution_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, 4096], 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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_convolution_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_2(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 1024
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y0 = yindex % 256
y1 = (yindex // 256)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (256*x2) + (1048576*y1)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (y0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + (4096*y3)), tmp2, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rw/crwczp7wd3om4xrznty4rti6kiafa2tzqyvn6po6scyvmhmtcnmw.py
# Topologically Sorted Source Nodes: [conv2d_2, layer_norm, sigmoid], Original ATen: [aten.convolution, aten.native_layer_norm, aten.sigmoid]
# Source node to ATen node mapping:
# conv2d_2 => convolution_2
# layer_norm => add, add_1, mul, mul_1, rsqrt, sub_1, var_mean
# sigmoid => sigmoid
# Graph fragment:
# %convolution_2 : [num_users=2] = call_function[target=torch.ops.aten.convolution.default](args = (%unsqueeze, %primals_6, %primals_7, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%permute, [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=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%permute, %getitem_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %rsqrt), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %primals_8), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_9), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%add_1,), kwargs = {})
triton_per_fused_convolution_native_layer_norm_sigmoid_3 = async_compile.triton('triton_per_fused_convolution_native_layer_norm_sigmoid_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=[4, 512],
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, 8), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_convolution_native_layer_norm_sigmoid_3', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1'], 'no_x_dim': True, '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_layer_norm_sigmoid_3(in_out_ptr0, in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel):
xnumel = 4
XBLOCK: tl.constexpr = 1
rnumel = 512
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 = tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (r1 + (512*x0)), None)
tmp1 = tl.load(in_ptr0 + (r1), None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr2 + (r1), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = tl.broadcast_to(tmp3, [RBLOCK])
tmp7 = triton_helpers.promote_to_tensor(tl.sum(tmp5, 0))
tmp8 = tl.full([1], 512, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp3 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 512.0
tmp17 = tmp15 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tmp21 = tmp2 - tmp10
tmp22 = tmp21 * tmp20
tmp24 = tmp22 * tmp23
tmp26 = tmp24 + tmp25
tmp27 = tl.sigmoid(tmp26)
tl.store(in_out_ptr0 + (r1 + (512*x0)), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + (x0), tmp20, None)
tl.store(out_ptr1 + (r1 + (512*x0)), tmp27, None)
tl.store(out_ptr0 + (x0), tmp10, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/7i/c7imzedyptzfxt2vgr3k6bffmwfpaeyua4jsjocwv6fjlirpf7x5.py
# Topologically Sorted Source Nodes: [channel_out], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# channel_out => mul_2
# Graph fragment:
# %mul_2 : [num_users=4] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_6, %primals_1), kwargs = {})
triton_poi_fused_mul_4 = async_compile.triton('triton_poi_fused_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=[8388608],
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_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_mul_4(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8388608
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 512
x2 = (xindex // 2097152)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (512*x2)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x3), None)
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + (x3), tmp2, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/au/caujl2tq6bs3qovx53ootns3c3h5xsrg3e7jadpmwpwwfswihcgz.py
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
# Source node to ATen node mapping:
# spatial_wq => convolution_4
# spatial_wq_1 => mean
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%mul_2, %primals_12, %primals_13, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%convolution_4, [-1, -2], True), kwargs = {})
triton_red_fused_convolution_mean_5 = async_compile.triton('triton_red_fused_convolution_mean_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.reduction(
size_hints=[32768, 128],
reduction_hint=ReductionHint.OUTER,
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_red_fused_convolution_mean_5', '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_red_fused_convolution_mean_5(in_ptr0, in_ptr1, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 32768
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 256
x1 = (xindex // 256)
tmp1 = tl.load(in_ptr1 + (x0), None, eviction_policy='evict_last')
_tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + (256*r2) + (32768*x1)), rmask, eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = _tmp4 + tmp3
_tmp4 = tl.where(rmask, tmp5, _tmp4)
tmp4 = tl.sum(_tmp4, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/fn/cfnj7f7h2gjgdklco4terukop2ftgjuqvstzhw36dsvm4f77sb5r.py
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
# Source node to ATen node mapping:
# spatial_wq => convolution_4
# spatial_wq_1 => mean
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%mul_2, %primals_12, %primals_13, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%convolution_4, [-1, -2], True), kwargs = {})
triton_per_fused_convolution_mean_6 = async_compile.triton('triton_per_fused_convolution_mean_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.OUTER,
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_convolution_mean_6', '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_convolution_mean_6(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 1024
rnumel = 32
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 = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 256
x1 = (xindex // 256)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (256*r2) + (8192*x1)), 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 + (x3), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/a3/ca3tf2cm62zevu3v2rtbukpzj6gti7pbckdddwcpizhzhqpcvstv.py
# Topologically Sorted Source Nodes: [spatial_wq_3], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# spatial_wq_3 => amax_1, div_1, exp_1, sub_2, sum_2
# Graph fragment:
# %amax_1 : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_8, [-1], True), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_8, %amax_1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_2,), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_1, [-1], True), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_1, %sum_2), kwargs = {})
triton_per_fused__softmax_7 = async_compile.triton('triton_per_fused__softmax_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=[4, 256],
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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_7', 'mutated_arg_names': [], 'no_x_dim': True, '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_7(in_ptr0, out_ptr2, xnumel, rnumel):
xnumel = 4
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)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (256*x0)), None)
tmp1 = 4096.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(triton_helpers.max2(tmp3, 0))
tmp6 = tmp2 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = tmp7 / tmp10
tl.store(out_ptr2 + (r1 + (256*x0)), tmp11, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/3j/c3j3xc5yosmtafu2xjvpoq3fymvzyjnac4rstke5d445pc7rwhy2.py
# Topologically Sorted Source Nodes: [spatial_weight, spatial_out], Original ATen: [aten.sigmoid, aten.mul]
# Source node to ATen node mapping:
# spatial_out => mul_3
# spatial_weight => sigmoid_1
# Graph fragment:
# %sigmoid_1 : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%view_12,), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sigmoid_1, %mul_2), kwargs = {})
triton_poi_fused_mul_sigmoid_8 = async_compile.triton('triton_poi_fused_mul_sigmoid_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=[2048, 4096], 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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_mul_sigmoid_8', '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_8(in_ptr0, in_ptr1, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 2048
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y1 = (yindex // 512)
y0 = yindex % 512
y3 = yindex
tmp0 = tl.load(in_ptr0 + (x2 + (4096*y1)), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr1 + (y0 + (512*x2) + (2097152*y1)), None, eviction_policy='evict_last')
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tl.store(out_ptr0 + (x2 + (4096*y3)), tmp3, 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 = args
args.clear()
assert_size_stride(primals_1, (4, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_3, (256, ), (1, ))
assert_size_stride(primals_4, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_5, (1, ), (1, ))
assert_size_stride(primals_6, (512, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_7, (512, ), (1, ))
assert_size_stride(primals_8, (512, ), (1, ))
assert_size_stride(primals_9, (512, ), (1, ))
assert_size_stride(primals_10, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (256, ), (1, ))
assert_size_stride(primals_12, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (256, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(primals_1, buf0, 2048, 4096, grid=grid(2048, 4096), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [channel_wv], 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, 256, 64, 64), (1048576, 1, 16384, 256))
# Topologically Sorted Source Nodes: [channel_wq], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf0, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 64, 64), (4096, 1, 64, 1))
buf5 = empty_strided_cuda((4, 4096, 1), (4096, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [channel_wq_2], Original ATen: [aten._softmax]
triton_red_fused__softmax_1.run(buf2, primals_5, buf5, 4, 4096, grid=grid(4), stream=stream0)
del primals_5
buf6 = empty_strided_cuda((4, 256, 64, 64), (1048576, 4096, 64, 1), torch.float32)
# Topologically Sorted Source Nodes: [channel_wv], Original ATen: [aten.convolution]
triton_poi_fused_convolution_2.run(buf1, primals_3, buf6, 1024, 4096, grid=grid(1024, 4096), stream=stream0)
del buf1
del primals_3
buf7 = empty_strided_cuda((4, 256, 1), (256, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 256, 4096), (1048576, 4096, 1), 0), buf5, out=buf7)
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(reinterpret_tensor(buf7, (4, 256, 1, 1), (256, 1, 1, 1), 0), primals_6, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 512, 1, 1), (512, 1, 1, 1))
buf9 = reinterpret_tensor(buf8, (4, 512, 1, 1), (512, 1, 512, 512), 0); del buf8 # reuse
buf10 = empty_strided_cuda((4, 1, 1), (1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((4, 1, 1), (1, 4, 4), torch.float32)
buf13 = reinterpret_tensor(buf11, (4, 1, 1), (1, 1, 1), 0); del buf11 # reuse
buf14 = empty_strided_cuda((4, 1, 512), (512, 2048, 1), torch.float32)
# Topologically Sorted Source Nodes: [conv2d_2, layer_norm, sigmoid], Original ATen: [aten.convolution, aten.native_layer_norm, aten.sigmoid]
triton_per_fused_convolution_native_layer_norm_sigmoid_3.run(buf9, buf13, primals_7, primals_8, primals_9, buf10, buf14, 4, 512, grid=grid(4), stream=stream0)
del primals_7
buf15 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512), torch.float32)
# Topologically Sorted Source Nodes: [channel_out], Original ATen: [aten.mul]
triton_poi_fused_mul_4.run(buf14, buf0, buf15, 8388608, grid=grid(8388608), stream=stream0)
del buf14
# Topologically Sorted Source Nodes: [spatial_wv], Original ATen: [aten.convolution]
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 256, 64, 64), (1048576, 1, 16384, 256))
# Topologically Sorted Source Nodes: [spatial_wq], Original ATen: [aten.convolution]
buf17 = extern_kernels.convolution(buf15, primals_12, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf17, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf18 = empty_strided_cuda((4, 256, 1, 1, 32), (8192, 1, 32768, 32768, 256), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
triton_red_fused_convolution_mean_5.run(buf17, primals_13, buf18, 32768, 128, grid=grid(32768), stream=stream0)
del primals_13
buf19 = empty_strided_cuda((4, 256, 1, 1), (256, 1, 1024, 1024), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq, spatial_wq_1], Original ATen: [aten.convolution, aten.mean]
triton_per_fused_convolution_mean_6.run(buf18, buf19, 1024, 32, grid=grid(1024), stream=stream0)
del buf18
buf22 = empty_strided_cuda((4, 1, 256), (256, 256, 1), torch.float32)
# Topologically Sorted Source Nodes: [spatial_wq_3], Original ATen: [aten._softmax]
triton_per_fused__softmax_7.run(buf19, buf22, 4, 256, grid=grid(4), stream=stream0)
del buf19
buf23 = reinterpret_tensor(buf17, (4, 256, 64, 64), (1048576, 4096, 64, 1), 0); del buf17 # reuse
# Topologically Sorted Source Nodes: [spatial_wv], Original ATen: [aten.convolution]
triton_poi_fused_convolution_2.run(buf16, primals_11, buf23, 1024, 4096, grid=grid(1024, 4096), stream=stream0)
del buf16
del primals_11
buf24 = reinterpret_tensor(buf2, (4, 1, 4096), (4096, 4096, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [spatial_wz], Original ATen: [aten.bmm]
extern_kernels.bmm(buf22, reinterpret_tensor(buf23, (4, 256, 4096), (1048576, 4096, 1), 0), out=buf24)
buf25 = empty_strided_cuda((4, 512, 64, 64), (2097152, 4096, 64, 1), torch.float32)
# Topologically Sorted Source Nodes: [spatial_weight, spatial_out], Original ATen: [aten.sigmoid, aten.mul]
triton_poi_fused_mul_sigmoid_8.run(buf24, buf15, buf25, 2048, 4096, grid=grid(2048, 4096), stream=stream0)
return (buf25, buf0, primals_2, primals_4, primals_6, primals_8, primals_9, primals_10, primals_12, buf5, reinterpret_tensor(buf7, (4, 256, 1, 1), (256, 1, 1, 1), 0), buf9, buf10, buf13, buf15, buf22, buf24, reinterpret_tensor(buf23, (4, 4096, 256), (1048576, 1, 4096), 0), reinterpret_tensor(buf6, (4, 4096, 256), (1048576, 1, 4096), 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, 512, 64, 64), (2097152, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((1, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((1, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((512, 256, 1, 1), (256, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((256, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((256, ), (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 torch
from torch import nn
class SequentialPolarizedSelfAttention(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, x):
b, c, h, w = x.size()
channel_wv = self.ch_wv(x)
channel_wq = self.ch_wq(x)
channel_wv = channel_wv.reshape(b, c // 2, -1)
channel_wq = channel_wq.reshape(b, -1, 1)
channel_wq = self.softmax_channel(channel_wq)
channel_wz = torch.matmul(channel_wv, channel_wq).unsqueeze(-1)
channel_weight = self.sigmoid(self.ln(self.ch_wz(channel_wz).
reshape(b, c, 1).permute(0, 2, 1))).permute(0, 2, 1).reshape(b,
c, 1, 1)
channel_out = channel_weight * x
spatial_wv = self.sp_wv(channel_out)
spatial_wq = self.sp_wq(channel_out)
spatial_wq = self.agp(spatial_wq)
spatial_wv = spatial_wv.reshape(b, c // 2, -1)
spatial_wq = spatial_wq.permute(0, 2, 3, 1).reshape(b, 1, c // 2)
spatial_wq = self.softmax_spatial(spatial_wq)
spatial_wz = torch.matmul(spatial_wq, spatial_wv)
spatial_weight = self.sigmoid(spatial_wz.reshape(b, 1, h, w))
spatial_out = spatial_weight * channel_out
return spatial_out
def get_inputs():
return [torch.rand([4, 512, 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 libdevice, math as tl_math
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_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = yindex // 512
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y3), None, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 512 * x2 + 2097152 * y1), tmp0, None)
@triton.jit
def triton_red_fused__softmax_1(in_ptr0, in_ptr1, out_ptr2, xnumel, rnumel,
XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 4
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
_tmp5 = tl.full([XBLOCK, RBLOCK], float('-inf'), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_last', other=0.0)
tmp3 = tmp0 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = triton_helpers.maximum(_tmp5, tmp4)
_tmp5 = tl.where(rmask & xmask, tmp6, _tmp5)
tmp5 = triton_helpers.max2(_tmp5, 1)[:, None]
tmp8 = tl.load(in_ptr1 + 0)
tmp9 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
_tmp14 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp7 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tmp7 + tmp9
tmp11 = tmp10 - tmp5
tmp12 = tl_math.exp(tmp11)
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp15 = _tmp14 + tmp13
_tmp14 = tl.where(rmask & xmask, tmp15, _tmp14)
tmp14 = tl.sum(_tmp14, 1)[:, None]
tmp17 = tl.load(in_ptr1 + 0)
tmp18 = tl.broadcast_to(tmp17, [XBLOCK, RBLOCK])
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp16 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp19 = tmp16 + tmp18
tmp20 = tmp19 - tmp5
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp21 / tmp14
tl.store(out_ptr2 + (r1 + 4096 * x0), tmp22, rmask & xmask)
@triton.jit
def triton_poi_fused_convolution_2(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y0 = yindex % 256
y1 = yindex // 256
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 256 * x2 + 1048576 * y1), None,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 4096 * y3), tmp2, None)
@triton.jit
def triton_per_fused_convolution_native_layer_norm_sigmoid_3(in_out_ptr0,
in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, xnumel, rnumel
):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (r1 + 512 * x0), None)
tmp1 = tl.load(in_ptr0 + r1, None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr2 + r1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = tl.broadcast_to(tmp3, [RBLOCK])
tmp7 = triton_helpers.promote_to_tensor(tl.sum(tmp5, 0))
tmp8 = tl.full([1], 512, tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 / tmp9
tmp11 = tmp3 - tmp10
tmp12 = tmp11 * tmp11
tmp13 = tl.broadcast_to(tmp12, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = 512.0
tmp17 = tmp15 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tmp21 = tmp2 - tmp10
tmp22 = tmp21 * tmp20
tmp24 = tmp22 * tmp23
tmp26 = tmp24 + tmp25
tmp27 = tl.sigmoid(tmp26)
tl.store(in_out_ptr0 + (r1 + 512 * x0), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x0, tmp20, None)
tl.store(out_ptr1 + (r1 + 512 * x0), tmp27, None)
tl.store(out_ptr0 + x0, tmp10, None)
@triton.jit
def triton_poi_fused_mul_4(in_ptr0, in_ptr1, 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 % 512
x2 = xindex // 2097152
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 512 * x2), None, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr1 + x3, None)
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x3, tmp2, None)
@triton.jit
def triton_red_fused_convolution_mean_5(in_ptr0, in_ptr1, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
rnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 256
x1 = xindex // 256
tmp1 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
_tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
x3 = xindex
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_ptr0 + (x0 + 256 * r2 + 32768 * x1), rmask,
eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = _tmp4 + tmp3
_tmp4 = tl.where(rmask, tmp5, _tmp4)
tmp4 = tl.sum(_tmp4, 1)[:, None]
tl.store(out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_per_fused_convolution_mean_6(in_ptr0, out_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
xnumel = 1024
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, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 256
x1 = xindex // 256
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 256 * r2 + 8192 * x1), 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 + x3, tmp4, xmask)
@triton.jit
def triton_per_fused__softmax_7(in_ptr0, out_ptr2, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 256 * x0), None)
tmp1 = 4096.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(triton_helpers.max2(tmp3, 0))
tmp6 = tmp2 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tl.broadcast_to(tmp7, [RBLOCK])
tmp10 = triton_helpers.promote_to_tensor(tl.sum(tmp8, 0))
tmp11 = tmp7 / tmp10
tl.store(out_ptr2 + (r1 + 256 * x0), tmp11, None)
@triton.jit
def triton_poi_fused_mul_sigmoid_8(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y1 = yindex // 512
y0 = yindex % 512
y3 = yindex
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y1), None, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr1 + (y0 + 512 * x2 + 2097152 * y1), None,
eviction_policy='evict_last')
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * tmp2
tl.store(out_ptr0 + (x2 + 4096 * y3), tmp3, 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) = args
args.clear()
assert_size_stride(primals_1, (4, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_3, (256,), (1,))
assert_size_stride(primals_4, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_5, (1,), (1,))
assert_size_stride(primals_6, (512, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_7, (512,), (1,))
assert_size_stride(primals_8, (512,), (1,))
assert_size_stride(primals_9, (512,), (1,))
assert_size_stride(primals_10, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (256,), (1,))
assert_size_stride(primals_12, (256, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (256,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512
), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(2048, 4096)](primals_1, buf0, 2048, 4096,
XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
del primals_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, 256, 64, 64), (1048576, 1, 16384, 256))
buf2 = extern_kernels.convolution(buf0, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 64, 64), (4096, 1, 64, 1))
buf5 = empty_strided_cuda((4, 4096, 1), (4096, 1, 1), torch.float32)
triton_red_fused__softmax_1[grid(4)](buf2, primals_5, buf5, 4, 4096,
XBLOCK=1, RBLOCK=2048, num_warps=16, num_stages=1)
del primals_5
buf6 = empty_strided_cuda((4, 256, 64, 64), (1048576, 4096, 64, 1),
torch.float32)
triton_poi_fused_convolution_2[grid(1024, 4096)](buf1, primals_3,
buf6, 1024, 4096, XBLOCK=64, YBLOCK=64, num_warps=8, num_stages=1)
del buf1
del primals_3
buf7 = empty_strided_cuda((4, 256, 1), (256, 1, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf6, (4, 256, 4096), (
1048576, 4096, 1), 0), buf5, out=buf7)
buf8 = extern_kernels.convolution(reinterpret_tensor(buf7, (4, 256,
1, 1), (256, 1, 1, 1), 0), primals_6, stride=(1, 1), padding=(0,
0), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf8, (4, 512, 1, 1), (512, 1, 1, 1))
buf9 = reinterpret_tensor(buf8, (4, 512, 1, 1), (512, 1, 512, 512), 0)
del buf8
buf10 = empty_strided_cuda((4, 1, 1), (1, 1, 1), torch.float32)
buf11 = empty_strided_cuda((4, 1, 1), (1, 4, 4), torch.float32)
buf13 = reinterpret_tensor(buf11, (4, 1, 1), (1, 1, 1), 0)
del buf11
buf14 = empty_strided_cuda((4, 1, 512), (512, 2048, 1), torch.float32)
triton_per_fused_convolution_native_layer_norm_sigmoid_3[grid(4)](buf9,
buf13, primals_7, primals_8, primals_9, buf10, buf14, 4, 512,
num_warps=4, num_stages=1)
del primals_7
buf15 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768,
512), torch.float32)
triton_poi_fused_mul_4[grid(8388608)](buf14, buf0, buf15, 8388608,
XBLOCK=1024, num_warps=4, num_stages=1)
del buf14
buf16 = extern_kernels.convolution(buf15, primals_10, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf17 = extern_kernels.convolution(buf15, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf17, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf18 = empty_strided_cuda((4, 256, 1, 1, 32), (8192, 1, 32768,
32768, 256), torch.float32)
triton_red_fused_convolution_mean_5[grid(32768)](buf17, primals_13,
buf18, 32768, 128, XBLOCK=64, RBLOCK=8, num_warps=4, num_stages=1)
del primals_13
buf19 = empty_strided_cuda((4, 256, 1, 1), (256, 1, 1024, 1024),
torch.float32)
triton_per_fused_convolution_mean_6[grid(1024)](buf18, buf19, 1024,
32, XBLOCK=128, num_warps=8, num_stages=1)
del buf18
buf22 = empty_strided_cuda((4, 1, 256), (256, 256, 1), torch.float32)
triton_per_fused__softmax_7[grid(4)](buf19, buf22, 4, 256,
num_warps=2, num_stages=1)
del buf19
buf23 = reinterpret_tensor(buf17, (4, 256, 64, 64), (1048576, 4096,
64, 1), 0)
del buf17
triton_poi_fused_convolution_2[grid(1024, 4096)](buf16, primals_11,
buf23, 1024, 4096, XBLOCK=64, YBLOCK=64, num_warps=8, num_stages=1)
del buf16
del primals_11
buf24 = reinterpret_tensor(buf2, (4, 1, 4096), (4096, 4096, 1), 0)
del buf2
extern_kernels.bmm(buf22, reinterpret_tensor(buf23, (4, 256, 4096),
(1048576, 4096, 1), 0), out=buf24)
buf25 = empty_strided_cuda((4, 512, 64, 64), (2097152, 4096, 64, 1),
torch.float32)
triton_poi_fused_mul_sigmoid_8[grid(2048, 4096)](buf24, buf15,
buf25, 2048, 4096, XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
return (buf25, buf0, primals_2, primals_4, primals_6, primals_8,
primals_9, primals_10, primals_12, buf5, reinterpret_tensor(buf7, (
4, 256, 1, 1), (256, 1, 1, 1), 0), buf9, buf10, buf13, buf15, buf22,
buf24, reinterpret_tensor(buf23, (4, 4096, 256), (1048576, 1, 4096),
0), reinterpret_tensor(buf6, (4, 4096, 256), (1048576, 1, 4096), 0))
class SequentialPolarizedSelfAttentionNew(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, input_0):
primals_2 = self.ch_wv.weight
primals_3 = self.ch_wv.bias
primals_4 = self.ch_wq.weight
primals_5 = self.ch_wq.bias
primals_6 = self.ch_wz.weight
primals_7 = self.ch_wz.bias
primals_8 = self.ln.weight
primals_9 = self.ln.bias
primals_10 = self.sp_wv.weight
primals_11 = self.sp_wv.bias
primals_12 = self.sp_wq.weight
primals_13 = self.sp_wq.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,
primals_10, primals_11, primals_12, primals_13])
return output[0]
| rushirajsherlocked/External-Attention-pytorch | SequentialPolarizedSelfAttention | false | 4,311 | [
"MIT"
] | 0 | 7d6814b2d90909adf81c62f3f8a89e30a59d6481 | https://github.com/rushirajsherlocked/External-Attention-pytorch/tree/7d6814b2d90909adf81c62f3f8a89e30a59d6481 | import torch
from torch import nn
class Model(nn.Module):
def __init__(self, channel=512):
super().__init__()
self.ch_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.ch_wq = nn.Conv2d(channel, 1, kernel_size=(1, 1))
self.softmax_channel = nn.Softmax(1)
self.softmax_spatial = nn.Softmax(-1)
self.ch_wz = nn.Conv2d(channel // 2, channel, kernel_size=(1, 1))
self.ln = nn.LayerNorm(channel)
self.sigmoid = nn.Sigmoid()
self.sp_wv = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.sp_wq = nn.Conv2d(channel, channel // 2, kernel_size=(1, 1))
self.agp = nn.AdaptiveAvgPool2d((1, 1))
def forward(self, x):
b, c, h, w = x.size()
channel_wv = self.ch_wv(x)
channel_wq = self.ch_wq(x)
channel_wv = channel_wv.reshape(b, c // 2, -1)
channel_wq = channel_wq.reshape(b, -1, 1)
channel_wq = self.softmax_channel(channel_wq)
channel_wz = torch.matmul(channel_wv, channel_wq).unsqueeze(-1)
channel_weight = self.sigmoid(self.ln(self.ch_wz(channel_wz).
reshape(b, c, 1).permute(0, 2, 1))).permute(0, 2, 1).reshape(b,
c, 1, 1)
channel_out = channel_weight * x
spatial_wv = self.sp_wv(channel_out)
spatial_wq = self.sp_wq(channel_out)
spatial_wq = self.agp(spatial_wq)
spatial_wv = spatial_wv.reshape(b, c // 2, -1)
spatial_wq = spatial_wq.permute(0, 2, 3, 1).reshape(b, 1, c // 2)
spatial_wq = self.softmax_spatial(spatial_wq)
spatial_wz = torch.matmul(spatial_wq, spatial_wv)
spatial_weight = self.sigmoid(spatial_wz.reshape(b, 1, h, w))
spatial_out = spatial_weight * channel_out
return spatial_out
def get_inputs():
return [torch.rand([4, 512, 64, 64])]
def get_init_inputs():
return []
|
SimulatorReward | # 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_7/inductor_cache/5j/c5ji4mfxenghd3ccczky5osir42aijmeisydrv7ufxv2edv4ktf6.py
# Topologically Sorted Source Nodes: [conv2d, x_1], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# x_1 => relu
# Graph fragment:
# %convolution : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%view, %primals_2, %primals_3, [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=[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_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 = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 16) % 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_7/inductor_cache/l5/cl567j2pbsgcv6nt7ux6tlu6b7q2zlzodpyhx5vc6sfgkf4lnrcu.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 = (%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_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=[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_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 = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 16) % 16
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_7/inductor_cache/om/comxhxljgusfvchwfatyy3zlyu63ectx7cc43vhpq4slnbeelc4a.py
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_3 => 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 = {})
triton_poi_fused_convolution_2 = async_compile.triton('triton_poi_fused_convolution_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: 'i32'}, 'device': 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_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_2(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)
x3 = xindex
x1 = (xindex // 16) % 32
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')
# kernel path: runs/run_shard_7/inductor_cache/hx/chxm27gfgqrmlk4dtntfyb4eusceejl3zp6b5gugzys3ftbpi4v5.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_5 => relu_2
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_9), kwargs = {})
# %relu_2 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_3 = async_compile.triton('triton_poi_fused_relu_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: 'i32'}, 'device': 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_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_relu_3(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 800
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 200
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')
# kernel path: runs/run_shard_7/inductor_cache/5d/c5dvgve24hbvrdcz5w4qrfj4r34h3se3x2w5fculeyheewlhp45h.py
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_6 => relu_3
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_11), kwargs = {})
# %relu_3 : [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=[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_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 = 400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 100
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')
# kernel path: runs/run_shard_7/inductor_cache/3y/c3yxwuwmqvwyaal5sg5yv4hb66vvtdtpd7juzykvfeirynnm6y36.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 = (%addmm_2, [1], True), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%addmm_2, %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_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=[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_5', '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__softmax_5(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 // 3)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (3*x1), xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + (3*x1)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + (3*x1)), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp0 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp1 - tmp5
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp2 - tmp5
tmp11 = tl_math.exp(tmp10)
tmp12 = tmp9 + tmp11
tmp13 = tmp4 - tmp5
tmp14 = tl_math.exp(tmp13)
tmp15 = tmp12 + tmp14
tmp16 = tmp7 / tmp15
tl.store(out_ptr0 + (x2), tmp16, 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, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (8, 4, 3, 3), (36, 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, (32, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_7, (32, ), (1, ))
assert_size_stride(primals_8, (200, 512), (512, 1))
assert_size_stride(primals_9, (200, ), (1, ))
assert_size_stride(primals_10, (100, 200), (200, 1))
assert_size_stride(primals_11, (100, ), (1, ))
assert_size_stride(primals_12, (3, 100), (100, 1))
assert_size_stride(primals_13, (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_1, primals_2, 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, 4, 4), (128, 16, 4, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [conv2d, x_1], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_3, 512, grid=grid(512), stream=stream0)
del primals_3
# 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, 16, 4, 4), (256, 16, 4, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [conv2d_1, x_2], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_1.run(buf3, primals_5, 1024, grid=grid(1024), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [x_3], 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, 32, 4, 4), (512, 16, 4, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.convolution]
triton_poi_fused_convolution_2.run(buf5, primals_7, 2048, grid=grid(2048), stream=stream0)
del primals_7
buf6 = empty_strided_cuda((4, 200), (200, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf5, (4, 512), (512, 1), 0), reinterpret_tensor(primals_8, (512, 200), (1, 512), 0), out=buf6)
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.relu]
triton_poi_fused_relu_3.run(buf7, primals_9, 800, grid=grid(800), stream=stream0)
del primals_9
buf8 = empty_strided_cuda((4, 100), (100, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf7, reinterpret_tensor(primals_10, (200, 100), (1, 200), 0), out=buf8)
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.relu]
triton_poi_fused_relu_4.run(buf9, primals_11, 400, grid=grid(400), stream=stream0)
del primals_11
buf10 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_7], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_13, buf9, reinterpret_tensor(primals_12, (100, 3), (1, 100), 0), alpha=1, beta=1, out=buf10)
del primals_13
buf11 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
# Topologically Sorted Source Nodes: [softmax], Original ATen: [aten._softmax]
triton_poi_fused__softmax_5.run(buf10, buf11, 12, grid=grid(12), stream=stream0)
del buf10
return (buf11, primals_2, primals_4, primals_6, primals_1, buf1, buf3, reinterpret_tensor(buf5, (4, 512), (512, 1), 0), buf7, buf9, buf11, primals_12, primals_10, primals_8, )
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((8, 4, 3, 3), (36, 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((32, 16, 3, 3), (144, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((200, 512), (512, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((200, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((100, 200), (200, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((100, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((3, 100), (100, 1), device='cuda:0', dtype=torch.float32)
primals_13 = 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])
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
class SimulatorReward(torch.nn.Module):
def __init__(self):
super(SimulatorReward, self).__init__()
self.conv1 = torch.nn.Conv2d(4, 8, kernel_size=3, padding=1)
self.conv2 = torch.nn.Conv2d(8, 16, kernel_size=3, padding=1)
self.conv3 = torch.nn.Conv2d(16, 32, kernel_size=3, padding=1)
self.fc1 = torch.nn.Linear(512, 200)
self.fc2 = torch.nn.Linear(200, 100)
self.fc3 = torch.nn.Linear(100, 3)
def forward(self, x):
x = x.reshape(-1, 4, 4, 4)
x = F.relu(self.conv1(x))
x = F.relu(self.conv2(x))
x = self.conv3(x)
x = x.view(-1, 512)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return F.softmax(x)
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 math as tl_math
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 = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 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_convolution_relu_1(in_out_ptr0, in_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 // 16 % 16
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_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 // 16 % 32
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)
@triton.jit
def triton_poi_fused_relu_3(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 800
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 200
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)
@triton.jit
def triton_poi_fused_relu_4(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 100
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)
@triton.jit
def triton_poi_fused__softmax_5(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 // 3
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 3 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 3 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 3 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp0 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp1 - tmp5
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp2 - tmp5
tmp11 = tl_math.exp(tmp10)
tmp12 = tmp9 + tmp11
tmp13 = tmp4 - tmp5
tmp14 = tl_math.exp(tmp13)
tmp15 = tmp12 + tmp14
tmp16 = tmp7 / tmp15
tl.store(out_ptr0 + x2, tmp16, 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, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (8, 4, 3, 3), (36, 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, (32, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_7, (32,), (1,))
assert_size_stride(primals_8, (200, 512), (512, 1))
assert_size_stride(primals_9, (200,), (1,))
assert_size_stride(primals_10, (100, 200), (200, 1))
assert_size_stride(primals_11, (100,), (1,))
assert_size_stride(primals_12, (3, 100), (100, 1))
assert_size_stride(primals_13, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_1, primals_2, 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, 4, 4), (128, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(512)](buf1, primals_3, 512,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_3
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, 16, 4, 4), (256, 16, 4, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_1[grid(1024)](buf3, primals_5,
1024, XBLOCK=128, 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, 32, 4, 4), (512, 16, 4, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_2[grid(2048)](buf5, primals_7, 2048,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((4, 200), (200, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf5, (4, 512), (512, 1), 0),
reinterpret_tensor(primals_8, (512, 200), (1, 512), 0), out=buf6)
buf7 = buf6
del buf6
triton_poi_fused_relu_3[grid(800)](buf7, primals_9, 800, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((4, 100), (100, 1), torch.float32)
extern_kernels.mm(buf7, reinterpret_tensor(primals_10, (200, 100),
(1, 200), 0), out=buf8)
buf9 = buf8
del buf8
triton_poi_fused_relu_4[grid(400)](buf9, primals_11, 400, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_11
buf10 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
extern_kernels.addmm(primals_13, buf9, reinterpret_tensor(
primals_12, (100, 3), (1, 100), 0), alpha=1, beta=1, out=buf10)
del primals_13
buf11 = empty_strided_cuda((4, 3), (3, 1), torch.float32)
triton_poi_fused__softmax_5[grid(12)](buf10, buf11, 12, XBLOCK=16,
num_warps=1, num_stages=1)
del buf10
return (buf11, primals_2, primals_4, primals_6, primals_1, buf1, buf3,
reinterpret_tensor(buf5, (4, 512), (512, 1), 0), buf7, buf9, buf11,
primals_12, primals_10, primals_8)
class SimulatorRewardNew(torch.nn.Module):
def __init__(self):
super(SimulatorRewardNew, self).__init__()
self.conv1 = torch.nn.Conv2d(4, 8, kernel_size=3, padding=1)
self.conv2 = torch.nn.Conv2d(8, 16, kernel_size=3, padding=1)
self.conv3 = torch.nn.Conv2d(16, 32, kernel_size=3, padding=1)
self.fc1 = torch.nn.Linear(512, 200)
self.fc2 = torch.nn.Linear(200, 100)
self.fc3 = torch.nn.Linear(100, 3)
def forward(self, input_0):
primals_2 = self.conv1.weight
primals_3 = 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.fc1.weight
primals_9 = self.fc1.bias
primals_10 = self.fc2.weight
primals_11 = self.fc2.bias
primals_12 = self.fc3.weight
primals_13 = self.fc3.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,
primals_10, primals_11, primals_12, primals_13])
return output[0]
| seulbinHwang/DeepReinforcementLearningInAction | SimulatorReward | false | 4,312 | [
"MIT"
] | 0 | c9039fd6951c46c8902cda04580c69159d172c82 | https://github.com/seulbinHwang/DeepReinforcementLearningInAction/tree/c9039fd6951c46c8902cda04580c69159d172c82 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
def __init__(self):
super().__init__()
self.conv1 = torch.nn.Conv2d(4, 8, kernel_size=3, padding=1)
self.conv2 = torch.nn.Conv2d(8, 16, kernel_size=3, padding=1)
self.conv3 = torch.nn.Conv2d(16, 32, kernel_size=3, padding=1)
self.fc1 = torch.nn.Linear(512, 200)
self.fc2 = torch.nn.Linear(200, 100)
self.fc3 = torch.nn.Linear(100, 3)
def forward(self, x):
x = x.reshape(-1, 4, 4, 4)
x = F.relu(self.conv1(x))
x = F.relu(self.conv2(x))
x = self.conv3(x)
x = x.view(-1, 512)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return F.softmax(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
VNLinear | # 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_7/inductor_cache/4b/c4bin6v3nraanry5cu4t7z3qycurd6acdwc4l5azpezbv2erqt7w.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 = (%permute,), 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=[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_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, 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 % 4
x3 = (xindex // 4)
y0 = yindex % 4
y1 = (yindex // 4)
x5 = xindex
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x3) + (16*x2) + (64*y1)), xmask & ymask)
tl.store(out_ptr0 + (x5 + (16*y4)), 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, 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.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(primals_1, buf0, 16, 16, grid=grid(16, 16), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf0, (64, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf1)
del primals_2
return (reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 1, 4, 16), 0), reinterpret_tensor(buf0, (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, 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 torch
import torch.nn as nn
import torch.utils.data
import torch
import torch.nn.parallel
class VNLinear(nn.Module):
def __init__(self, in_channels, out_channels):
super(VNLinear, self).__init__()
self.map_to_feat = nn.Linear(in_channels, out_channels, bias=False)
def forward(self, x):
"""
x: point features of shape [B, N_feat, 3, N_samples, ...]
"""
x_out = self.map_to_feat(x.transpose(1, -1)).transpose(1, -1)
return x_out
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
import torch.utils.data
import torch
import torch.nn.parallel
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, 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 % 4
x3 = xindex // 4
y0 = yindex % 4
y1 = yindex // 4
x5 = xindex
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x3 + 16 * x2 + 64 * y1), xmask & ymask)
tl.store(out_ptr0 + (x5 + 16 * y4), tmp0, xmask & ymask)
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.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(16, 16)](primals_1, buf0, 16, 16,
XBLOCK=16, YBLOCK=16, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf1)
del primals_2
return reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 1, 4, 16), 0
), reinterpret_tensor(buf0, (64, 4), (4, 1), 0)
class VNLinearNew(nn.Module):
def __init__(self, in_channels, out_channels):
super(VNLinearNew, self).__init__()
self.map_to_feat = nn.Linear(in_channels, out_channels, bias=False)
def forward(self, input_0):
primals_2 = self.map_to_feat.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| shiyani21/vnn | VNLinear | false | 4,313 | [
"MIT"
] | 0 | 921be51d6651ff32bff895f4da99ef83d50900da | https://github.com/shiyani21/vnn/tree/921be51d6651ff32bff895f4da99ef83d50900da | import torch
import torch.nn as nn
import torch.utils.data
import torch
import torch.nn.parallel
class Model(nn.Module):
def __init__(self, in_channels, out_channels):
super().__init__()
self.map_to_feat = nn.Linear(in_channels, out_channels, bias=False)
def forward(self, x):
"""
x: point features of shape [B, N_feat, 3, N_samples, ...]
"""
x_out = self.map_to_feat(x.transpose(1, -1)).transpose(1, -1)
return x_out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
ConvAutoencoder | # 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_7/inductor_cache/lp/clp5td7lbqtje3pt7v6xbcp766swgazqemomz2nzsxtdtmjesxht.py
# Topologically Sorted Source Nodes: [conv2d, out], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d => convolution
# out => 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=[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_7/inductor_cache/wy/cwyx3wa4jndgnwzcjpr33hhlviahccyeckxfax46ztwjbjc22gd7.py
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# out_1 => 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=[65536],
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 = 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)
x2 = xindex
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)
tl.store(out_ptr0 + (x2), tmp6, None)
tl.store(out_ptr1 + (x2), tmp16, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/n2/cn23y7jhz2xygynfpumprse4torkd7nlohv4kacxis56jwb54k5e.py
# Topologically Sorted Source Nodes: [conv2d_1, out_2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv2d_1 => convolution_1
# out_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], [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_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_7/inductor_cache/2c/c2ckayeaosayoubmvaniltgwqroshneuww3zj4rpc6cvstbdf2g7.py
# Topologically Sorted Source Nodes: [out_3], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# out_3 => getitem_2, getitem_3
# Graph fragment:
# %getitem_2 : [num_users=2] = call_function[target=operator.getitem](args = (%_low_memory_max_pool2d_with_offsets_1, 0), 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=[4096],
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_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
x1 = (xindex // 16)
x2 = xindex
tmp0 = tl.load(in_ptr0 + ((2*x0) + (64*x1)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (2*x0) + (64*x1)), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (32 + (2*x0) + (64*x1)), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (33 + (2*x0) + (64*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)
tl.store(out_ptr0 + (x2), tmp6, None)
tl.store(out_ptr1 + (x2), tmp16, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ob/cobzstvceif2lfgn6xtfergsweplymenx2njyr3rvqynnghgnjjn.py
# Topologically Sorted Source Nodes: [conv_transpose2d, out_4], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# conv_transpose2d => convolution_2
# out_4 => relu_2
# Graph fragment:
# %convolution_2 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%getitem_2, %primals_6, %primals_7, [2, 2], [0, 0], [1, 1], True, [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_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=[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_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 = 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)
tl.store(in_out_ptr0 + (x3), tmp4, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/t7/ct7cf34m4g63ojfteengkc3tcdxkjvs4wde47kna4a7bol6sdtyb.py
# Topologically Sorted Source Nodes: [conv_transpose2d_1, out_5], Original ATen: [aten.convolution, aten.sigmoid]
# Source node to ATen node mapping:
# conv_transpose2d_1 => convolution_3
# out_5 => sigmoid
# 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], True, [0, 0], 1), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%convolution_3,), kwargs = {})
triton_poi_fused_convolution_sigmoid_5 = async_compile.triton('triton_poi_fused_convolution_sigmoid_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: 'i32'}, 'device': 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_sigmoid_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_sigmoid_5(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)
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.sigmoid(tmp3)
tl.store(in_out_ptr0 + (x0), 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 = args
args.clear()
assert_size_stride(primals_1, (16, 1, 3, 3), (9, 9, 3, 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, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 16, 2, 2), (64, 4, 2, 1))
assert_size_stride(primals_7, (16, ), (1, ))
assert_size_stride(primals_8, (16, 1, 2, 2), (4, 4, 2, 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=(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: [conv2d, out], 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.int8)
# Topologically Sorted Source Nodes: [out_1], 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=(1, 1), 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, out_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.float32)
buf7 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch.int8)
# Topologically Sorted Source Nodes: [out_3], 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: [conv_transpose2d], Original ATen: [aten.convolution]
buf8 = extern_kernels.convolution(buf6, primals_6, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 16, 32, 32), (16384, 1024, 32, 1))
buf9 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [conv_transpose2d, out_4], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_4.run(buf9, primals_7, 65536, grid=grid(65536), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [conv_transpose2d_1], Original ATen: [aten.convolution]
buf10 = extern_kernels.convolution(buf9, primals_8, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=True, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 1, 64, 64), (4096, 4096, 64, 1))
buf11 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [conv_transpose2d_1, out_5], Original ATen: [aten.convolution, aten.sigmoid]
triton_poi_fused_convolution_sigmoid_5.run(buf11, primals_9, 16384, grid=grid(16384), stream=stream0)
del primals_9
return (buf11, primals_1, primals_3, primals_4, primals_6, primals_8, buf1, buf2, buf3, buf5, buf6, buf7, buf9, 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((16, 1, 3, 3), (9, 9, 3, 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, 3, 3), (144, 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((4, 16, 2, 2), (64, 4, 2, 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, 2, 2), (4, 4, 2, 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
import torch.nn as nn
import torch.nn.functional as F
class ConvAutoencoder(nn.Module):
"""Simple convolutional autoencoder
...
Methods
-------
forward(x)
Forward pass of x
"""
def __init__(self):
super(ConvAutoencoder, self).__init__()
self.conv_1 = nn.Conv2d(1, 16, 3, padding=1)
self.conv_2 = nn.Conv2d(16, 4, 3, padding=1)
self.pooling_func = nn.MaxPool2d(2, 2)
self.trans_conv_1 = nn.ConvTranspose2d(4, 16, 2, stride=2)
self.trans_conv_2 = nn.ConvTranspose2d(16, 1, 2, stride=2)
def forward(self, x):
out = F.relu(self.conv_1(x))
out = self.pooling_func(out)
out = F.relu(self.conv_2(out))
out = self.pooling_func(out)
out = F.relu(self.trans_conv_1(out))
out = torch.sigmoid(self.trans_conv_2(out))
return out
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
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
x2 = xindex
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)
tl.store(out_ptr0 + x2, tmp6, None)
tl.store(out_ptr1 + x2, tmp16, 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
x1 = xindex // 16
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 64 * x1), None, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 64 * x1), None, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (32 + 2 * x0 + 64 * x1), None, eviction_policy
='evict_last')
tmp5 = tl.load(in_ptr0 + (33 + 2 * x0 + 64 * 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)
tl.store(out_ptr0 + x2, tmp6, None)
tl.store(out_ptr1 + x2, tmp16, None)
@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)
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)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_sigmoid_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)
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.sigmoid(tmp3)
tl.store(in_out_ptr0 + x0, tmp4, 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, 3, 3), (9, 9, 3, 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, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 16, 2, 2), (64, 4, 2, 1))
assert_size_stride(primals_7, (16,), (1,))
assert_size_stride(primals_8, (16, 1, 2, 2), (4, 4, 2, 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=(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_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.int8)
triton_poi_fused_max_pool2d_with_indices_1[grid(65536)](buf1, buf2,
buf3, 65536, XBLOCK=512, num_warps=4, num_stages=1)
buf4 = extern_kernels.convolution(buf2, 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, 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
.float32)
buf7 = empty_strided_cuda((4, 4, 16, 16), (1024, 256, 16, 1), torch
.int8)
triton_poi_fused_max_pool2d_with_indices_3[grid(4096)](buf5, buf6,
buf7, 4096, XBLOCK=256, num_warps=4, num_stages=1)
buf8 = extern_kernels.convolution(buf6, primals_6, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 16, 32, 32), (16384, 1024, 32, 1))
buf9 = buf8
del buf8
triton_poi_fused_convolution_relu_4[grid(65536)](buf9, primals_7,
65536, XBLOCK=512, num_warps=4, num_stages=1)
del primals_7
buf10 = extern_kernels.convolution(buf9, primals_8, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 1, 64, 64), (4096, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_convolution_sigmoid_5[grid(16384)](buf11,
primals_9, 16384, XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
return (buf11, primals_1, primals_3, primals_4, primals_6, primals_8,
buf1, buf2, buf3, buf5, buf6, buf7, buf9, buf11)
class ConvAutoencoderNew(nn.Module):
"""Simple convolutional autoencoder
...
Methods
-------
forward(x)
Forward pass of x
"""
def __init__(self):
super(ConvAutoencoderNew, self).__init__()
self.conv_1 = nn.Conv2d(1, 16, 3, padding=1)
self.conv_2 = nn.Conv2d(16, 4, 3, padding=1)
self.pooling_func = nn.MaxPool2d(2, 2)
self.trans_conv_1 = nn.ConvTranspose2d(4, 16, 2, stride=2)
self.trans_conv_2 = nn.ConvTranspose2d(16, 1, 2, stride=2)
def forward(self, input_0):
primals_1 = self.conv_1.weight
primals_2 = self.conv_1.bias
primals_4 = self.conv_2.weight
primals_5 = self.conv_2.bias
primals_6 = self.trans_conv_1.weight
primals_7 = self.trans_conv_1.bias
primals_8 = self.trans_conv_2.weight
primals_9 = self.trans_conv_2.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]
| shankal17/Autoencoders | ConvAutoencoder | false | 4,314 | [
"MIT"
] | 0 | 17aa9f1fe573008fa84694e30e9d395127684191 | https://github.com/shankal17/Autoencoders/tree/17aa9f1fe573008fa84694e30e9d395127684191 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
"""Simple convolutional autoencoder
...
Methods
-------
forward(x)
Forward pass of x
"""
def __init__(self):
super().__init__()
self.conv_1 = nn.Conv2d(1, 16, 3, padding=1)
self.conv_2 = nn.Conv2d(16, 4, 3, padding=1)
self.pooling_func = nn.MaxPool2d(2, 2)
self.trans_conv_1 = nn.ConvTranspose2d(4, 16, 2, stride=2)
self.trans_conv_2 = nn.ConvTranspose2d(16, 1, 2, stride=2)
def forward(self, x):
out = F.relu(self.conv_1(x))
out = self.pooling_func(out)
out = F.relu(self.conv_2(out))
out = self.pooling_func(out)
out = F.relu(self.trans_conv_1(out))
out = torch.sigmoid(self.trans_conv_2(out))
return out
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return []
|
GatedLinearUnit | # 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_7/inductor_cache/ia/ciaahs7cbg42mojr2yfpyjkwofxdhoy7qgax7xuox6b5crfrzhnr.py
# Topologically Sorted Source Nodes: [sigmoid, output], Original ATen: [aten.sigmoid, aten.mul]
# Source node to ATen node mapping:
# output => mul
# sigmoid => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = 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, %view_3), 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=[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_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 = 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)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * 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, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1, 4), (4, 1))
assert_size_stride(primals_3, (1, ), (1, ))
assert_size_stride(primals_4, (1, 4), (4, 1))
assert_size_stride(primals_5, (1, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_3, reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 1), (1, 4), 0), alpha=1, beta=1, out=buf1)
del primals_2
del primals_3
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 1), (1, 4), 0), alpha=1, beta=1, out=buf3)
del primals_4
del primals_5
buf4 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [sigmoid, output], Original ATen: [aten.sigmoid, aten.mul]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_sigmoid_0.run(buf1, buf3, buf4, 64, grid=grid(64), stream=stream0)
return (buf4, reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), buf1, 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((1, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((1, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((1, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((1, ), (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 GatedLinearUnit(nn.Module):
def __init__(self, input_size, hidden_layer_size, dropout_rate,
activation=None):
super(GatedLinearUnit, self).__init__()
self.input_size = input_size
self.hidden_layer_size = hidden_layer_size
self.dropout_rate = dropout_rate
self.activation_name = activation
if self.dropout_rate:
self.dropout = nn.Dropout(p=self.dropout_rate)
self.W4 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
self.W5 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
if self.activation_name:
self.activation = getattr(nn, self.activation_name)()
self.sigmoid = nn.Sigmoid()
self.init_weights()
def init_weights(self):
for n, p in self.named_parameters():
if 'bias' not in n:
torch.nn.init.xavier_uniform_(p)
elif 'bias' in n:
torch.nn.init.zeros_(p)
def forward(self, x):
if self.dropout_rate:
x = self.dropout(x)
if self.activation_name:
output = self.sigmoid(self.W4(x)) * self.activation(self.W5(x))
else:
output = self.sigmoid(self.W4(x)) * self.W5(x)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_size': 4, 'hidden_layer_size': 1, 'dropout_rate': 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
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 = 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)
tmp1 = tl.sigmoid(tmp0)
tmp3 = tmp1 * 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, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1, 4), (4, 1))
assert_size_stride(primals_3, (1,), (1,))
assert_size_stride(primals_4, (1, 4), (4, 1))
assert_size_stride(primals_5, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(primals_1, (64,
4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 1), (1, 4), 0
), alpha=1, beta=1, out=buf1)
del primals_2
del primals_3
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_1, (64,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 1), (1, 4), 0
), alpha=1, beta=1, out=buf3)
del primals_4
del primals_5
buf4 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_sigmoid_0[grid(64)](buf1, buf3, buf4, 64,
XBLOCK=64, num_warps=1, num_stages=1)
return buf4, reinterpret_tensor(primals_1, (64, 4), (4, 1), 0), buf1, buf3
class GatedLinearUnitNew(nn.Module):
def __init__(self, input_size, hidden_layer_size, dropout_rate,
activation=None):
super(GatedLinearUnitNew, self).__init__()
self.input_size = input_size
self.hidden_layer_size = hidden_layer_size
self.dropout_rate = dropout_rate
self.activation_name = activation
if self.dropout_rate:
self.dropout = nn.Dropout(p=self.dropout_rate)
self.W4 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
self.W5 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
if self.activation_name:
self.activation = getattr(nn, self.activation_name)()
self.sigmoid = nn.Sigmoid()
self.init_weights()
def init_weights(self):
for n, p in self.named_parameters():
if 'bias' not in n:
torch.nn.init.xavier_uniform_(p)
elif 'bias' in n:
torch.nn.init.zeros_(p)
def forward(self, input_0):
primals_2 = self.W4.weight
primals_3 = self.W4.bias
primals_4 = self.W5.weight
primals_5 = self.W5.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| sherpahu/AutoX | GatedLinearUnit | false | 4,315 | [
"Apache-2.0"
] | 0 | 37aca6bb848ecfdde6868b9f8eb869563fece3eb | https://github.com/sherpahu/AutoX/tree/37aca6bb848ecfdde6868b9f8eb869563fece3eb | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, input_size, hidden_layer_size, dropout_rate,
activation=None):
super().__init__()
self.input_size = input_size
self.hidden_layer_size = hidden_layer_size
self.dropout_rate = dropout_rate
self.activation_name = activation
if self.dropout_rate:
self.dropout = nn.Dropout(p=self.dropout_rate)
self.W4 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
self.W5 = torch.nn.Linear(self.input_size, self.hidden_layer_size)
if self.activation_name:
self.activation = getattr(nn, self.activation_name)()
self.sigmoid = nn.Sigmoid()
self.init_weights()
def init_weights(self):
for n, p in self.named_parameters():
if 'bias' not in n:
torch.nn.init.xavier_uniform_(p)
elif 'bias' in n:
torch.nn.init.zeros_(p)
def forward(self, x):
if self.dropout_rate:
x = self.dropout(x)
if self.activation_name:
output = self.sigmoid(self.W4(x)) * self.activation(self.W5(x))
else:
output = self.sigmoid(self.W4(x)) * self.W5(x)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 1, 0.5]
|
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_7/inductor_cache/sr/csrxdjbtbkq5mhx4lx76hdeti625uy52jalpuc5xjwghomvl635m.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=[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_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 = 12800
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 200
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_7/inductor_cache/qw/cqwdvuc66lglzux6l2qprkpkyfvgktk37ixovgb2zaonbhwfr275.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=[8192],
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 = 6400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x4 = xindex
x0 = xindex % 100
x2 = xindex % 1600
x3 = (xindex // 1600)
tmp0 = tl.load(in_out_ptr0 + (x4), 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 + (x4), tmp4, xmask)
tl.store(out_ptr0 + (x2 + (1664*x3)), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/pv/cpvkbqma42nsdzqpqpkanztnk3kwdfx7iozdpzeynueg2vowikim.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# x_5 => amax, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_5, [1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_5, %amax), 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=[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__log_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__log_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
tl.store(out_ptr0 + (x3), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/pu/cpuo34qggtttrulwolbrgtcihegg5pnmq65f46vdbaenzdat4oka.py
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# x_5 => 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_3 = async_compile.triton('triton_poi_fused__log_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__log_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__log_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')
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 = 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
tl.store(out_ptr0 + (x3), 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, (200, 4), (4, 1))
assert_size_stride(primals_2, (200, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (100, 200), (200, 1))
assert_size_stride(primals_5, (100, ), (1, ))
assert_size_stride(primals_6, (4, 100), (100, 1))
assert_size_stride(primals_7, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 200), (200, 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, 200), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 200), (3200, 800, 200, 1), 0); del buf0 # reuse
buf8 = empty_strided_cuda((4, 4, 4, 200), (3200, 800, 200, 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, 12800, grid=grid(12800), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 100), (100, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf1, (64, 200), (200, 1), 0), reinterpret_tensor(primals_4, (200, 100), (1, 200), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 100), (1600, 400, 100, 1), 0); del buf2 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 100), (1664, 400, 100, 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, 6400, grid=grid(6400), 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, 100), (100, 1), 0), reinterpret_tensor(primals_6, (100, 4), (1, 100), 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._log_softmax]
triton_poi_fused__log_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._log_softmax]
triton_poi_fused__log_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, 200), (200, 1), 0), reinterpret_tensor(buf3, (64, 100), (100, 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((200, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((200, ), (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((100, 200), (200, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((100, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((4, 100), (100, 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
from torch.nn import functional as F
class MLP(torch.nn.Module):
"""MLP for patch segmentation."""
def __init__(self, n_classes, input_dim):
super().__init__()
self.layer_1 = nn.Linear(input_dim, 200)
self.layer_2 = nn.Linear(200, 100)
self.layer_3 = nn.Linear(100, n_classes)
def forward(self, x):
x = self.layer_1(x)
x = F.relu(x)
x = self.layer_2(x)
x = F.relu(x)
x = self.layer_3(x)
x = F.log_softmax(x, dim=1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'n_classes': 4, 'input_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
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 = 12800
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 200
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_relu_threshold_backward_1(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 6400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x4 = xindex
x0 = xindex % 100
x2 = xindex % 1600
x3 = xindex // 1600
tmp0 = tl.load(in_out_ptr0 + x4, 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 + x4, tmp4, xmask)
tl.store(out_ptr0 + (x2 + 1664 * x3), tmp6, xmask)
@triton.jit
def triton_poi_fused__log_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
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_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')
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 = 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
tl.store(out_ptr0 + x3, 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, (200, 4), (4, 1))
assert_size_stride(primals_2, (200,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (100, 200), (200, 1))
assert_size_stride(primals_5, (100,), (1,))
assert_size_stride(primals_6, (4, 100), (100, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 200), (200, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 200), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 200), (3200, 800, 200, 1), 0)
del buf0
buf8 = empty_strided_cuda((4, 4, 4, 200), (3200, 800, 200, 1),
torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(12800)](buf1,
primals_2, buf8, 12800, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 100), (100, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 200), (200, 1), 0),
reinterpret_tensor(primals_4, (200, 100), (1, 200), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 100), (1600, 400, 100, 1), 0)
del buf2
buf7 = empty_strided_cuda((4, 4, 4, 100), (1664, 400, 100, 1),
torch.bool)
triton_poi_fused_relu_threshold_backward_1[grid(6400)](buf3,
primals_5, buf7, 6400, 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, 100),
(100, 1), 0), reinterpret_tensor(primals_6, (100, 4), (1, 100),
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__log_softmax_2[grid(256)](buf4, buf5, 256, XBLOCK=
256, num_warps=4, num_stages=1)
buf6 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf4
triton_poi_fused__log_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, 200), (200, 1), 0
), reinterpret_tensor(buf3, (64, 100), (100, 1), 0
), buf6, primals_6, buf7, primals_4, buf8
class MLPNew(torch.nn.Module):
"""MLP for patch segmentation."""
def __init__(self, n_classes, input_dim):
super().__init__()
self.layer_1 = nn.Linear(input_dim, 200)
self.layer_2 = nn.Linear(200, 100)
self.layer_3 = nn.Linear(100, n_classes)
def forward(self, input_0):
primals_1 = self.layer_1.weight
primals_2 = self.layer_1.bias
primals_4 = self.layer_2.weight
primals_5 = self.layer_2.bias
primals_6 = self.layer_3.weight
primals_7 = self.layer_3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| sachaMorin/dino | MLP | false | 4,316 | [
"Apache-2.0"
] | 0 | b5c42ecffb535a8e6735c63ddc314118927cfd52 | https://github.com/sachaMorin/dino/tree/b5c42ecffb535a8e6735c63ddc314118927cfd52 | import torch
from torch import nn
from torch.nn import functional as F
class Model(torch.nn.Module):
"""MLP for patch segmentation."""
def __init__(self, n_classes, input_dim):
super().__init__()
self.layer_1 = nn.Linear(input_dim, 200)
self.layer_2 = nn.Linear(200, 100)
self.layer_3 = nn.Linear(100, n_classes)
def forward(self, x):
x = self.layer_1(x)
x = F.relu(x)
x = self.layer_2(x)
x = F.relu(x)
x = self.layer_3(x)
x = F.log_softmax(x, dim=1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
ContinuousLoss_L2 | # 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_7/inductor_cache/dr/cdripuddp72w42y6bgeqhjaf4iaiubpdzv2w3gopy272fdmidz2n.py
# Topologically Sorted Source Nodes: [sub, labs, loss, setitem, sum_1], Original ATen: [aten.sub, aten.abs, aten.pow, aten.lift_fresh, aten.index_put, aten.sum]
# Source node to ATen node mapping:
# labs => abs_1
# loss => pow_1
# setitem => full_default, index_put
# sub => sub
# sum_1 => sum_1
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %abs_1 : [num_users=2] = 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), 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: cpu, pin_memory: False})
# %index_put : [num_users=1] = call_function[target=torch.ops.aten.index_put_.default](args = (%pow_1, [%lt], %full_default), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.default](args = (%index_put,), kwargs = {})
triton_per_fused_abs_index_put_lift_fresh_pow_sub_sum_0 = async_compile.triton('triton_per_fused_abs_index_put_lift_fresh_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: '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_index_put_lift_fresh_pow_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_abs_index_put_lift_fresh_pow_sub_sum_0(in_ptr0, in_ptr1, 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
tmp0 = tl.load(in_ptr0 + (r0), None)
tmp1 = tl.load(in_ptr1 + (r0), None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = 1.0
tmp5 = tmp3 < tmp4
tmp6 = tmp3 * tmp3
tmp7 = 0.0
tmp8 = tl.where(tmp5, tmp7, tmp6)
tmp9 = tl.broadcast_to(tmp8, [RBLOCK])
tmp11 = triton_helpers.promote_to_tensor(tl.sum(tmp9, 0))
tl.store(out_ptr1 + (tl.full([1], 0, tl.int32)), tmp11, 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)
buf1 = empty_strided_cuda((), (), torch.float32)
# Topologically Sorted Source Nodes: [sub, labs, loss, setitem, sum_1], Original ATen: [aten.sub, aten.abs, aten.pow, aten.lift_fresh, aten.index_put, aten.sum]
stream0 = get_raw_stream(0)
triton_per_fused_abs_index_put_lift_fresh_pow_sub_sum_0.run(arg0_1, arg1_1, buf1, 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 ContinuousLoss_L2(nn.Module):
""" Class to measure loss between continuous emotion dimension predictions and labels. Using l2 loss as base. """
def __init__(self, margin=1):
super(ContinuousLoss_L2, self).__init__()
self.margin = margin
def forward(self, pred, target):
labs = torch.abs(pred - target)
loss = labs ** 2
loss[labs < self.margin] = 0.0
return loss.sum()
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_index_put_lift_fresh_pow_sub_sum_0(in_ptr0,
in_ptr1, 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
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.abs(tmp2)
tmp4 = 1.0
tmp5 = tmp3 < tmp4
tmp6 = tmp3 * tmp3
tmp7 = 0.0
tmp8 = tl.where(tmp5, tmp7, tmp6)
tmp9 = tl.broadcast_to(tmp8, [RBLOCK])
tmp11 = triton_helpers.promote_to_tensor(tl.sum(tmp9, 0))
tl.store(out_ptr1 + tl.full([1], 0, tl.int32), tmp11, 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)
buf1 = empty_strided_cuda((), (), torch.float32)
get_raw_stream(0)
triton_per_fused_abs_index_put_lift_fresh_pow_sub_sum_0[grid(1)](arg0_1
, arg1_1, buf1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
class ContinuousLoss_L2New(nn.Module):
""" Class to measure loss between continuous emotion dimension predictions and labels. Using l2 loss as base. """
def __init__(self, margin=1):
super(ContinuousLoss_L2New, self).__init__()
self.margin = margin
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| shrookehab/Body-Language-and-Emotion-Recognition | ContinuousLoss_L2 | false | 4,317 | [
"MIT"
] | 0 | a13068be1f8599fa2df6db925a98ac64fd2adf42 | https://github.com/shrookehab/Body-Language-and-Emotion-Recognition/tree/a13068be1f8599fa2df6db925a98ac64fd2adf42 | import torch
import torch.nn as nn
class Model(nn.Module):
""" Class to measure loss between continuous emotion dimension predictions and labels. Using l2 loss as base. """
def __init__(self, margin=1):
super().__init__()
self.margin = margin
def forward(self, pred, target):
labs = torch.abs(pred - target)
loss = labs ** 2
loss[labs < self.margin] = 0.0
return loss.sum()
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 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_7/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_7/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_7/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_7/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_7/inductor_cache/tl/ctlxctn7eg6nwvpdhdhyqadp63cm2ogdwxsotfynexn2zw62nfbb.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_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_7/inductor_cache/ku/ckuw5gg26ddjp4n4da74yttcx6jxcy2y4vb2npxdoq42pzni2oot.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, [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_7/inductor_cache/m3/cm3haovccm7lav2s6wgp3wthu7in42r335z2o7yva4d7olh5begj.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, [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_7/inductor_cache/ir/cirx6nbkabstacj3yb3umtzb7ustxzn5ha5etdpsewqc2v53x42u.py
# Topologically Sorted Source Nodes: [conv2d_3, x_act], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_3 => convolution_3
# x_act => 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_2 : [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_7/inductor_cache/mx/cmxxnyyr5kmtzdpzon3a5fqr6k4jayrantx3iscrc5pevtb6lc52.py
# Topologically Sorted Source Nodes: [x_act_2], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# x_act_2 => amax, exp, log, sub, sub_1, sum_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%addmm, [1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%addmm, %amax), kwargs = {})
# %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_per_fused__log_softmax_8 = async_compile.triton('triton_per_fused__log_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.persistent_reduction(
size_hints=[4, 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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__log_softmax_8', '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__log_softmax_8(in_ptr0, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 4
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, 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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tl_math.log(tmp10)
tmp12 = tmp5 - tmp11
tl.store(out_ptr2 + (r1 + (16*x0)), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/er/cerrhd6dfklfkghffr4w4v6k4tkknpp6pjf2fpylkzd3qma7oygl.py
# Topologically Sorted Source Nodes: [conv2d_4, x_val], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# conv2d_4 => convolution_4
# x_val => 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_1 : [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_7/inductor_cache/dy/cdyy4l65r6roouxfxf2rt7jc3yi26kd72lw6ykhtgaiqpacjtrts.py
# Topologically Sorted Source Nodes: [x_val_2], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_val_2 => relu_5
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_15), kwargs = {})
# %relu_5 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_10 = async_compile.triton('triton_poi_fused_relu_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=[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_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_relu_10(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')
# kernel path: runs/run_shard_7/inductor_cache/jp/cjpw3qv3xbeeitqrvxn6apmx6vcxqlrxpbbtjnzsyqgtt4tatr6q.py
# Topologically Sorted Source Nodes: [x_val_3], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# x_val_3 => tanh
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_17), kwargs = {})
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%add_tensor,), kwargs = {})
triton_poi_fused_tanh_11 = async_compile.triton('triton_poi_fused_tanh_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=[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_tanh_11', '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_11(in_out_ptr0, in_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_out_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr0 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = libdevice.tanh(tmp3)
tl.store(in_out_ptr0 + (x0), 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, primals_10, primals_11, primals_12, primals_13, primals_14, primals_15, primals_16, primals_17 = 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, (16, 64), (64, 1))
assert_size_stride(primals_11, (16, ), (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, (64, 32), (32, 1))
assert_size_stride(primals_15, (64, ), (1, ))
assert_size_stride(primals_16, (1, 64), (64, 1))
assert_size_stride(primals_17, (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, x], 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, x_1], 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, x_2], 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)
buf23 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_3, x_act], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_7.run(buf10, primals_9, buf11, buf23, 16, 16, grid=grid(16, 16), stream=stream0)
del primals_9
buf12 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf11, (4, 64), (64, 1), 0), reinterpret_tensor(primals_10, (64, 16), (1, 64), 0), alpha=1, beta=1, out=buf12)
del primals_11
buf15 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_act_2], Original ATen: [aten._log_softmax]
triton_per_fused__log_softmax_8.run(buf12, buf15, 4, 16, grid=grid(4), stream=stream0)
del buf12
# Topologically Sorted Source Nodes: [conv2d_4], Original ATen: [aten.convolution]
buf16 = 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(buf16, (4, 2, 4, 4), (32, 1, 8, 2))
buf17 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
buf22 = empty_strided_cuda((4, 2, 4, 4), (32, 1, 8, 2), torch.bool)
# Topologically Sorted Source Nodes: [conv2d_4, x_val], Original ATen: [aten.convolution, aten.relu, aten.threshold_backward]
triton_poi_fused_convolution_relu_threshold_backward_9.run(buf16, primals_13, buf17, buf22, 8, 16, grid=grid(8, 16), stream=stream0)
del buf16
del primals_13
buf18 = reinterpret_tensor(buf10, (4, 64), (64, 1), 0); del buf10 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf17, (4, 32), (32, 1), 0), reinterpret_tensor(primals_14, (32, 64), (1, 32), 0), out=buf18)
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [x_val_2], Original ATen: [aten.relu]
triton_poi_fused_relu_10.run(buf19, primals_15, 256, grid=grid(256), stream=stream0)
del primals_15
buf20 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf19, reinterpret_tensor(primals_16, (64, 1), (1, 64), 0), out=buf20)
buf21 = buf20; del buf20 # reuse
# Topologically Sorted Source Nodes: [x_val_3], Original ATen: [aten.tanh]
triton_poi_fused_tanh_11.run(buf21, primals_17, 4, grid=grid(4), stream=stream0)
del primals_17
return (buf15, buf21, buf0, buf1, buf2, buf3, primals_8, primals_12, buf5, buf7, buf9, reinterpret_tensor(buf11, (4, 64), (64, 1), 0), buf15, reinterpret_tensor(buf17, (4, 32), (32, 1), 0), buf19, buf21, primals_16, primals_14, buf22, primals_10, buf23, )
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((16, 64), (64, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((16, ), (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((64, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((1, 64), (64, 1), device='cuda:0', dtype=torch.float32)
primals_17 = 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, 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.nn.functional as F
class Net(nn.Module):
"""policy-value network module"""
def __init__(self, board_width, board_height):
super(Net, self).__init__()
self.board_width = board_width
self.board_height = board_height
self.conv1 = nn.Conv2d(4, 32, kernel_size=3, padding=1)
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, padding=1)
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, padding=1)
self.act_conv1 = nn.Conv2d(128, 4, kernel_size=1)
self.act_fc1 = nn.Linear(4 * board_width * board_height,
board_width * board_height)
self.val_conv1 = nn.Conv2d(128, 2, kernel_size=1)
self.val_fc1 = nn.Linear(2 * board_width * board_height, 64)
self.val_fc2 = nn.Linear(64, 1)
def forward(self, state_input):
x = F.relu(self.conv1(state_input))
x = F.relu(self.conv2(x))
x = F.relu(self.conv3(x))
x_act = F.relu(self.act_conv1(x))
x_act = x_act.view(-1, 4 * self.board_width * self.board_height)
x_act = F.log_softmax(self.act_fc1(x_act))
x_val = F.relu(self.val_conv1(x))
x_val = x_val.view(-1, 2 * self.board_width * self.board_height)
x_val = F.relu(self.val_fc1(x_val))
x_val = F.tanh(self.val_fc2(x_val))
return x_act, x_val
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'board_width': 4, 'board_height': 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_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_per_fused__log_softmax_8(in_ptr0, out_ptr2, xnumel, rnumel,
XBLOCK: tl.constexpr):
xnumel = 4
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, 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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tl_math.log(tmp10)
tmp12 = tmp5 - tmp11
tl.store(out_ptr2 + (r1 + 16 * x0), tmp12, 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_relu_10(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)
@triton.jit
def triton_poi_fused_tanh_11(in_out_ptr0, in_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_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = libdevice.tanh(tmp3)
tl.store(in_out_ptr0 + x0, tmp4, 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, (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, (16, 64), (64, 1))
assert_size_stride(primals_11, (16,), (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, (64, 32), (32, 1))
assert_size_stride(primals_15, (64,), (1,))
assert_size_stride(primals_16, (1, 64), (64, 1))
assert_size_stride(primals_17, (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=256, 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)
buf23 = 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, buf23, 16, 16, XBLOCK=16, YBLOCK=16,
num_warps=4, num_stages=1)
del primals_9
buf12 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf11, (4, 64),
(64, 1), 0), reinterpret_tensor(primals_10, (64, 16), (1, 64),
0), alpha=1, beta=1, out=buf12)
del primals_11
buf15 = empty_strided_cuda((4, 16), (16, 1), torch.float32)
triton_per_fused__log_softmax_8[grid(4)](buf12, buf15, 4, 16,
XBLOCK=1, num_warps=2, num_stages=1)
del buf12
buf16 = 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(buf16, (4, 2, 4, 4), (32, 1, 8, 2))
buf17 = empty_strided_cuda((4, 2, 4, 4), (32, 16, 4, 1), torch.float32)
buf22 = empty_strided_cuda((4, 2, 4, 4), (32, 1, 8, 2), torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_9[grid(8, 16)](
buf16, primals_13, buf17, buf22, 8, 16, XBLOCK=16, YBLOCK=2,
num_warps=1, num_stages=1)
del buf16
del primals_13
buf18 = reinterpret_tensor(buf10, (4, 64), (64, 1), 0)
del buf10
extern_kernels.mm(reinterpret_tensor(buf17, (4, 32), (32, 1), 0),
reinterpret_tensor(primals_14, (32, 64), (1, 32), 0), out=buf18)
buf19 = buf18
del buf18
triton_poi_fused_relu_10[grid(256)](buf19, primals_15, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_15
buf20 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.mm(buf19, reinterpret_tensor(primals_16, (64, 1), (1,
64), 0), out=buf20)
buf21 = buf20
del buf20
triton_poi_fused_tanh_11[grid(4)](buf21, primals_17, 4, XBLOCK=4,
num_warps=1, num_stages=1)
del primals_17
return (buf15, buf21, buf0, buf1, buf2, buf3, primals_8, primals_12,
buf5, buf7, buf9, reinterpret_tensor(buf11, (4, 64), (64, 1), 0),
buf15, reinterpret_tensor(buf17, (4, 32), (32, 1), 0), buf19, buf21,
primals_16, primals_14, buf22, primals_10, buf23)
class NetNew(nn.Module):
"""policy-value network module"""
def __init__(self, board_width, board_height):
super(NetNew, self).__init__()
self.board_width = board_width
self.board_height = board_height
self.conv1 = nn.Conv2d(4, 32, kernel_size=3, padding=1)
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, padding=1)
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, padding=1)
self.act_conv1 = nn.Conv2d(128, 4, kernel_size=1)
self.act_fc1 = nn.Linear(4 * board_width * board_height,
board_width * board_height)
self.val_conv1 = nn.Conv2d(128, 2, kernel_size=1)
self.val_fc1 = nn.Linear(2 * board_width * board_height, 64)
self.val_fc2 = nn.Linear(64, 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.act_conv1.weight
primals_9 = self.act_conv1.bias
primals_10 = self.act_fc1.weight
primals_11 = self.act_fc1.bias
primals_12 = self.val_conv1.weight
primals_13 = self.val_conv1.bias
primals_14 = self.val_fc1.weight
primals_15 = self.val_fc1.bias
primals_16 = self.val_fc2.weight
primals_17 = self.val_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, primals_16, primals_17])
return output[0], output[1]
| sewon0918/pj4 | Net | false | 4,318 | [
"MIT"
] | 0 | 144996e7f99e7639f1fffb34770ab9713307428d | https://github.com/sewon0918/pj4/tree/144996e7f99e7639f1fffb34770ab9713307428d | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
"""policy-value network module"""
def __init__(self, board_width, board_height):
super().__init__()
self.board_width = board_width
self.board_height = board_height
self.conv1 = nn.Conv2d(4, 32, kernel_size=3, padding=1)
self.conv2 = nn.Conv2d(32, 64, kernel_size=3, padding=1)
self.conv3 = nn.Conv2d(64, 128, kernel_size=3, padding=1)
self.act_conv1 = nn.Conv2d(128, 4, kernel_size=1)
self.act_fc1 = nn.Linear(4 * board_width * board_height,
board_width * board_height)
self.val_conv1 = nn.Conv2d(128, 2, kernel_size=1)
self.val_fc1 = nn.Linear(2 * board_width * board_height, 64)
self.val_fc2 = nn.Linear(64, 1)
def forward(self, state_input):
x = F.relu(self.conv1(state_input))
x = F.relu(self.conv2(x))
x = F.relu(self.conv3(x))
x_act = F.relu(self.act_conv1(x))
x_act = x_act.view(-1, 4 * self.board_width * self.board_height)
x_act = F.log_softmax(self.act_fc1(x_act))
x_val = F.relu(self.val_conv1(x))
x_val = x_val.view(-1, 2 * self.board_width * self.board_height)
x_val = F.relu(self.val_fc1(x_val))
x_val = F.tanh(self.val_fc2(x_val))
return x_act, x_val
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
MyBatchNorm | # 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_7/inductor_cache/zm/czmag3kfjhprfl74ex2gtrbasibb5zez6geq7q2q34kdf6ae4lps.py
# Topologically Sorted Source Nodes: [var_mean, add, sqrt], Original ATen: [aten.var_mean, aten.add, aten.sqrt]
# Source node to ATen node mapping:
# add => add
# sqrt => sqrt
# var_mean => var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_1, [0]), kwargs = {correction: 1})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add,), kwargs = {})
triton_poi_fused_add_sqrt_var_mean_0 = async_compile.triton('triton_poi_fused_add_sqrt_var_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.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_sqrt_var_mean_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_add_sqrt_var_mean_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 + (x0), xmask)
tmp1 = tl.load(in_ptr0 + (64 + x0), xmask)
tmp3 = tl.load(in_ptr0 + (128 + x0), xmask)
tmp5 = tl.load(in_ptr0 + (192 + x0), xmask)
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 = 3.0
tmp21 = tmp19 / tmp20
tmp22 = 1e-05
tmp23 = tmp21 + tmp22
tmp24 = libdevice.sqrt(tmp23)
tl.store(out_ptr0 + (x0), tmp8, xmask)
tl.store(out_ptr1 + (x0), tmp24, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ub/cubnh6zgxyxindhc5zulhxvecgwfwmjretrkc27g3y2xg56gt47o.py
# Topologically Sorted Source Nodes: [var_mean, sub, add, sqrt, zed_prime, mul, zed_norm], Original ATen: [aten.var_mean, aten.sub, aten.add, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# add => add
# mul => mul
# sqrt => sqrt
# sub => sub
# var_mean => var_mean
# zed_norm => add_1
# zed_prime => div
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_1, [0]), kwargs = {correction: 1})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %getitem_1), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem, 1e-05), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add,), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, %sqrt), 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_mul_sqrt_sub_var_mean_1 = async_compile.triton('triton_poi_fused_add_div_mul_sqrt_sub_var_mean_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: 'i32'}, 'device': 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_add_div_mul_sqrt_sub_var_mean_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_add_div_mul_sqrt_sub_var_mean_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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
x3 = xindex
x4 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x3), xmask)
tmp2 = tl.load(in_ptr2 + (x4), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr3 + (x4), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp1 - tmp2
tmp5 = tmp3 / tmp4
tmp6 = tmp0 * tmp5
tmp8 = tmp6 + 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 = 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((1, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [var_mean, add, sqrt], Original ATen: [aten.var_mean, aten.add, aten.sqrt]
stream0 = get_raw_stream(0)
triton_poi_fused_add_sqrt_var_mean_0.run(primals_1, 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: [var_mean, sub, add, sqrt, zed_prime, mul, zed_norm], Original ATen: [aten.var_mean, aten.sub, aten.add, aten.sqrt, aten.div, aten.mul]
triton_poi_fused_add_div_mul_sqrt_sub_var_mean_1.run(primals_2, primals_1, buf0, buf1, primals_3, buf2, 256, grid=grid(256), stream=stream0)
del buf0
del buf1
del primals_2
del primals_3
return (buf2, 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 MyBatchNorm(nn.Module):
def __init__(self, size, epsilon=1e-05):
super(MyBatchNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(size))
self.beta = nn.Parameter(torch.zeros(size))
self.epsilon = epsilon
def forward(self, x):
var, meu = torch.var_mean(x, axis=0)
zed_prime = (x - meu) / torch.sqrt(var + self.epsilon)
zed_norm = self.gamma * zed_prime + self.beta
return zed_norm
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'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.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_sqrt_var_mean_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 + x0, xmask)
tmp1 = tl.load(in_ptr0 + (64 + x0), xmask)
tmp3 = tl.load(in_ptr0 + (128 + x0), xmask)
tmp5 = tl.load(in_ptr0 + (192 + x0), xmask)
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 = 3.0
tmp21 = tmp19 / tmp20
tmp22 = 1e-05
tmp23 = tmp21 + tmp22
tmp24 = libdevice.sqrt(tmp23)
tl.store(out_ptr0 + x0, tmp8, xmask)
tl.store(out_ptr1 + x0, tmp24, xmask)
@triton.jit
def triton_poi_fused_add_div_mul_sqrt_sub_var_mean_1(in_ptr0, in_ptr1,
in_ptr2, in_ptr3, in_ptr4, 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
x3 = xindex
x4 = xindex % 64
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x3, xmask)
tmp2 = tl.load(in_ptr2 + x4, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr3 + x4, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp1 - tmp2
tmp5 = tmp3 / tmp4
tmp6 = tmp0 * tmp5
tmp8 = tmp6 + tmp7
tl.store(out_ptr0 + x3, tmp8, 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((1, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_sqrt_var_mean_0[grid(64)](primals_1, 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_div_mul_sqrt_sub_var_mean_1[grid(256)](primals_2,
primals_1, buf0, buf1, primals_3, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf0
del buf1
del primals_2
del primals_3
return buf2, primals_1
class MyBatchNormNew(nn.Module):
def __init__(self, size, epsilon=1e-05):
super(MyBatchNormNew, self).__init__()
self.gamma = nn.Parameter(torch.ones(size))
self.beta = nn.Parameter(torch.zeros(size))
self.epsilon = epsilon
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]
| shohamda/deep-learning | MyBatchNorm | false | 4,319 | [
"MIT"
] | 0 | 160296c403cefd5351ffe5161e07789c22637284 | https://github.com/shohamda/deep-learning/tree/160296c403cefd5351ffe5161e07789c22637284 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, size, epsilon=1e-05):
super().__init__()
self.gamma = nn.Parameter(torch.ones(size))
self.beta = nn.Parameter(torch.zeros(size))
self.epsilon = epsilon
def forward(self, x):
var, meu = torch.var_mean(x, axis=0)
zed_prime = (x - meu) / torch.sqrt(var + self.epsilon)
zed_norm = self.gamma * zed_prime + self.beta
return zed_norm
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
MSELoss | # 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_7/inductor_cache/55/c55fkw5xylmimncpbxp5cat6marcjia37272mv6jzuip2gyy6yad.py
# Topologically Sorted Source Nodes: [sub, error, class_wise_loss], Original ATen: [aten.sub, aten.pow, aten.mean]
# Source node to ATen node mapping:
# class_wise_loss => mean
# error => pow_1
# sub => sub
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [2, 3]), kwargs = {})
triton_per_fused_mean_pow_sub_0 = async_compile.triton('triton_per_fused_mean_pow_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=[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_mean_pow_sub_0', '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_mean_pow_sub_0(in_ptr0, in_ptr1, 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.load(in_ptr1 + (r1 + (16*x0)), xmask, other=0.0)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tl.store(out_ptr0 + (x0), tmp7, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/y3/cy3kkuyuy55mr7trhrol7nuumhuzagr2ln4zbfocuxav52hzj34j.py
# Topologically Sorted Source Nodes: [sub, error, class_wise_loss, loss, loss_1], Original ATen: [aten.sub, aten.pow, aten.mean, aten.mul]
# Source node to ATen node mapping:
# class_wise_loss => mean
# error => pow_1
# loss => mean_1
# loss_1 => mul
# sub => sub
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [2, 3]), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%mean,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mean_1, 1.0), kwargs = {})
triton_per_fused_mean_mul_pow_sub_1 = async_compile.triton('triton_per_fused_mean_mul_pow_sub_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, 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': {2: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 3), equal_to_1=(2,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mean_mul_pow_sub_1', '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_mul_pow_sub_1(in_out_ptr0, in_ptr0, 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 + (r0), None)
tmp1 = 16.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tmp6 = tmp5 / tmp1
tmp7 = 1.0
tmp8 = tmp6 * tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp8, 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, 1), torch.float32)
# Topologically Sorted Source Nodes: [sub, error, class_wise_loss], Original ATen: [aten.sub, aten.pow, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_mean_pow_sub_0.run(arg0_1, arg1_1, buf0, 16, 16, grid=grid(16), stream=stream0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [sub, error, class_wise_loss, loss, loss_1], Original ATen: [aten.sub, aten.pow, aten.mean, aten.mul]
triton_per_fused_mean_mul_pow_sub_1.run(buf2, buf0, 1, 16, grid=grid(1), stream=stream0)
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)
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class MSELoss(nn.Module):
"""DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(MSELoss, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
error = (predict - target) ** 2
class_wise_loss = torch.mean(error, dim=(2, 3))
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor)
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
import torch._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
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_mean_pow_sub_0(in_ptr0, in_ptr1, 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.load(in_ptr1 + (r1 + 16 * x0), xmask, other=0.0)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tl.store(out_ptr0 + x0, tmp7, xmask)
@triton.jit
def triton_per_fused_mean_mul_pow_sub_1(in_out_ptr0, in_ptr0, 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 + r0, None)
tmp1 = 16.0
tmp2 = tmp0 / tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = tl.sum(tmp3, 1)[:, None]
tmp6 = tmp5 / tmp1
tmp7 = 1.0
tmp8 = tmp6 * tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp8, 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, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_mean_pow_sub_0[grid(16)](arg0_1, arg1_1, buf0, 16,
16, XBLOCK=8, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1
del buf1
triton_per_fused_mean_mul_pow_sub_1[grid(1)](buf2, buf0, 1, 16,
XBLOCK=1, num_warps=2, num_stages=1)
del buf0
return buf2,
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class MSELossNew(nn.Module):
"""DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(MSELossNew, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_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]
| shuaizzZ/mmsegmentation | MSELoss | false | 4,320 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class Model(nn.Module):
"""DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super().__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
error = (predict - target) ** 2
class_wise_loss = torch.mean(error, dim=(2, 3))
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GCN | # 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_7/inductor_cache/be/cbej2f3myglhqo2dienhyo4fp7tbscq32k7imbgc2psgl6gaxxhi.py
# Topologically Sorted Source Nodes: [add, x], Original ATen: [aten.add, aten.relu]
# Source node to ATen node mapping:
# add => add
# x => relu
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_1, %primals_4), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add,), kwargs = {})
triton_poi_fused_add_relu_0 = async_compile.triton('triton_poi_fused_add_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_add_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_add_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')
# kernel path: runs/run_shard_7/inductor_cache/ul/culvxc5xcnacfjypzxghwcyc2445sqsz25ci4rib6axjxs3fv3so.py
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# log_softmax => amax, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%addmm_default, [1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%addmm_default, %amax), kwargs = {})
triton_poi_fused__log_softmax_1 = async_compile.triton('triton_poi_fused__log_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: 'i32'}, 'device': 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_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__log_softmax_1(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
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
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/yr/cyr6fatjcqc5np3quy6arljtkkff4qjmueyb5b4pk5xvkxgrzuvd.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=[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__log_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__log_softmax_2(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
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, 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
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 = 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, 4), (4, 1))
assert_size_stride(primals_6, (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: [support], Original ATen: [aten.mm]
extern_kernels.mm(primals_2, primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.mm]
extern_kernels.mm(primals_3, buf0, out=buf1)
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [add, x], Original ATen: [aten.add, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_add_relu_0.run(buf2, primals_4, 16, grid=grid(16), stream=stream0)
del primals_4
buf3 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [support_1], Original ATen: [aten.mm]
extern_kernels.mm(buf2, primals_5, out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.addmm(primals_6, primals_3, buf3, alpha=1, beta=1, out=buf4)
del primals_6
buf5 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
triton_poi_fused__log_softmax_1.run(buf4, buf5, 16, grid=grid(16), stream=stream0)
buf6 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
triton_poi_fused__log_softmax_2.run(buf5, buf6, 16, grid=grid(16), stream=stream0)
del buf5
return (buf4, buf6, buf2, buf6, reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), reinterpret_tensor(primals_2, (4, 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, 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, 4), (4, 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)
| from torch.nn import Module
import math
import torch
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
import torch.nn as nn
import torch.nn.functional as F
class GraphConvolution(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907
"""
def __init__(self, in_features, out_features, bias=True):
super(GraphConvolution, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.FloatTensor(in_features, out_features))
if bias:
self.bias = Parameter(torch.FloatTensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
if self.bias is not None:
return output + self.bias
else:
return output
def __repr__(self):
return self.__class__.__name__ + ' (' + str(self.in_features
) + ' -> ' + str(self.out_features) + ')'
class GCN(nn.Module):
def __init__(self, nfeat, nhid, nclass, dropout):
super(GCN, self).__init__()
self.gc1 = GraphConvolution(nfeat, nhid)
self.gc2 = GraphConvolution(nhid, nclass)
self.dropout = dropout
def forward(self, x, adj):
x = F.relu(self.gc1(x, adj))
x = F.dropout(x, self.dropout, training=self.training)
x = self.gc2(x, adj)
return x, F.log_softmax(x, dim=1)
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'nfeat': 4, 'nhid': 4, 'nclass': 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 import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch.nn import Module
import math
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
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_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)
@triton.jit
def triton_poi_fused__log_softmax_1(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
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
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_softmax_2(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
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, 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
tl.store(out_ptr0 + x2, tmp13, 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, 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, 4), (4, 1))
assert_size_stride(primals_6, (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_2, primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_3, buf0, out=buf1)
buf2 = buf1
del buf1
get_raw_stream(0)
triton_poi_fused_add_relu_0[grid(16)](buf2, primals_4, 16, XBLOCK=
16, num_warps=1, num_stages=1)
del primals_4
buf3 = buf0
del buf0
extern_kernels.mm(buf2, primals_5, out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_6, primals_3, buf3, alpha=1, beta=1,
out=buf4)
del primals_6
buf5 = buf3
del buf3
triton_poi_fused__log_softmax_1[grid(16)](buf4, buf5, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__log_softmax_2[grid(16)](buf5, buf6, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf5
return buf4, buf6, buf2, buf6, reinterpret_tensor(primals_3, (4, 4), (1,
4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0
), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0)
class GraphConvolution(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907
"""
def __init__(self, in_features, out_features, bias=True):
super(GraphConvolution, self).__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.FloatTensor(in_features, out_features))
if bias:
self.bias = Parameter(torch.FloatTensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
if self.bias is not None:
return output + self.bias
else:
return output
def __repr__(self):
return self.__class__.__name__ + ' (' + str(self.in_features
) + ' -> ' + str(self.out_features) + ')'
class GCNNew(nn.Module):
def __init__(self, nfeat, nhid, nclass, dropout):
super(GCNNew, self).__init__()
self.gc1 = GraphConvolution(nfeat, nhid)
self.gc2 = GraphConvolution(nhid, nclass)
self.dropout = dropout
def forward(self, input_0, input_1):
primals_1 = self.gc1.weight
primals_4 = self.gc1.bias
primals_2 = self.gc2.weight
primals_6 = self.gc2.bias
primals_3 = input_0
primals_5 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0], output[1]
| shovalf/OGRE-1 | GCN | false | 4,321 | [
"MIT"
] | 0 | 08efad50fac27e8c9621897838e122a2e8fdae1c | https://github.com/shovalf/OGRE-1/tree/08efad50fac27e8c9621897838e122a2e8fdae1c | from torch.nn import Module
import math
import torch
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
import torch.nn as nn
import torch.nn.functional as F
class GraphConvolution(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907
"""
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.in_features = in_features
self.out_features = out_features
self.weight = Parameter(torch.FloatTensor(in_features, out_features))
if bias:
self.bias = Parameter(torch.FloatTensor(out_features))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
if self.bias is not None:
return output + self.bias
else:
return output
def __repr__(self):
return self.__class__.__name__ + ' (' + str(self.in_features
) + ' -> ' + str(self.out_features) + ')'
class Model(nn.Module):
def __init__(self, nfeat, nhid, nclass, dropout):
super().__init__()
self.gc1 = GraphConvolution(nfeat, nhid)
self.gc2 = GraphConvolution(nhid, nclass)
self.dropout = dropout
def forward(self, x, adj):
x = F.relu(self.gc1(x, adj))
x = F.dropout(x, self.dropout, training=self.training)
x = self.gc2(x, adj)
return x, F.log_softmax(x, dim=1)
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [4, 4, 4, 0.5]
|
ECA | # 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_7/inductor_cache/l3/cl35tzbhrd24dhunkbb6gjs54aklpyr46oikqhoylcgmkcmhujil.py
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# y => 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')
# kernel path: runs/run_shard_7/inductor_cache/g2/cg2ce7tlyrhfqtl6vsyvllceqhjdvbyx4guw5z2sfzdgy6ig2lwc.py
# Topologically Sorted Source Nodes: [y_2, mul], Original ATen: [aten.sigmoid, aten.mul]
# Source node to ATen node mapping:
# mul => mul
# y_2 => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%permute_1,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_1, %sigmoid), kwargs = {})
triton_poi_fused_mul_sigmoid_1 = async_compile.triton('triton_poi_fused_mul_sigmoid_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: 'i32'}, 'device': 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_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_mul_sigmoid_1(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
x2 = xindex
x1 = (xindex // 16)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + (x2), tmp3, 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, 1, 3, 3), (9, 9, 3, 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: [y], 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)
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (4, 1, 1, 4), (4, 0, 0, 1), 0), primals_2, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 1, 4), (4, 4, 4, 1))
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [y_2, mul], Original ATen: [aten.sigmoid, aten.mul]
triton_poi_fused_mul_sigmoid_1.run(primals_1, buf2, buf3, 256, grid=grid(256), stream=stream0)
return (buf3, primals_1, primals_2, reinterpret_tensor(buf1, (4, 1, 1, 4), (4, 1, 1, 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, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((1, 1, 3, 3), (9, 9, 3, 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._C
import torch.serialization
from torch import nn
from typing import *
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class ECA(nn.Module):
"""Constructs a ECA module.
Args:
channel: Number of channels of the input feature map
k_size: Adaptive selection of kernel size
"""
def __init__(self, in_channels, k_size=3):
super(ECA, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.conv = nn.Conv2d(1, 1, kernel_size=k_size, padding=(k_size - 1
) // 2, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, x):
_b, _c, _h, _w = int_size(x)
y = self.avg_pool(x)
y = self.conv(y.permute(0, 3, 2, 1)).permute(0, 3, 2, 1)
y = self.sigmoid(y)
return x * y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_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._C
import torch.serialization
from torch import nn
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_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)
@triton.jit
def triton_poi_fused_mul_sigmoid_1(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
x2 = xindex
x1 = xindex // 16
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + x2, tmp3, 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, 1, 3, 3), (9, 9, 3, 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=1,
num_warps=2, num_stages=1)
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (4, 1, 1,
4), (4, 0, 0, 1), 0), primals_2, stride=(1, 1), padding=(1, 1),
dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 1, 4), (4, 4, 4, 1))
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_mul_sigmoid_1[grid(256)](primals_1, buf2, buf3,
256, XBLOCK=128, num_warps=4, num_stages=1)
return buf3, primals_1, primals_2, reinterpret_tensor(buf1, (4, 1, 1, 4
), (4, 1, 1, 1), 0), buf2
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class ECANew(nn.Module):
"""Constructs a ECA module.
Args:
channel: Number of channels of the input feature map
k_size: Adaptive selection of kernel size
"""
def __init__(self, in_channels, k_size=3):
super(ECANew, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.conv = nn.Conv2d(1, 1, kernel_size=k_size, padding=(k_size - 1
) // 2, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, input_0):
primals_2 = self.conv.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| shuaizzZ/mmsegmentation | ECA | false | 4,322 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
from typing import *
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class Model(nn.Module):
"""Constructs a ECA module.
Args:
channel: Number of channels of the input feature map
k_size: Adaptive selection of kernel size
"""
def __init__(self, in_channels, k_size=3):
super().__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.conv = nn.Conv2d(1, 1, kernel_size=k_size, padding=(k_size - 1
) // 2, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, x):
_b, _c, _h, _w = int_size(x)
y = self.avg_pool(x)
y = self.conv(y.permute(0, 3, 2, 1)).permute(0, 3, 2, 1)
y = self.sigmoid(y)
return x * y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
Mix2Pooling | # 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_7/inductor_cache/4v/c4vc2jsp72fmrcszakst46cabkdn7c6m2udmayb4fexbdm7r2tkk.py
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# out => cat
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%_adaptive_avg_pool2d, %getitem_2], 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=[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_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, 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 // 16) % 4
x0 = xindex % 16
x2 = (xindex // 64)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, 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], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr0 + (32 + x0 + (16*((-2) + x1)) + (64*x2)), tmp6 & xmask, other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x3), tmp10, 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: [out], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_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._C
import torch.serialization
from torch import nn
from typing import *
class Mix2Pooling(nn.Module):
def __init__(self, size):
super(Mix2Pooling, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(size)
self.max_pool = nn.AdaptiveMaxPool2d(size)
def forward(self, x):
spx = torch.chunk(x, 2, 1)
out = torch.cat((self.avg_pool(spx[0]), self.max_pool(spx[1])), 1)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'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._C
import torch.serialization
from torch import 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_poi_fused_cat_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 // 16 % 4
x0 = xindex % 16
x2 = xindex // 64
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr0 + (32 + x0 + 16 * (-2 + x1) + 64 * x2), tmp6 &
xmask, other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x3, tmp10, 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_cat_0[grid(256)](arg0_1, buf0, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class Mix2PoolingNew(nn.Module):
def __init__(self, size):
super(Mix2PoolingNew, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(size)
self.max_pool = nn.AdaptiveMaxPool2d(size)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| shuaizzZ/mmsegmentation | Mix2Pooling | false | 4,323 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
from typing import *
class Model(nn.Module):
def __init__(self, size):
super().__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(size)
self.max_pool = nn.AdaptiveMaxPool2d(size)
def forward(self, x):
spx = torch.chunk(x, 2, 1)
out = torch.cat((self.avg_pool(spx[0]), self.max_pool(spx[1])), 1)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
CDiceLoss | # 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_7/inductor_cache/qz/cqza6p5fjiie2hfiu5dfjqqugrnzziwuwxzlhzy2aa7khopxjbym.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%arg0_1, [1], 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
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_7/inductor_cache/5o/c5oxgtzmewpalxb2zodalaja2tjnelcx7pomom5mq4tbrnbcpwgh.py
# Topologically Sorted Source Nodes: [probs, scatter_, mul, intersection, pow_1, sum_2, sum_3], Original ATen: [aten._softmax, aten.scatter, aten.mul, aten.sum, aten.pow]
# Source node to ATen node mapping:
# intersection => sum_2
# mul => mul
# pow_1 => pow_1
# probs => div, sum_1
# scatter_ => scatter_upon_const_tensor
# sum_2 => sum_3
# sum_3 => sum_4
# 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 = {})
# %scatter_upon_const_tensor : [num_users=2] = call_function[target=torch._inductor.fx_passes.post_grad.scatter_upon_const_tensor](args = (), kwargs = {shape: [4, 4, 4, 4], background_val: 0, dtype: torch.int64, dim: 1, selector: %arg1_1, val: 1})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %scatter_upon_const_tensor), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [2, 3]), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%div, 2), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [2, 3]), kwargs = {})
# %sum_4 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%scatter_upon_const_tensor, [2, 3]), kwargs = {})
triton_per_fused__softmax_mul_pow_scatter_sum_1 = async_compile.triton('triton_per_fused__softmax_mul_pow_scatter_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=[16, 16],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 2: '*fp32', 3: '*fp32', 4: '*i64', 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, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_mul_pow_scatter_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, '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__softmax_mul_pow_scatter_sum_1(in_ptr0, in_ptr1, out_ptr1, out_ptr2, out_ptr3, 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)
r2 = rindex
x3 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (r2 + (16*x3)), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp2 = tl.load(in_ptr0 + (16 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp4 = tl.load(in_ptr0 + (32 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (48 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp9 = tl.load(in_ptr1 + (r2 + (16*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp10 = x0
tmp11 = tmp9 == tmp10
tmp12 = tl.full([1, 1], 1, tl.int64)
tmp13 = tl.full([1, 1], 0, tl.int64)
tmp14 = tl.where(tmp11, tmp12, tmp13)
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp8 * tmp15
tmp17 = tl.broadcast_to(tmp16, [XBLOCK, RBLOCK])
tmp19 = tl.where(xmask, tmp17, 0)
tmp20 = tl.sum(tmp19, 1)[:, None]
tmp21 = tmp8 * tmp8
tmp22 = tl.broadcast_to(tmp21, [XBLOCK, RBLOCK])
tmp24 = tl.where(xmask, tmp22, 0)
tmp25 = tl.sum(tmp24, 1)[:, None]
tmp26 = tl.broadcast_to(tmp14, [XBLOCK, RBLOCK])
tmp28 = tl.where(xmask, tmp26, 0)
tmp29 = tl.sum(tmp28, 1)[:, None]
tl.store(out_ptr1 + (x3), tmp20, xmask)
tl.store(out_ptr2 + (x3), tmp25, xmask)
tl.store(out_ptr3 + (x3), tmp29, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ds/cdsf5jlq46zrnt5hdmvx6wcnyrvwffqvyury4y3ho32i4yhbpqeb.py
# Topologically Sorted Source Nodes: [mul_1, add_1, union, add_2, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.mul, aten.add, aten.div, aten.mean, aten.rsub]
# Source node to ATen node mapping:
# add_1 => add_1
# add_2 => add_2
# class_wise_loss => div_1
# loss => mean
# loss_1 => mul_2
# mul_1 => mul_1
# sub => sub_1
# union => add
# Graph fragment:
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sum_2, 2), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, 1e-06), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_3, %sum_4), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add, 1e-06), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%add_1, %add_2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%div_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %mean), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, 1.0), kwargs = {})
triton_per_fused_add_div_mean_mul_rsub_2 = async_compile.triton('triton_per_fused_add_div_mean_mul_rsub_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, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i64', 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, 5), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mean_mul_rsub_2', '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_mean_mul_rsub_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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 + (r0), None)
tmp5 = tl.load(in_ptr1 + (r0), None)
tmp6 = tl.load(in_ptr2 + (r0), None)
tmp1 = 2.0
tmp2 = tmp0 * tmp1
tmp3 = 1e-06
tmp4 = tmp2 + tmp3
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 + tmp7
tmp9 = tmp8 + tmp3
tmp10 = tmp4 / tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.sum(tmp11, 1)[:, None]
tmp14 = 16.0
tmp15 = tmp13 / tmp14
tmp16 = 1.0
tmp17 = tmp16 - tmp15
tmp18 = tmp17 * tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp18, 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, 1, 4, 4), (16, 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: [probs], 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
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
# Topologically Sorted Source Nodes: [probs, scatter_, mul, intersection, pow_1, sum_2, sum_3], Original ATen: [aten._softmax, aten.scatter, aten.mul, aten.sum, aten.pow]
triton_per_fused__softmax_mul_pow_scatter_sum_1.run(buf0, arg1_1, buf2, buf3, buf4, 16, 16, grid=grid(16), stream=stream0)
del arg1_1
del buf0
buf5 = empty_strided_cuda((), (), torch.float32)
buf6 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [mul_1, add_1, union, add_2, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.mul, aten.add, aten.div, aten.mean, aten.rsub]
triton_per_fused_add_div_mean_mul_rsub_2.run(buf6, buf2, buf3, buf4, 1, 16, grid=grid(1), stream=stream0)
del buf2
del buf3
del buf4
return (buf6, )
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, 1, 4, 4), (16, 16, 4, 1), device='cuda:0', dtype=torch.int64)
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class CDiceLoss(nn.Module):
"""class-wise DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(CDiceLoss, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
probs = F.softmax(predict, dim=1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
intersection = torch.sum(probs * target_onehot, dim=(2, 3))
union = torch.sum(probs.pow(2), dim=(2, 3)) + torch.sum(target_onehot,
dim=(2, 3))
class_wise_loss = (2 * intersection + self.smooth) / (union + self.
smooth)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
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_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_per_fused__softmax_mul_pow_scatter_sum_1(in_ptr0, in_ptr1,
out_ptr1, out_ptr2, out_ptr3, 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)
r2 = rindex
x3 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (r2 + 16 * x3), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp2 = tl.load(in_ptr0 + (16 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp4 = tl.load(in_ptr0 + (32 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (48 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp9 = tl.load(in_ptr1 + (r2 + 16 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp10 = x0
tmp11 = tmp9 == tmp10
tmp12 = tl.full([1, 1], 1, tl.int64)
tmp13 = tl.full([1, 1], 0, tl.int64)
tmp14 = tl.where(tmp11, tmp12, tmp13)
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp8 * tmp15
tmp17 = tl.broadcast_to(tmp16, [XBLOCK, RBLOCK])
tmp19 = tl.where(xmask, tmp17, 0)
tmp20 = tl.sum(tmp19, 1)[:, None]
tmp21 = tmp8 * tmp8
tmp22 = tl.broadcast_to(tmp21, [XBLOCK, RBLOCK])
tmp24 = tl.where(xmask, tmp22, 0)
tmp25 = tl.sum(tmp24, 1)[:, None]
tmp26 = tl.broadcast_to(tmp14, [XBLOCK, RBLOCK])
tmp28 = tl.where(xmask, tmp26, 0)
tmp29 = tl.sum(tmp28, 1)[:, None]
tl.store(out_ptr1 + x3, tmp20, xmask)
tl.store(out_ptr2 + x3, tmp25, xmask)
tl.store(out_ptr3 + x3, tmp29, xmask)
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_2(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, 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 + r0, None)
tmp5 = tl.load(in_ptr1 + r0, None)
tmp6 = tl.load(in_ptr2 + r0, None)
tmp1 = 2.0
tmp2 = tmp0 * tmp1
tmp3 = 1e-06
tmp4 = tmp2 + tmp3
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 + tmp7
tmp9 = tmp8 + tmp3
tmp10 = tmp4 / tmp9
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.sum(tmp11, 1)[:, None]
tmp14 = 16.0
tmp15 = tmp13 / tmp14
tmp16 = 1.0
tmp17 = tmp16 - tmp15
tmp18 = tmp17 * tmp16
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp18, 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, 1, 4, 4), (16, 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=
256, num_warps=4, num_stages=1)
del arg0_1
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
triton_per_fused__softmax_mul_pow_scatter_sum_1[grid(16)](buf0,
arg1_1, buf2, buf3, buf4, 16, 16, XBLOCK=8, num_warps=2,
num_stages=1)
del arg1_1
del buf0
buf5 = empty_strided_cuda((), (), torch.float32)
buf6 = buf5
del buf5
triton_per_fused_add_div_mean_mul_rsub_2[grid(1)](buf6, buf2, buf3,
buf4, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del buf2
del buf3
del buf4
return buf6,
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class CDiceLossNew(nn.Module):
"""class-wise DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(CDiceLossNew, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| shuaizzZ/mmsegmentation | CDiceLoss | false | 4,324 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class Model(nn.Module):
"""class-wise DiceLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super().__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
probs = F.softmax(predict, dim=1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
intersection = torch.sum(probs * target_onehot, dim=(2, 3))
union = torch.sum(probs.pow(2), dim=(2, 3)) + torch.sum(target_onehot,
dim=(2, 3))
class_wise_loss = (2 * intersection + self.smooth) / (union + self.
smooth)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
def get_init_inputs():
return []
|
SpatialAttention | # 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_7/inductor_cache/46/c46mg7rvdztu6n5oosf5c4if7ziag6obrxhwbn43lcdfibfuom7w.py
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# out => cat
# Graph fragment:
# %cat : [num_users=2] = call_function[target=torch.ops.aten.cat.default](args = ([%mean, %getitem], 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 = tmp5 + tmp6
tmp8 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp9 = tmp7 + tmp8
tmp10 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp11 = tmp9 + tmp10
tmp12 = 4.0
tmp13 = tmp11 / tmp12
tmp14 = tl.full(tmp13.shape, 0.0, tmp13.dtype)
tmp15 = tl.where(tmp4, tmp13, tmp14)
tmp16 = tmp0 >= tmp3
tmp17 = tl.full([1], 2, tl.int64)
tmp18 = tmp0 < tmp17
tmp19 = tl.load(in_ptr0 + (x0 + (64*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp20 = tl.load(in_ptr0 + (16 + x0 + (64*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp21 = triton_helpers.maximum(tmp19, tmp20)
tmp22 = tl.load(in_ptr0 + (32 + x0 + (64*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp23 = triton_helpers.maximum(tmp21, tmp22)
tmp24 = tl.load(in_ptr0 + (48 + x0 + (64*x2)), tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp25 = triton_helpers.maximum(tmp23, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp16, tmp25, tmp26)
tmp28 = tl.where(tmp4, tmp15, tmp27)
tl.store(out_ptr0 + (x3), tmp28, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/6q/c6qyrmvchep2lyeodxjgze7brt2fv4khvsx2os2smplvfajckxaz.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 = (%convolution,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_1, %sigmoid), kwargs = {})
triton_poi_fused_mul_sigmoid_1 = async_compile.triton('triton_poi_fused_mul_sigmoid_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: 'i32'}, 'device': 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_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_mul_sigmoid_1(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_ptr1 + (x0 + (16*x2)), xmask, eviction_policy='evict_last')
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + (x3), tmp3, 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, 2, 7, 7), (98, 49, 7, 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: [out], Original ATen: [aten.cat]
stream0 = get_raw_stream(0)
triton_poi_fused_cat_0.run(primals_1, buf0, 128, grid=grid(128), stream=stream0)
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1), padding=(3, 3), 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))
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]
triton_poi_fused_mul_sigmoid_1.run(primals_1, buf1, buf2, 256, grid=grid(256), stream=stream0)
return (buf2, primals_1, primals_2, 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, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((1, 2, 7, 7), (98, 49, 7, 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._C
import torch.serialization
from torch import nn
from typing import *
class SpatialAttention(nn.Module):
def __init__(self, kernel_size=7):
super(SpatialAttention, self).__init__()
assert kernel_size in (3, 7), 'kernel size must be 3 or 7'
padding = 3 if kernel_size == 7 else 1
self.conv = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, x):
avgout = torch.mean(x, dim=1, keepdim=True)
maxout, _ = torch.max(x, dim=1, keepdim=True)
out = torch.cat([avgout, maxout], dim=1)
out = self.conv(out)
return x * self.sigmoid(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
import torch._C
import torch.serialization
from torch import 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_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 = tmp5 + tmp6
tmp8 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp9 = tmp7 + tmp8
tmp10 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = tmp9 + tmp10
tmp12 = 4.0
tmp13 = tmp11 / tmp12
tmp14 = tl.full(tmp13.shape, 0.0, tmp13.dtype)
tmp15 = tl.where(tmp4, tmp13, tmp14)
tmp16 = tmp0 >= tmp3
tl.full([1], 2, tl.int64)
tmp19 = tl.load(in_ptr0 + (x0 + 64 * x2), tmp16 & xmask,
eviction_policy='evict_last', other=0.0)
tmp20 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), tmp16 & xmask,
eviction_policy='evict_last', other=0.0)
tmp21 = triton_helpers.maximum(tmp19, tmp20)
tmp22 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), tmp16 & xmask,
eviction_policy='evict_last', other=0.0)
tmp23 = triton_helpers.maximum(tmp21, tmp22)
tmp24 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), tmp16 & xmask,
eviction_policy='evict_last', other=0.0)
tmp25 = triton_helpers.maximum(tmp23, tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp16, tmp25, tmp26)
tmp28 = tl.where(tmp4, tmp15, tmp27)
tl.store(out_ptr0 + x3, tmp28, xmask)
@triton.jit
def triton_poi_fused_mul_sigmoid_1(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_ptr1 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp0 * tmp2
tl.store(out_ptr0 + x3, tmp3, 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, 2, 7, 7), (98, 49, 7, 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)](primals_1, buf0, 128, XBLOCK=128,
num_warps=4, num_stages=1)
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(3, 3), 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))
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_mul_sigmoid_1[grid(256)](primals_1, buf1, buf2,
256, XBLOCK=256, num_warps=4, num_stages=1)
return buf2, primals_1, primals_2, buf0, buf1
class SpatialAttentionNew(nn.Module):
def __init__(self, kernel_size=7):
super(SpatialAttentionNew, self).__init__()
assert kernel_size in (3, 7), 'kernel size must be 3 or 7'
padding = 3 if kernel_size == 7 else 1
self.conv = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, input_0):
primals_2 = self.conv.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
| shuaizzZ/mmsegmentation | SpatialAttention | false | 4,325 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
from typing import *
class Model(nn.Module):
def __init__(self, kernel_size=7):
super().__init__()
assert kernel_size in (3, 7), 'kernel size must be 3 or 7'
padding = 3 if kernel_size == 7 else 1
self.conv = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False)
self.sigmoid = nn.Sigmoid()
def forward(self, x):
avgout = torch.mean(x, dim=1, keepdim=True)
maxout, _ = torch.max(x, dim=1, keepdim=True)
out = torch.cat([avgout, maxout], dim=1)
out = self.conv(out)
return x * self.sigmoid(out)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
RecallLoss | # 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_7/inductor_cache/qz/cqza6p5fjiie2hfiu5dfjqqugrnzziwuwxzlhzy2aa7khopxjbym.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%arg0_1, [1], 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
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_7/inductor_cache/ik/cikxhjmzv22zd3qirv4mpbxbcmbl4xa3ybyp6cafd5txbi6gbe3u.py
# Topologically Sorted Source Nodes: [probs, scatter_, mul, true_positive, total_target], Original ATen: [aten._softmax, aten.scatter, aten.mul, aten.sum]
# Source node to ATen node mapping:
# mul => mul
# probs => div, sum_1
# scatter_ => scatter_upon_const_tensor
# total_target => sum_3
# true_positive => 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 = {})
# %scatter_upon_const_tensor : [num_users=2] = call_function[target=torch._inductor.fx_passes.post_grad.scatter_upon_const_tensor](args = (), kwargs = {shape: [4, 4, 4, 4], background_val: 0, dtype: torch.int64, dim: 1, selector: %arg1_1, val: 1})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, %scatter_upon_const_tensor), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [2, 3]), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%scatter_upon_const_tensor, [2, 3]), kwargs = {})
triton_per_fused__softmax_mul_scatter_sum_1 = async_compile.triton('triton_per_fused__softmax_mul_scatter_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=[16, 16],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 2: '*fp32', 3: '*i64', 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_per_fused__softmax_mul_scatter_sum_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 6, '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_mul_scatter_sum_1(in_ptr0, in_ptr1, out_ptr0, out_ptr1, 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)
r2 = rindex
x3 = xindex
x1 = (xindex // 4)
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (r2 + (16*x3)), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp2 = tl.load(in_ptr0 + (16 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp4 = tl.load(in_ptr0 + (32 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (48 + r2 + (64*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp9 = tl.load(in_ptr1 + (r2 + (16*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp10 = x0
tmp11 = tmp9 == tmp10
tmp12 = tl.full([1, 1], 1, tl.int64)
tmp13 = tl.full([1, 1], 0, tl.int64)
tmp14 = tl.where(tmp11, tmp12, tmp13)
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp8 * tmp15
tmp17 = tl.broadcast_to(tmp16, [XBLOCK, RBLOCK])
tmp19 = tl.where(xmask, tmp17, 0)
tmp20 = tl.sum(tmp19, 1)[:, None]
tmp21 = tl.broadcast_to(tmp14, [XBLOCK, RBLOCK])
tmp23 = tl.where(xmask, tmp21, 0)
tmp24 = tl.sum(tmp23, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp20, xmask)
tl.store(out_ptr1 + (x3), tmp24, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/qv/cqv2pcywfused3w2zzzmleabdaurod2e6tqrx36l6l2hz52fnoqe.py
# Topologically Sorted Source Nodes: [add, add_1, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.add, aten.div, aten.mean, aten.rsub, aten.mul]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# class_wise_loss => div_1
# loss => mean
# loss_1 => mul_1
# sub => sub_1
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_2, 1e-06), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_3, 1e-06), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%add, %add_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%div_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %mean), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, 1.0), kwargs = {})
triton_per_fused_add_div_mean_mul_rsub_2 = async_compile.triton('triton_per_fused_add_div_mean_mul_rsub_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, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i64', 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_2', 'mutated_arg_names': ['in_out_ptr0'], '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_add_div_mean_mul_rsub_2(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 + (r0), None)
tmp3 = tl.load(in_ptr1 + (r0), None)
tmp1 = 1e-06
tmp2 = tmp0 + tmp1
tmp4 = tmp3.to(tl.float32)
tmp5 = tmp4 + tmp1
tmp6 = tmp2 / tmp5
tmp7 = tl.broadcast_to(tmp6, [XBLOCK, RBLOCK])
tmp9 = tl.sum(tmp7, 1)[:, None]
tmp10 = 16.0
tmp11 = tmp9 / tmp10
tmp12 = 1.0
tmp13 = tmp12 - tmp11
tmp14 = tmp13 * tmp12
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 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, 1, 4, 4), (16, 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: [probs], 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, 1), torch.float32)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
# Topologically Sorted Source Nodes: [probs, scatter_, mul, true_positive, total_target], Original ATen: [aten._softmax, aten.scatter, aten.mul, aten.sum]
triton_per_fused__softmax_mul_scatter_sum_1.run(buf0, arg1_1, buf1, buf2, 16, 16, grid=grid(16), stream=stream0)
del arg1_1
del buf0
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [add, add_1, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.add, aten.div, aten.mean, aten.rsub, aten.mul]
triton_per_fused_add_div_mean_mul_rsub_2.run(buf4, buf1, buf2, 1, 16, grid=grid(1), stream=stream0)
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, 1, 4, 4), (16, 16, 4, 1), device='cuda:0', dtype=torch.int64)
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class RecallLoss(nn.Module):
"""RecallLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(RecallLoss, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
probs = F.softmax(predict, dim=1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
true_positive = torch.sum(probs * target_onehot, dim=(2, 3))
total_target = torch.sum(target_onehot, dim=(2, 3))
class_wise_loss = (true_positive + self.smooth) / (total_target +
self.smooth)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
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_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_per_fused__softmax_mul_scatter_sum_1(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, 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)
r2 = rindex
x3 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + (r2 + 16 * x3), xmask, other=0.0)
tmp1 = tl.load(in_ptr0 + (r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp2 = tl.load(in_ptr0 + (16 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp4 = tl.load(in_ptr0 + (32 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + (48 + r2 + 64 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp9 = tl.load(in_ptr1 + (r2 + 16 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tmp10 = x0
tmp11 = tmp9 == tmp10
tmp12 = tl.full([1, 1], 1, tl.int64)
tmp13 = tl.full([1, 1], 0, tl.int64)
tmp14 = tl.where(tmp11, tmp12, tmp13)
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp8 * tmp15
tmp17 = tl.broadcast_to(tmp16, [XBLOCK, RBLOCK])
tmp19 = tl.where(xmask, tmp17, 0)
tmp20 = tl.sum(tmp19, 1)[:, None]
tmp21 = tl.broadcast_to(tmp14, [XBLOCK, RBLOCK])
tmp23 = tl.where(xmask, tmp21, 0)
tmp24 = tl.sum(tmp23, 1)[:, None]
tl.store(out_ptr0 + x3, tmp20, xmask)
tl.store(out_ptr1 + x3, tmp24, xmask)
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_2(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 + r0, None)
tmp3 = tl.load(in_ptr1 + r0, None)
tmp1 = 1e-06
tmp2 = tmp0 + tmp1
tmp4 = tmp3.to(tl.float32)
tmp5 = tmp4 + tmp1
tmp6 = tmp2 / tmp5
tmp7 = tl.broadcast_to(tmp6, [XBLOCK, RBLOCK])
tmp9 = tl.sum(tmp7, 1)[:, None]
tmp10 = 16.0
tmp11 = tmp9 / tmp10
tmp12 = 1.0
tmp13 = tmp12 - tmp11
tmp14 = tmp13 * tmp12
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 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, 1, 4, 4), (16, 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=
256, num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
triton_per_fused__softmax_mul_scatter_sum_1[grid(16)](buf0, arg1_1,
buf1, buf2, 16, 16, XBLOCK=1, num_warps=2, num_stages=1)
del arg1_1
del buf0
buf3 = empty_strided_cuda((), (), torch.float32)
buf4 = buf3
del buf3
triton_per_fused_add_div_mean_mul_rsub_2[grid(1)](buf4, buf1, buf2,
1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del buf1
del buf2
return buf4,
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class RecallLossNew(nn.Module):
"""RecallLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super(RecallLossNew, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| shuaizzZ/mmsegmentation | RecallLoss | false | 4,326 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class Model(nn.Module):
"""RecallLoss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', class_weight=None, loss_weight=1.0):
super().__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
probs = F.softmax(predict, dim=1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
true_positive = torch.sum(probs * target_onehot, dim=(2, 3))
total_target = torch.sum(target_onehot, dim=(2, 3))
class_wise_loss = (true_positive + self.smooth) / (total_target +
self.smooth)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
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_7/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], True), 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_7/inductor_cache/r3/cr374vaxi6epsuiifrjdtytnqrgm37y2ovtb2khul3f5x6eqbvt7.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 = (5.0, %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
r1 = (rindex // 4)
tmp0 = tl.load(in_ptr0 + (r2), None)
tmp3 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp3 * tmp3
tmp5 = tmp2 * tmp4
tmp6 = 5.0
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, 1), (16, 4, 1, 64), 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
from torch import nn
from torch.nn import functional as F
class ContrastiveLoss(nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=5.0):
super(ContrastiveLoss, self).__init__()
self.margin = margin
def forward(self, output1, output2, label):
euclidean_distance = F.pairwise_distance(output1, output2, keepdim=True
)
loss_contrastive = torch.mean((1 - label) * torch.pow(
euclidean_distance, 2) + label * 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
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_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
r1 = rindex // 4
tmp0 = tl.load(in_ptr0 + r2, None)
tmp3 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp3 * tmp3
tmp5 = tmp2 * tmp4
tmp6 = 5.0
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, 1), (16, 4, 1, 64), 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(nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=5.0):
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]
| shuuchen/siamese_network | ContrastiveLoss | false | 4,327 | [
"Apache-2.0"
] | 0 | 54a952d320800c6bb5618cb40386e4c25bdde6fb | https://github.com/shuuchen/siamese_network/tree/54a952d320800c6bb5618cb40386e4c25bdde6fb | import torch
from torch import nn
from torch.nn import functional as F
class Model(nn.Module):
"""
Contrastive loss function.
Based on: http://yann.lecun.com/exdb/publis/pdf/hadsell-chopra-lecun-06.pdf
"""
def __init__(self, margin=5.0):
super().__init__()
self.margin = margin
def forward(self, output1, output2, label):
euclidean_distance = F.pairwise_distance(output1, output2, keepdim=True
)
loss_contrastive = torch.mean((1 - label) * torch.pow(
euclidean_distance, 2) + label * 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 []
|
TripletLoss | # 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_7/inductor_cache/ny/cnyi4x3au2afjh4vggmfheesqq2dmjfewktqnpj53l522remfvxt.py
# Topologically Sorted Source Nodes: [pos_dist, add, neg_dist, sub, hinge_dist, loss], Original ATen: [aten.sub, aten.add, aten.norm, aten.clamp, aten.mean]
# Source node to ATen node mapping:
# add => add_2
# hinge_dist => clamp_min
# loss => mean
# neg_dist => add_1, pow_3, pow_4, sub_1, sum_2
# pos_dist => add, pow_1, pow_2, sub, sum_1
# sub => sub_2
# 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=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_1, 0.5), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%pow_2, 5.0), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg1_1, %arg2_1), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Scalar](args = (%sub_1, 1e-06), kwargs = {})
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%add_1, 2.0), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_3, [3]), kwargs = {})
# %pow_4 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sum_2, 0.5), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_2, %pow_4), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_2, 0.0), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%clamp_min,), kwargs = {})
triton_per_fused_add_clamp_mean_norm_sub_0 = async_compile.triton('triton_per_fused_add_clamp_mean_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.persistent_reduction(
size_hints=[1, 64],
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, 5), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_clamp_mean_norm_sub_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 12, '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_norm_sub_0(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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 + (4*r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*r0), None, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + (4*r0)), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + (4*r0)), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + (4*r0)), None, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (2 + (4*r0)), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (3 + (4*r0)), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (3 + (4*r0)), None, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr2 + (4*r0), None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr2 + (1 + (4*r0)), None, eviction_policy='evict_last')
tmp36 = tl.load(in_ptr2 + (2 + (4*r0)), None, eviction_policy='evict_last')
tmp41 = tl.load(in_ptr2 + (3 + (4*r0)), None, 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)
tmp25 = 5.0
tmp26 = tmp24 + tmp25
tmp28 = tmp0 - tmp27
tmp29 = tmp28 + tmp3
tmp30 = tmp29 * tmp29
tmp32 = tmp6 - tmp31
tmp33 = tmp32 + tmp3
tmp34 = tmp33 * tmp33
tmp35 = tmp30 + tmp34
tmp37 = tmp12 - tmp36
tmp38 = tmp37 + tmp3
tmp39 = tmp38 * tmp38
tmp40 = tmp35 + tmp39
tmp42 = tmp18 - tmp41
tmp43 = tmp42 + tmp3
tmp44 = tmp43 * tmp43
tmp45 = tmp40 + tmp44
tmp46 = libdevice.sqrt(tmp45)
tmp47 = tmp26 - tmp46
tmp48 = 0.0
tmp49 = triton_helpers.maximum(tmp47, tmp48)
tmp50 = tl.broadcast_to(tmp49, [XBLOCK, RBLOCK])
tmp52 = tl.sum(tmp50, 1)[:, None]
tmp53 = 64.0
tmp54 = tmp52 / tmp53
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp54, 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)
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [pos_dist, add, neg_dist, sub, hinge_dist, loss], Original ATen: [aten.sub, aten.add, aten.norm, aten.clamp, aten.mean]
stream0 = get_raw_stream(0)
triton_per_fused_add_clamp_mean_norm_sub_0.run(buf2, arg1_1, arg0_1, arg2_1, 1, 64, grid=grid(1), stream=stream0)
del arg0_1
del arg1_1
del arg2_1
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
from torch import nn
from torch.nn.modules.distance import PairwiseDistance
class TripletLoss(nn.Module):
def __init__(self, margin=5.0):
super(TripletLoss, self).__init__()
self.margin = margin
self.pdist = PairwiseDistance(2)
def forward(self, anchor, negative, positive):
pos_dist = self.pdist.forward(anchor, positive)
neg_dist = self.pdist.forward(anchor, negative)
hinge_dist = torch.clamp(self.margin + pos_dist - neg_dist, min=0.0)
loss = torch.mean(hinge_dist)
return loss
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
from torch import nn
from torch.nn.modules.distance import PairwiseDistance
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_clamp_mean_norm_sub_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, 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 + 4 * r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * r0, None, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + 4 * r0), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (1 + 4 * r0), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + 4 * r0), None, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (2 + 4 * r0), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (3 + 4 * r0), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (3 + 4 * r0), None, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr2 + 4 * r0, None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr2 + (1 + 4 * r0), None, eviction_policy='evict_last')
tmp36 = tl.load(in_ptr2 + (2 + 4 * r0), None, eviction_policy='evict_last')
tmp41 = tl.load(in_ptr2 + (3 + 4 * r0), None, 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)
tmp25 = 5.0
tmp26 = tmp24 + tmp25
tmp28 = tmp0 - tmp27
tmp29 = tmp28 + tmp3
tmp30 = tmp29 * tmp29
tmp32 = tmp6 - tmp31
tmp33 = tmp32 + tmp3
tmp34 = tmp33 * tmp33
tmp35 = tmp30 + tmp34
tmp37 = tmp12 - tmp36
tmp38 = tmp37 + tmp3
tmp39 = tmp38 * tmp38
tmp40 = tmp35 + tmp39
tmp42 = tmp18 - tmp41
tmp43 = tmp42 + tmp3
tmp44 = tmp43 * tmp43
tmp45 = tmp40 + tmp44
tmp46 = libdevice.sqrt(tmp45)
tmp47 = tmp26 - tmp46
tmp48 = 0.0
tmp49 = triton_helpers.maximum(tmp47, tmp48)
tmp50 = tl.broadcast_to(tmp49, [XBLOCK, RBLOCK])
tmp52 = tl.sum(tmp50, 1)[:, None]
tmp53 = 64.0
tmp54 = tmp52 / tmp53
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp54, 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)
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1
del buf1
get_raw_stream(0)
triton_per_fused_add_clamp_mean_norm_sub_0[grid(1)](buf2, arg1_1,
arg0_1, arg2_1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
return buf2,
class TripletLossNew(nn.Module):
def __init__(self, margin=5.0):
super(TripletLossNew, self).__init__()
self.margin = margin
self.pdist = PairwiseDistance(2)
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]
| shuuchen/siamese_network | TripletLoss | false | 4,328 | [
"Apache-2.0"
] | 0 | 54a952d320800c6bb5618cb40386e4c25bdde6fb | https://github.com/shuuchen/siamese_network/tree/54a952d320800c6bb5618cb40386e4c25bdde6fb | import torch
from torch import nn
from torch.nn.modules.distance import PairwiseDistance
class Model(nn.Module):
def __init__(self, margin=5.0):
super().__init__()
self.margin = margin
self.pdist = PairwiseDistance(2)
def forward(self, anchor, negative, positive):
pos_dist = self.pdist.forward(anchor, positive)
neg_dist = self.pdist.forward(anchor, negative)
hinge_dist = torch.clamp(self.margin + pos_dist - neg_dist, min=0.0)
loss = torch.mean(hinge_dist)
return loss
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 []
|
BinaryFocalLossWithLogits | # 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_7/inductor_cache/hj/chjcbk2hdljkxzvt3ki2g7h34oz6yzea55wveanzmbe64apv4thx.py
# Topologically Sorted Source Nodes: [neg, probs_neg, pow_1, mul, mul_1, log_sigmoid, mul_2, probs_pos, pow_2, mul_3, sub, mul_4, log_sigmoid_1, neg_1, mul_5, loss_tmp], Original ATen: [aten.neg, aten.sigmoid, aten.pow, aten.mul, aten.log_sigmoid_forward, aten.rsub, aten.sub]
# Source node to ATen node mapping:
# log_sigmoid => abs_1, exp, full_default, log1p, minimum, neg_1, sub
# log_sigmoid_1 => abs_2, exp_1, full_default_1, log1p_1, minimum_1, neg_3, sub_2
# loss_tmp => sub_3
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# mul_4 => mul_4
# mul_5 => mul_5
# neg => neg
# neg_1 => neg_2
# pow_1 => pow_1
# pow_2 => pow_2
# probs_neg => sigmoid_1
# probs_pos => sigmoid
# sub => sub_1
# Graph fragment:
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%arg0_1,), kwargs = {})
# %sigmoid_1 : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%neg,), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sigmoid_1, 2.0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_1, -4), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %arg1_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_1 : [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_1,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%minimum, %log1p), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_1, %sub), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%arg0_1,), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sigmoid, 2.0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_2, -3), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1.0, %arg1_1), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_3, %sub_1), kwargs = {})
# %full_default_1 : [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})
# %neg_2 : [num_users=2] = call_function[target=torch.ops.aten.neg.default](args = (%arg0_1,), kwargs = {})
# %minimum_1 : [num_users=1] = call_function[target=torch.ops.aten.minimum.default](args = (%full_default_1, %neg_2), kwargs = {})
# %abs_2 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%neg_2,), kwargs = {})
# %neg_3 : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%abs_2,), kwargs = {})
# %exp_1 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%neg_3,), kwargs = {})
# %log1p_1 : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp_1,), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%minimum_1, %log1p_1), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_4, %sub_2), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_2, %mul_5), kwargs = {})
triton_poi_fused_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_0 = async_compile.triton('triton_poi_fused_log_sigmoid_forward_mul_neg_pow_rsub_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.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_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_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_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_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)
tmp6 = tl.load(in_ptr1 + (x0), xmask)
tmp1 = -tmp0
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp2 * tmp2
tmp4 = -4.0
tmp5 = tmp3 * tmp4
tmp7 = tmp5 * tmp6
tmp8 = 0.0
tmp9 = triton_helpers.minimum(tmp8, tmp0)
tmp10 = tl_math.abs(tmp0)
tmp11 = -tmp10
tmp12 = tl_math.exp(tmp11)
tmp13 = libdevice.log1p(tmp12)
tmp14 = tmp9 - tmp13
tmp15 = tmp7 * tmp14
tmp16 = tl.sigmoid(tmp0)
tmp17 = tmp16 * tmp16
tmp18 = -3.0
tmp19 = tmp17 * tmp18
tmp20 = 1.0
tmp21 = tmp20 - tmp6
tmp22 = tmp19 * tmp21
tmp23 = triton_helpers.minimum(tmp8, tmp1)
tmp24 = tl_math.abs(tmp1)
tmp25 = -tmp24
tmp26 = tl_math.exp(tmp25)
tmp27 = libdevice.log1p(tmp26)
tmp28 = tmp23 - tmp27
tmp29 = tmp22 * tmp28
tmp30 = tmp15 - tmp29
tl.store(out_ptr0 + (x0), tmp30, 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: [neg, probs_neg, pow_1, mul, mul_1, log_sigmoid, mul_2, probs_pos, pow_2, mul_3, sub, mul_4, log_sigmoid_1, neg_1, mul_5, loss_tmp], Original ATen: [aten.neg, aten.sigmoid, aten.pow, aten.mul, aten.log_sigmoid_forward, aten.rsub, aten.sub]
stream0 = get_raw_stream(0)
triton_poi_fused_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_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)
| import torch
import warnings
from typing import Optional
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.data
def binary_focal_loss_with_logits(input: 'torch.Tensor', target:
'torch.Tensor', alpha: 'float'=0.25, gamma: 'float'=2.0, reduction:
'str'='none', eps: 'Optional[float]'=None) ->torch.Tensor:
"""Function that computes Binary Focal loss.
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
input: input data tensor of arbitrary shape.
target: the target tensor with shape matching input.
alpha: Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
eps: Deprecated: scalar for numerically stability when dividing. This is no longer used.
Returns:
the computed loss.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> logits = torch.tensor([[[6.325]],[[5.26]],[[87.49]]])
>>> labels = torch.tensor([[[1.]],[[1.]],[[0.]]])
>>> binary_focal_loss_with_logits(logits, labels, **kwargs)
tensor(21.8725)
"""
if eps is not None and not torch.jit.is_scripting():
warnings.warn(
'`binary_focal_loss_with_logits` has been reworked for improved numerical stability and the `eps` argument is no longer necessary'
, DeprecationWarning, stacklevel=2)
if not isinstance(input, torch.Tensor):
raise TypeError(f'Input type is not a torch.Tensor. Got {type(input)}')
if not len(input.shape) >= 2:
raise ValueError(
f'Invalid input shape, we expect BxCx*. Got: {input.shape}')
if input.size(0) != target.size(0):
raise ValueError(
f'Expected input batch_size ({input.size(0)}) to match target batch_size ({target.size(0)}).'
)
probs_pos = torch.sigmoid(input)
probs_neg = torch.sigmoid(-input)
loss_tmp = -alpha * torch.pow(probs_neg, gamma) * target * F.logsigmoid(
input) - (1 - alpha) * torch.pow(probs_pos, gamma) * (1.0 - target
) * F.logsigmoid(-input)
if reduction == 'none':
loss = loss_tmp
elif reduction == 'mean':
loss = torch.mean(loss_tmp)
elif reduction == 'sum':
loss = torch.sum(loss_tmp)
else:
raise NotImplementedError(f'Invalid reduction mode: {reduction}')
return loss
class BinaryFocalLossWithLogits(nn.Module):
"""Criterion that computes Focal loss.
According to :cite:`lin2018focal`, the Focal loss is computed as follows:
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
alpha): Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
Shape:
- Input: :math:`(N, *)`.
- Target: :math:`(N, *)`.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> loss = BinaryFocalLossWithLogits(**kwargs)
>>> input = torch.randn(1, 3, 5, requires_grad=True)
>>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(2)
>>> output = loss(input, target)
>>> output.backward()
"""
def __init__(self, alpha: 'float', gamma: 'float'=2.0, reduction: 'str'
='none') ->None:
super().__init__()
self.alpha: 'float' = alpha
self.gamma: 'float' = gamma
self.reduction: 'str' = reduction
def forward(self, input: 'torch.Tensor', target: 'torch.Tensor'
) ->torch.Tensor:
return binary_focal_loss_with_logits(input, target, self.alpha,
self.gamma, self.reduction)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'alpha': 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 libdevice, math as tl_math
import warnings
from typing import Optional
import torch.nn as nn
import torch.nn.functional as F
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_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_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)
tmp6 = tl.load(in_ptr1 + x0, xmask)
tmp1 = -tmp0
tmp2 = tl.sigmoid(tmp1)
tmp3 = tmp2 * tmp2
tmp4 = -4.0
tmp5 = tmp3 * tmp4
tmp7 = tmp5 * tmp6
tmp8 = 0.0
tmp9 = triton_helpers.minimum(tmp8, tmp0)
tmp10 = tl_math.abs(tmp0)
tmp11 = -tmp10
tmp12 = tl_math.exp(tmp11)
tmp13 = libdevice.log1p(tmp12)
tmp14 = tmp9 - tmp13
tmp15 = tmp7 * tmp14
tmp16 = tl.sigmoid(tmp0)
tmp17 = tmp16 * tmp16
tmp18 = -3.0
tmp19 = tmp17 * tmp18
tmp20 = 1.0
tmp21 = tmp20 - tmp6
tmp22 = tmp19 * tmp21
tmp23 = triton_helpers.minimum(tmp8, tmp1)
tmp24 = tl_math.abs(tmp1)
tmp25 = -tmp24
tmp26 = tl_math.exp(tmp25)
tmp27 = libdevice.log1p(tmp26)
tmp28 = tmp23 - tmp27
tmp29 = tmp22 * tmp28
tmp30 = tmp15 - tmp29
tl.store(out_ptr0 + x0, tmp30, 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_log_sigmoid_forward_mul_neg_pow_rsub_sigmoid_sub_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 binary_focal_loss_with_logits(input: 'torch.Tensor', target:
'torch.Tensor', alpha: 'float'=0.25, gamma: 'float'=2.0, reduction:
'str'='none', eps: 'Optional[float]'=None) ->torch.Tensor:
"""Function that computes Binary Focal loss.
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
input: input data tensor of arbitrary shape.
target: the target tensor with shape matching input.
alpha: Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
eps: Deprecated: scalar for numerically stability when dividing. This is no longer used.
Returns:
the computed loss.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> logits = torch.tensor([[[6.325]],[[5.26]],[[87.49]]])
>>> labels = torch.tensor([[[1.]],[[1.]],[[0.]]])
>>> binary_focal_loss_with_logits(logits, labels, **kwargs)
tensor(21.8725)
"""
if eps is not None and not torch.jit.is_scripting():
warnings.warn(
'`binary_focal_loss_with_logits` has been reworked for improved numerical stability and the `eps` argument is no longer necessary'
, DeprecationWarning, stacklevel=2)
if not isinstance(input, torch.Tensor):
raise TypeError(f'Input type is not a torch.Tensor. Got {type(input)}')
if not len(input.shape) >= 2:
raise ValueError(
f'Invalid input shape, we expect BxCx*. Got: {input.shape}')
if input.size(0) != target.size(0):
raise ValueError(
f'Expected input batch_size ({input.size(0)}) to match target batch_size ({target.size(0)}).'
)
probs_pos = torch.sigmoid(input)
probs_neg = torch.sigmoid(-input)
loss_tmp = -alpha * torch.pow(probs_neg, gamma) * target * F.logsigmoid(
input) - (1 - alpha) * torch.pow(probs_pos, gamma) * (1.0 - target
) * F.logsigmoid(-input)
if reduction == 'none':
loss = loss_tmp
elif reduction == 'mean':
loss = torch.mean(loss_tmp)
elif reduction == 'sum':
loss = torch.sum(loss_tmp)
else:
raise NotImplementedError(f'Invalid reduction mode: {reduction}')
return loss
class BinaryFocalLossWithLogitsNew(nn.Module):
"""Criterion that computes Focal loss.
According to :cite:`lin2018focal`, the Focal loss is computed as follows:
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
alpha): Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
Shape:
- Input: :math:`(N, *)`.
- Target: :math:`(N, *)`.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> loss = BinaryFocalLossWithLogits(**kwargs)
>>> input = torch.randn(1, 3, 5, requires_grad=True)
>>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(2)
>>> output = loss(input, target)
>>> output.backward()
"""
def __init__(self, alpha: 'float', gamma: 'float'=2.0, reduction: 'str'
='none') ->None:
super().__init__()
self.alpha: 'float' = alpha
self.gamma: 'float' = gamma
self.reduction: 'str' = 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]
| shubham-gupta-iitr/mmmlX | BinaryFocalLossWithLogits | false | 4,329 | [
"Apache-2.0"
] | 0 | 3485e6191e0e45bf1c8168e4e928a36ab9264d22 | https://github.com/shubham-gupta-iitr/mmmlX/tree/3485e6191e0e45bf1c8168e4e928a36ab9264d22 | import torch
import warnings
from typing import Optional
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.data
def binary_focal_loss_with_logits(input: 'torch.Tensor', target:
'torch.Tensor', alpha: 'float'=0.25, gamma: 'float'=2.0, reduction:
'str'='none', eps: 'Optional[float]'=None) ->torch.Tensor:
"""Function that computes Binary Focal loss.
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
input: input data tensor of arbitrary shape.
target: the target tensor with shape matching input.
alpha: Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
eps: Deprecated: scalar for numerically stability when dividing. This is no longer used.
Returns:
the computed loss.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> logits = torch.tensor([[[6.325]],[[5.26]],[[87.49]]])
>>> labels = torch.tensor([[[1.]],[[1.]],[[0.]]])
>>> binary_focal_loss_with_logits(logits, labels, **kwargs)
tensor(21.8725)
"""
if eps is not None and not torch.jit.is_scripting():
warnings.warn(
'`binary_focal_loss_with_logits` has been reworked for improved numerical stability and the `eps` argument is no longer necessary'
, DeprecationWarning, stacklevel=2)
if not isinstance(input, torch.Tensor):
raise TypeError(f'Input type is not a torch.Tensor. Got {type(input)}')
if not len(input.shape) >= 2:
raise ValueError(
f'Invalid input shape, we expect BxCx*. Got: {input.shape}')
if input.size(0) != target.size(0):
raise ValueError(
f'Expected input batch_size ({input.size(0)}) to match target batch_size ({target.size(0)}).'
)
probs_pos = torch.sigmoid(input)
probs_neg = torch.sigmoid(-input)
loss_tmp = -alpha * torch.pow(probs_neg, gamma) * target * F.logsigmoid(
input) - (1 - alpha) * torch.pow(probs_pos, gamma) * (1.0 - target
) * F.logsigmoid(-input)
if reduction == 'none':
loss = loss_tmp
elif reduction == 'mean':
loss = torch.mean(loss_tmp)
elif reduction == 'sum':
loss = torch.sum(loss_tmp)
else:
raise NotImplementedError(f'Invalid reduction mode: {reduction}')
return loss
class Model(nn.Module):
"""Criterion that computes Focal loss.
According to :cite:`lin2018focal`, the Focal loss is computed as follows:
.. math::
\\text{FL}(p_t) = -\\alpha_t (1 - p_t)^{\\gamma} \\, \\text{log}(p_t)
where:
- :math:`p_t` is the model's estimated probability for each class.
Args:
alpha): Weighting factor for the rare class :math:`\\alpha \\in [0, 1]`.
gamma: Focusing parameter :math:`\\gamma >= 0`.
reduction: Specifies the reduction to apply to the
output: ``'none'`` | ``'mean'`` | ``'sum'``. ``'none'``: no reduction
will be applied, ``'mean'``: the sum of the output will be divided by
the number of elements in the output, ``'sum'``: the output will be
summed.
Shape:
- Input: :math:`(N, *)`.
- Target: :math:`(N, *)`.
Examples:
>>> kwargs = {"alpha": 0.25, "gamma": 2.0, "reduction": 'mean'}
>>> loss = BinaryFocalLossWithLogits(**kwargs)
>>> input = torch.randn(1, 3, 5, requires_grad=True)
>>> target = torch.empty(1, 3, 5, dtype=torch.long).random_(2)
>>> output = l
# ... truncated (>4000 chars) for memory efficiency |
FCDiscriminator | # 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_7/inductor_cache/xd/cxdqslrdqajmcxikxhvxi7lkzd2yepfzcwkkltrpstapeq35h632.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=[256, 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 = 256
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_7/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_7/inductor_cache/ra/crarmf7s2qf36jg27hprl42qtwcxcnnoyrgzgevtstzj4qgsdzwl.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=[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_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 = 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_7/inductor_cache/yd/cydvmxsmzwizyj5fbgjnjeeo27as6zdlft5s5uj57ovvcxtlbfhh.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=[32768, 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 = 32768
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 % 128
y1 = (yindex // 128)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (128*x2) + (1152*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/7v/c7vlmp4ptmjjinootrsb47fer72573dvgxb4w77hrarddids2b3i.py
# Unsorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
triton_poi_fused_4 = async_compile.triton('triton_poi_fused_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=[131072, 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_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_4(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 131072
xnumel = 9
yoffset = (tl.program_id(1) + tl.program_id(2) * tl.num_programs(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 % 256
y1 = (yindex // 256)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (256*x2) + (2304*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ww/cww4y7dmuzvb4hfmxcbqu4r6ro4pwuk7n4nibbuo64ghx7nhx65h.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
# 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, [2, 2], [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_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=[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_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 = 1024
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')
# kernel path: runs/run_shard_7/inductor_cache/c4/cc4krnccw3e7e5e5mcvkqs3amzwd3nlyvdfcp3tv77ivjyllg2et.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# x_2 => convolution_1
# x_3 => 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], [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_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=[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_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 = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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')
# kernel path: runs/run_shard_7/inductor_cache/5i/c5i4deb7lauisk6ikglle4ujq5ggctuvn42wmhz4nr7vbb2fxq2g.py
# Topologically Sorted Source Nodes: [x_4, x_5], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# x_4 => convolution_2
# x_5 => 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], [1, 1], [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_7 = async_compile.triton('triton_poi_fused_convolution_relu_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=[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_relu_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_relu_7(in_out_ptr0, in_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)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/bu/cbud4un5il2p2q6a2golnodgakkneqctkb7n5xqp76h67aumolea.py
# Topologically Sorted Source Nodes: [x_6, x_7], Original ATen: [aten.convolution, aten.leaky_relu]
# Source node to ATen node mapping:
# x_6 => convolution_3
# x_7 => gt, mul, where
# Graph fragment:
# %convolution_3 : [num_users=3] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_2, %primals_8, %primals_9, [2, 2], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution_3, 0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution_3, 0.2), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %convolution_3, %mul), kwargs = {})
triton_poi_fused_convolution_leaky_relu_8 = async_compile.triton('triton_poi_fused_convolution_leaky_relu_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=[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_leaky_relu_8', '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_leaky_relu_8(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 % 512
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, 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(in_out_ptr0 + (x2), tmp7, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ox/coxtwdkja7w74ybai7cjb3ivcyxelx5hodsgzoo6oforq7jadda7.py
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# x_8 => convolution_4
# Graph fragment:
# %convolution_4 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%where, %primals_10, %primals_11, [1, 1], [0, 0], [1, 1], False, [0, 0], 1), kwargs = {})
triton_poi_fused_convolution_9 = async_compile.triton('triton_poi_fused_convolution_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=[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_convolution_9', '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_9(in_out_ptr0, in_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_out_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr0 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tl.store(in_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, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11 = args
args.clear()
assert_size_stride(primals_1, (64, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (64, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (128, ), (1, ))
assert_size_stride(primals_6, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (256, ), (1, ))
assert_size_stride(primals_8, (512, 256, 3, 3), (2304, 9, 3, 1))
assert_size_stride(primals_9, (512, ), (1, ))
assert_size_stride(primals_10, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (1, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 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, 256, 9, grid=grid(256, 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((128, 64, 3, 3), (576, 1, 192, 64), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(primals_4, buf2, 8192, 9, grid=grid(8192, 9), stream=stream0)
del primals_4
buf3 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(primals_6, buf3, 32768, 9, grid=grid(32768, 9), stream=stream0)
del primals_6
buf4 = empty_strided_cuda((512, 256, 3, 3), (2304, 1, 768, 256), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_4.run(primals_8, buf4, 131072, 9, grid=grid(131072, 9), stream=stream0)
del primals_8
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.convolution]
buf5 = extern_kernels.convolution(buf1, buf0, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 64, 2, 2), (256, 1, 128, 64))
buf6 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_5.run(buf6, primals_2, 1024, grid=grid(1024), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf7 = extern_kernels.convolution(buf6, buf2, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (4, 128, 1, 1), (128, 1, 128, 128))
buf8 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_6.run(buf8, primals_5, 512, grid=grid(512), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.convolution]
buf9 = extern_kernels.convolution(buf8, buf3, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 256, 1, 1), (256, 1, 256, 256))
buf10 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [x_4, x_5], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_7.run(buf10, primals_7, 1024, grid=grid(1024), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.convolution]
buf11 = extern_kernels.convolution(buf10, buf4, stride=(2, 2), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 512, 1, 1), (512, 1, 512, 512))
buf12 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [x_6, x_7], Original ATen: [aten.convolution, aten.leaky_relu]
triton_poi_fused_convolution_leaky_relu_8.run(buf12, primals_9, 2048, grid=grid(2048), stream=stream0)
del primals_9
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.convolution]
buf13 = extern_kernels.convolution(buf12, primals_10, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf13, (4, 1, 1, 1), (1, 1, 1, 1))
buf14 = buf13; del buf13 # reuse
# Topologically Sorted Source Nodes: [x_8], Original ATen: [aten.convolution]
triton_poi_fused_convolution_9.run(buf14, primals_11, 4, grid=grid(4), stream=stream0)
del primals_11
return (buf14, buf0, buf1, buf2, buf3, buf4, primals_10, buf6, buf8, buf10, buf12, )
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, 4, 3, 3), (36, 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, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((128, 64, 3, 3), (576, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((256, 128, 3, 3), (1152, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((512, 256, 3, 3), (2304, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((1, 512, 1, 1), (512, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_11 = 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])
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 FCDiscriminator(nn.Module):
"""
inplanes, planes. Patch-gan
"""
def __init__(self, inplanes, planes=64):
super(FCDiscriminator, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.conv4 = nn.Conv2d(planes * 4, planes * 8, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 8, 1, kernel_size=1)
def forward(self, x):
x = self.conv1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.relu(x)
x = self.conv3(x)
x = self.relu(x)
x = self.conv4(x)
x = self.leaky_relu(x)
x = self.classifier(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'inplanes': 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.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_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 256
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 % 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_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 % 128
y1 = yindex // 128
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 128 * x2 + 1152 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_4(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = (tl.program_id(1) + tl.program_id(2) * tl.num_programs(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 % 256
y1 = yindex // 256
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 256 * x2 + 2304 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_5(in_out_ptr0, in_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 % 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)
@triton.jit
def triton_poi_fused_convolution_relu_6(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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)
@triton.jit
def triton_poi_fused_convolution_relu_7(in_out_ptr0, in_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)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_leaky_relu_8(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 % 512
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, 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(in_out_ptr0 + x2, tmp7, None)
@triton.jit
def triton_poi_fused_convolution_9(in_out_ptr0, in_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_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tl.store(in_out_ptr0 + x0, tmp3, 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, (64, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (64,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (128,), (1,))
assert_size_stride(primals_6, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (256,), (1,))
assert_size_stride(primals_8, (512, 256, 3, 3), (2304, 9, 3, 1))
assert_size_stride(primals_9, (512,), (1,))
assert_size_stride(primals_10, (1, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_11, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4, 3, 3), (36, 1, 12, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(256, 9)](primals_1, buf0, 256, 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((128, 64, 3, 3), (576, 1, 192, 64), torch
.float32)
triton_poi_fused_2[grid(8192, 9)](primals_4, buf2, 8192, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(32768, 9)](primals_6, buf3, 32768, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_6
buf4 = empty_strided_cuda((512, 256, 3, 3), (2304, 1, 768, 256),
torch.float32)
triton_poi_fused_4[grid(131072, 9)](primals_8, buf4, 131072, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_8
buf5 = extern_kernels.convolution(buf1, buf0, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 64, 2, 2), (256, 1, 128, 64))
buf6 = buf5
del buf5
triton_poi_fused_convolution_relu_5[grid(1024)](buf6, primals_2,
1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf7 = extern_kernels.convolution(buf6, buf2, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (4, 128, 1, 1), (128, 1, 128, 128))
buf8 = buf7
del buf7
triton_poi_fused_convolution_relu_6[grid(512)](buf8, primals_5, 512,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf9 = extern_kernels.convolution(buf8, buf3, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 256, 1, 1), (256, 1, 256, 256))
buf10 = buf9
del buf9
triton_poi_fused_convolution_relu_7[grid(1024)](buf10, primals_7,
1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_7
buf11 = extern_kernels.convolution(buf10, buf4, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf11, (4, 512, 1, 1), (512, 1, 512, 512))
buf12 = buf11
del buf11
triton_poi_fused_convolution_leaky_relu_8[grid(2048)](buf12,
primals_9, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
buf13 = extern_kernels.convolution(buf12, primals_10, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf13, (4, 1, 1, 1), (1, 1, 1, 1))
buf14 = buf13
del buf13
triton_poi_fused_convolution_9[grid(4)](buf14, primals_11, 4,
XBLOCK=4, num_warps=1, num_stages=1)
del primals_11
return (buf14, buf0, buf1, buf2, buf3, buf4, primals_10, buf6, buf8,
buf10, buf12)
class FCDiscriminatorNew(nn.Module):
"""
inplanes, planes. Patch-gan
"""
def __init__(self, inplanes, planes=64):
super(FCDiscriminatorNew, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.conv4 = nn.Conv2d(planes * 4, planes * 8, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 8, 1, kernel_size=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.conv4.weight
primals_9 = self.conv4.bias
primals_10 = self.classifier.weight
primals_11 = self.classifier.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])
return output[0]
| shiyutang/ProDA | FCDiscriminator | false | 4,330 | [
"MIT"
] | 0 | 38209ced03c6044743273bb60e07cd915ac2ae12 | https://github.com/shiyutang/ProDA/tree/38209ced03c6044743273bb60e07cd915ac2ae12 | import torch
import torch.nn as nn
import torch.utils.data
class Model(nn.Module):
"""
inplanes, planes. Patch-gan
"""
def __init__(self, inplanes, planes=64):
super().__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=3, stride=2,
padding=1)
self.conv2 = nn.Conv2d(planes, planes * 2, kernel_size=3, stride=2,
padding=1)
self.conv3 = nn.Conv2d(planes * 2, planes * 4, kernel_size=3,
stride=2, padding=1)
self.conv4 = nn.Conv2d(planes * 4, planes * 8, kernel_size=3,
stride=2, padding=1)
self.relu = nn.ReLU(inplace=True)
self.leaky_relu = nn.LeakyReLU(negative_slope=0.2, inplace=True)
self.classifier = nn.Conv2d(planes * 8, 1, kernel_size=1)
def forward(self, x):
x = self.conv1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.relu(x)
x = self.conv3(x)
x = self.relu(x)
x = self.conv4(x)
x = self.leaky_relu(x)
x = self.classifier(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
F1Loss | # 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_7/inductor_cache/nw/cnwgisju4h5iwbbibpm7ry7jyqdrctyoxqysjbzmmiwisfsn62pt.py
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.max]
# Source node to ATen node mapping:
# max_1 => max_1
# Graph fragment:
# %max_1 : [num_users=1] = call_function[target=torch.ops.aten.max.dim](args = (%arg0_1, 1), kwargs = {})
triton_poi_fused_max_0 = async_compile.triton('triton_poi_fused_max_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: '*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_max_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_max_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)
tmp1 = tl.load(in_ptr0 + (16 + x0 + (64*x1)), xmask)
tmp17 = tl.load(in_ptr0 + (32 + x0 + (64*x1)), xmask)
tmp32 = tl.load(in_ptr0 + (48 + x0 + (64*x1)), xmask)
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], 0, tl.int64)
tmp11 = tl.full([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], 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], 3, tl.int64)
tmp42 = tmp31 < tmp41
tmp43 = tmp40 & tmp42
tmp44 = tmp38 | tmp43
tmp45 = tl.where(tmp44, tmp30, tmp32)
tmp46 = tl.where(tmp44, tmp31, tmp41)
tl.store(out_ptr0 + (x2), tmp46, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/by/cbyhv46ouyjdmjbqo63rraqaf25opgthyk3njgkcpe27siagcayn.py
# Topologically Sorted Source Nodes: [scatter_, scatter__1, mul, true_positive, total_target, total_predict], Original ATen: [aten.scatter, aten.mul, aten.sum]
# Source node to ATen node mapping:
# mul => mul
# scatter_ => scatter_upon_const_tensor_1
# scatter__1 => scatter_upon_const_tensor
# total_predict => sum_3
# total_target => sum_2
# true_positive => sum_1
# Graph fragment:
# %scatter_upon_const_tensor_1 : [num_users=2] = call_function[target=torch._inductor.fx_passes.post_grad.scatter_upon_const_tensor](args = (), kwargs = {shape: [4, 4, 4, 4], background_val: 0, dtype: torch.int64, dim: 1, selector: %view, val: 1})
# %scatter_upon_const_tensor : [num_users=2] = call_function[target=torch._inductor.fx_passes.post_grad.scatter_upon_const_tensor](args = (), kwargs = {shape: [4, 4, 4, 4], background_val: 0, dtype: torch.int64, dim: 1, selector: %arg1_1, val: 1})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%scatter_upon_const_tensor_1, %scatter_upon_const_tensor), kwargs = {})
# %sum_1 : [num_users=2] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%mul, [2, 3]), kwargs = {})
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%scatter_upon_const_tensor, [2, 3]), kwargs = {})
# %sum_3 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%scatter_upon_const_tensor_1, [2, 3]), kwargs = {})
triton_per_fused_mul_scatter_sum_1 = async_compile.triton('triton_per_fused_mul_scatter_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=[16, 16],
reduction_hint=ReductionHint.DEFAULT,
filename=__file__,
triton_meta={'signature': {0: '*i64', 1: '*i64', 2: '*i64', 3: '*i64', 4: '*i64', 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, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mul_scatter_sum_1', '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_scatter_sum_1(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)
r2 = rindex
x1 = (xindex // 4)
x0 = xindex % 4
x3 = xindex
tmp0 = tl.load(in_ptr0 + (r2 + (16*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tl.load(in_ptr1 + (r2 + (16*x1)), xmask, eviction_policy='evict_last', other=0.0)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = tl.full([1, 1], 1, tl.int64)
tmp4 = tl.full([1, 1], 0, tl.int64)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp7 = tmp6 == tmp1
tmp8 = tl.where(tmp7, tmp3, tmp4)
tmp9 = tmp5 * tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.where(xmask, tmp10, 0)
tmp13 = tl.sum(tmp12, 1)[:, None]
tmp14 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp16 = tl.where(xmask, tmp14, 0)
tmp17 = tl.sum(tmp16, 1)[:, None]
tmp18 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp20 = tl.where(xmask, tmp18, 0)
tmp21 = tl.sum(tmp20, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp13, xmask)
tl.store(out_ptr1 + (x3), tmp17, xmask)
tl.store(out_ptr2 + (x3), tmp21, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/tn/ctnphcfld2tlzq5mhmsktmctpqwpgh3jaw2v2ziqc44bxlcpnpod.py
# Topologically Sorted Source Nodes: [add, mul_1, add_1, recall, mul_3, add_2, mul_2, add_3, precision, mul_4, add_4, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.add, aten.mul, aten.div, aten.mean, aten.rsub]
# Source node to ATen node mapping:
# add => add
# add_1 => add_1
# add_2 => add_2
# add_3 => add_3
# add_4 => add_4
# class_wise_loss => div_2
# loss => mean
# loss_1 => mul_5
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# mul_4 => mul_4
# precision => div_1
# recall => div
# sub => sub
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_1, 1e-06), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 1.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_2, 1e-06), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_1, %add_1), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, 2), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_1, 1e-06), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_2, 1.0), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%sum_3, 1e-06), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_2, %add_3), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_3, %div_1), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, %div_1), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul_4, %add_4), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%div_2,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (1, %mean), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, 1.0), kwargs = {})
triton_per_fused_add_div_mean_mul_rsub_2 = async_compile.triton('triton_per_fused_add_div_mean_mul_rsub_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, 16],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i64', 2: '*i64', 3: '*i64', 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, 5), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_add_div_mean_mul_rsub_2', '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_mean_mul_rsub_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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 + (r0), None)
tmp6 = tl.load(in_ptr1 + (r0), None)
tmp12 = tl.load(in_ptr2 + (r0), None)
tmp1 = tmp0.to(tl.float32)
tmp2 = 1e-06
tmp3 = tmp1 + tmp2
tmp4 = 1.0
tmp5 = tmp3 * tmp4
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp7 + tmp2
tmp9 = tmp5 / tmp8
tmp10 = 2.0
tmp11 = tmp9 * tmp10
tmp13 = tmp12.to(tl.float32)
tmp14 = tmp13 + tmp2
tmp15 = tmp5 / tmp14
tmp16 = tmp11 * tmp15
tmp17 = tmp9 + tmp15
tmp18 = tmp16 / tmp17
tmp19 = tl.broadcast_to(tmp18, [XBLOCK, RBLOCK])
tmp21 = tl.sum(tmp19, 1)[:, None]
tmp22 = 16.0
tmp23 = tmp21 / tmp22
tmp24 = tmp4 - tmp23
tmp25 = tmp24 * tmp4
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp25, 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, 1, 4, 4), (16, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.int64)
# Topologically Sorted Source Nodes: [max_1], Original ATen: [aten.max]
stream0 = get_raw_stream(0)
triton_poi_fused_max_0.run(arg0_1, buf0, 64, grid=grid(64), stream=stream0)
del arg0_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
# Topologically Sorted Source Nodes: [scatter_, scatter__1, mul, true_positive, total_target, total_predict], Original ATen: [aten.scatter, aten.mul, aten.sum]
triton_per_fused_mul_scatter_sum_1.run(buf0, arg1_1, buf1, buf2, buf3, 16, 16, grid=grid(16), stream=stream0)
del arg1_1
del buf0
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [add, mul_1, add_1, recall, mul_3, add_2, mul_2, add_3, precision, mul_4, add_4, class_wise_loss, loss, sub, loss_1], Original ATen: [aten.add, aten.mul, aten.div, aten.mean, aten.rsub]
triton_per_fused_add_div_mean_mul_rsub_2.run(buf5, buf1, buf2, buf3, 1, 16, grid=grid(1), stream=stream0)
del buf1
del buf2
del buf3
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, 1, 4, 4), (16, 16, 4, 1), device='cuda:0', dtype=torch.int64)
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class F1Loss(nn.Module):
"""F1Loss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', rp_weight=[1.0, 1.0], class_weight
=None, loss_weight=1.0):
super(F1Loss, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.rp_weight = rp_weight
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
_, maxpred = torch.max(predict, 1)
predict_onehot = torch.zeros(predict.size()).type_as(maxpred)
predict_onehot.scatter_(1, maxpred.view(N, 1, H, W), 1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
true_positive = torch.sum(predict_onehot * target_onehot, dim=(2, 3))
total_target = torch.sum(target_onehot, dim=(2, 3))
total_predict = torch.sum(predict_onehot, dim=(2, 3))
recall = self.rp_weight[0] * (true_positive + self.smooth) / (
total_target + self.smooth)
precision = self.rp_weight[1] * (true_positive + self.smooth) / (
total_predict + self.smooth)
class_wise_loss = 2 * recall * precision / (recall + precision)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
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._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
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_poi_fused_max_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)
tmp1 = tl.load(in_ptr0 + (16 + x0 + 64 * x1), xmask)
tmp17 = tl.load(in_ptr0 + (32 + x0 + 64 * x1), xmask)
tmp32 = tl.load(in_ptr0 + (48 + x0 + 64 * x1), xmask)
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], 0, tl.int64)
tmp11 = tl.full([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], 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], 3, tl.int64)
tmp42 = tmp31 < tmp41
tmp43 = tmp40 & tmp42
tmp44 = tmp38 | tmp43
tl.where(tmp44, tmp30, tmp32)
tmp46 = tl.where(tmp44, tmp31, tmp41)
tl.store(out_ptr0 + x2, tmp46, xmask)
@triton.jit
def triton_per_fused_mul_scatter_sum_1(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)
r2 = rindex
x1 = xindex // 4
x0 = xindex % 4
x3 = xindex
tmp0 = tl.load(in_ptr0 + (r2 + 16 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr1 + (r2 + 16 * x1), xmask, eviction_policy=
'evict_last', other=0.0)
tmp1 = x0
tmp2 = tmp0 == tmp1
tmp3 = tl.full([1, 1], 1, tl.int64)
tmp4 = tl.full([1, 1], 0, tl.int64)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp7 = tmp6 == tmp1
tmp8 = tl.where(tmp7, tmp3, tmp4)
tmp9 = tmp5 * tmp8
tmp10 = tl.broadcast_to(tmp9, [XBLOCK, RBLOCK])
tmp12 = tl.where(xmask, tmp10, 0)
tmp13 = tl.sum(tmp12, 1)[:, None]
tmp14 = tl.broadcast_to(tmp8, [XBLOCK, RBLOCK])
tmp16 = tl.where(xmask, tmp14, 0)
tmp17 = tl.sum(tmp16, 1)[:, None]
tmp18 = tl.broadcast_to(tmp5, [XBLOCK, RBLOCK])
tmp20 = tl.where(xmask, tmp18, 0)
tmp21 = tl.sum(tmp20, 1)[:, None]
tl.store(out_ptr0 + x3, tmp13, xmask)
tl.store(out_ptr1 + x3, tmp17, xmask)
tl.store(out_ptr2 + x3, tmp21, xmask)
@triton.jit
def triton_per_fused_add_div_mean_mul_rsub_2(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, 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 + r0, None)
tmp6 = tl.load(in_ptr1 + r0, None)
tmp12 = tl.load(in_ptr2 + r0, None)
tmp1 = tmp0.to(tl.float32)
tmp2 = 1e-06
tmp3 = tmp1 + tmp2
tmp4 = 1.0
tmp5 = tmp3 * tmp4
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp7 + tmp2
tmp9 = tmp5 / tmp8
tmp10 = 2.0
tmp11 = tmp9 * tmp10
tmp13 = tmp12.to(tl.float32)
tmp14 = tmp13 + tmp2
tmp15 = tmp5 / tmp14
tmp16 = tmp11 * tmp15
tmp17 = tmp9 + tmp15
tmp18 = tmp16 / tmp17
tmp19 = tl.broadcast_to(tmp18, [XBLOCK, RBLOCK])
tmp21 = tl.sum(tmp19, 1)[:, None]
tmp22 = 16.0
tmp23 = tmp21 / tmp22
tmp24 = tmp4 - tmp23
tmp25 = tmp24 * tmp4
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp25, 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, 1, 4, 4), (16, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.int64)
get_raw_stream(0)
triton_poi_fused_max_0[grid(64)](arg0_1, buf0, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.int64)
triton_per_fused_mul_scatter_sum_1[grid(16)](buf0, arg1_1, buf1,
buf2, buf3, 16, 16, XBLOCK=1, num_warps=2, num_stages=1)
del arg1_1
del buf0
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4
del buf4
triton_per_fused_add_div_mean_mul_rsub_2[grid(1)](buf5, buf1, buf2,
buf3, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del buf1
del buf2
del buf3
return buf5,
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class F1LossNew(nn.Module):
"""F1Loss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', rp_weight=[1.0, 1.0], class_weight
=None, loss_weight=1.0):
super(F1LossNew, self).__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.rp_weight = rp_weight
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| shuaizzZ/mmsegmentation | F1Loss | false | 4,331 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
import torch.nn.functional as F
from typing import *
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:
assert weight.dim() == loss.dim()
if weight.dim() > 1:
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
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
class Model(nn.Module):
"""F1Loss.
Args:
reduction (str, optional): . Defaults to 'mean'.
Options are "none", "mean" and "sum".
class_weight (list[float], optional): Weight of each class.
Defaults to None.
loss_weight (float, optional): Weight of the loss. Defaults to 1.0.
"""
def __init__(self, reduction='mean', rp_weight=[1.0, 1.0], class_weight
=None, loss_weight=1.0):
super().__init__()
assert reduction in ('none', 'mean', 'sum')
self.reduction = reduction
self.rp_weight = rp_weight
self.loss_weight = loss_weight
self.class_weight = class_weight
self.smooth = 1e-06
def forward(self, predict, target, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
"""Forward function."""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if self.class_weight is not None:
class_weight = torch.tensor(self.class_weight).type_as(predict)
else:
class_weight = None
N, _C, H, W = predict.size()
_, maxpred = torch.max(predict, 1)
predict_onehot = torch.zeros(predict.size()).type_as(maxpred)
predict_onehot.scatter_(1, maxpred.view(N, 1, H, W), 1)
target_onehot = torch.zeros(predict.size()).type_as(target)
target_onehot.scatter_(1, target.view(N, 1, H, W), 1)
true_positive = torch.sum(predict_onehot * target_onehot, dim=(2, 3))
total_target = torch.sum(target_onehot, dim=(2, 3))
total_predict = torch.sum(predict_onehot, dim=(2, 3))
recall = self.rp_weight[0] * (true_positive + self.smooth) / (
total_target + self.smooth)
precision = self.rp_weight[1] * (true_positive + self.smooth) / (
total_predict + self.smooth)
class_wise_loss = 2 * recall * precision / (recall + precision)
if class_weight is not None:
class_wise_loss = class_wise_loss * class_weight
loss = self.loss_weight * (1 - weight_reduce_loss(class_wise_loss,
weight, reduction=reduction, avg_factor=avg_factor))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4, 1, 4, 4], dtype=torch.
int64)]
def get_init_inputs():
return []
|
NonLocal | # 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_7/inductor_cache/v2/cv2ouxsqm3ibwyduv3ncuj6czgk4hqfxu2z5kqnwbp7fsakanj7z.py
# Topologically Sorted Source Nodes: [x_theta], Original ATen: [aten.clone, aten.transpose]
# Source node to ATen node mapping:
# x_theta => clone
# Graph fragment:
# %clone : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute,), kwargs = {memory_format: torch.contiguous_format})
# %permute_6 : [num_users=1] = call_function[target=torch.ops.aten.permute.default](args = (%expand, [0, 2, 1]), kwargs = {})
triton_poi_fused_clone_transpose_0 = async_compile.triton('triton_poi_fused_clone_transpose_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_clone_transpose_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_clone_transpose_0(in_out_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
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
x1 = xindex
y0 = yindex
y2 = yindex % 4
y3 = (yindex // 4)
tmp0 = tl.load(in_out_ptr0 + (x1 + (8*y0)), xmask & ymask, eviction_policy='evict_last')
tl.debug_barrier()
tl.store(in_out_ptr0 + (x1 + (8*y0)), tmp0, xmask & ymask)
tl.store(out_ptr0 + (y2 + (4*x1) + (32*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/br/cbrfdr5npcq3x6tg57wssigjzvvzcx35pamxdd4ajtovfeddfzbz.py
# Topologically Sorted Source Nodes: [mul_theta_phi_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# mul_theta_phi_1 => amax, exp, sub, sum_1
# 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 = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [1], True), kwargs = {})
triton_per_fused__softmax_1 = async_compile.triton('triton_per_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.persistent_reduction(
size_hints=[32, 8],
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': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__softmax_1', '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_1(in_ptr0, out_ptr0, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 32
rnumel = 8
RBLOCK: tl.constexpr = 8
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
x0 = xindex % 8
x1 = (xindex // 8)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (8*r2) + (64*x1)), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tl.store(out_ptr0 + (x3), tmp4, xmask)
tl.store(out_ptr1 + (x3), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/f5/cf52j4jrhglkyce37k4fpaq5cdeovfdq25q4wtnaymuu6ylbumhr.py
# Topologically Sorted Source Nodes: [mul_theta_phi_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# mul_theta_phi_1 => div, exp, sub
# Graph fragment:
# %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 = {})
# %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=[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__softmax_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__softmax_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
x3 = xindex
x0 = xindex % 8
x2 = (xindex // 64)
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x0 + (8*x2)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (x0 + (8*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp3 = tl_math.exp(tmp2)
tmp5 = tmp3 / tmp4
tl.store(in_out_ptr0 + (x3), tmp5, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/gi/cgi4kf6ohfvkft7ahcg4eogwuv3zzp5uy3pehfpsij6m4yk4s2aw.py
# Topologically Sorted Source Nodes: [contiguous_2, mul_theta_phi_g_1], Original ATen: [aten.clone, aten.view]
# Source node to ATen node mapping:
# contiguous_2 => clone_2
# mul_theta_phi_g_1 => view_9
# 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_9 : [num_users=2] = call_function[target=torch.ops.aten.reshape.default](args = (%clone_2, [4, 2, 4, 4]), kwargs = {})
triton_poi_fused_clone_view_3 = async_compile.triton('triton_poi_fused_clone_view_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, 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_clone_view_3', '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_clone_view_3(in_out_ptr0, in_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 16
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 % 4
y1 = (yindex // 4)
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*x2) + (32*y1)), xmask & ymask, eviction_policy='evict_last')
tl.debug_barrier()
tl.store(in_out_ptr0 + (x2 + (8*y3)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/sz/cszpyxrnf5or42o33rncwa5m67aqfmsrvr7zvazievvfxzm2y2cl.py
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.add]
# Source node to ATen node mapping:
# out => add
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convolution_3, %primals_1), 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=[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_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_add_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
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + (x0), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask)
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x0), 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 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_3, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_4, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (4, 2, 1, 1), (2, 1, 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_1, 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, 2, 4, 4), (32, 16, 4, 1))
# Topologically Sorted Source Nodes: [conv2d_1], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(primals_1, primals_3, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 4, 4), (32, 16, 4, 1))
# Topologically Sorted Source Nodes: [conv2d_2], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(primals_1, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 2, 4, 4), (32, 16, 4, 1))
buf3 = reinterpret_tensor(buf1, (4, 8, 4), (32, 1, 8), 0); del buf1 # reuse
buf15 = empty_strided_cuda((4, 4, 8), (32, 1, 4), torch.float32)
# Topologically Sorted Source Nodes: [x_theta], Original ATen: [aten.clone, aten.transpose]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_transpose_0.run(buf3, buf15, 16, 8, grid=grid(16, 8), stream=stream0)
buf4 = empty_strided_cuda((4, 8, 8), (64, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_theta, mul_theta_phi], Original ATen: [aten.clone, aten.bmm]
extern_kernels.bmm(buf3, reinterpret_tensor(buf0, (4, 4, 8), (32, 8, 1), 0), out=buf4)
buf5 = empty_strided_cuda((4, 1, 8), (8, 32, 1), torch.float32)
buf6 = empty_strided_cuda((4, 1, 8), (8, 32, 1), torch.float32)
# Topologically Sorted Source Nodes: [mul_theta_phi_1], Original ATen: [aten._softmax]
triton_per_fused__softmax_1.run(buf4, buf5, buf6, 32, 8, grid=grid(32), stream=stream0)
buf7 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [mul_theta_phi_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf7, buf5, buf6, 256, grid=grid(256), stream=stream0)
del buf5
del buf6
buf8 = reinterpret_tensor(buf2, (4, 8, 4), (32, 1, 8), 0); del buf2 # reuse
buf14 = reinterpret_tensor(buf3, (4, 4, 8), (32, 1, 4), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [x_g], Original ATen: [aten.clone, aten.transpose]
triton_poi_fused_clone_transpose_0.run(buf8, buf14, 16, 8, grid=grid(16, 8), stream=stream0)
buf9 = empty_strided_cuda((4, 8, 4), (32, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_g, mul_theta_phi_g], Original ATen: [aten.clone, aten.bmm]
extern_kernels.bmm(buf7, buf8, out=buf9)
buf10 = reinterpret_tensor(buf8, (4, 4, 8), (32, 8, 1), 0); del buf8 # reuse
buf11 = reinterpret_tensor(buf10, (4, 2, 4, 4), (32, 16, 4, 1), 0); del buf10 # reuse
# Topologically Sorted Source Nodes: [contiguous_2, mul_theta_phi_g_1], Original ATen: [aten.clone, aten.view]
triton_poi_fused_clone_view_3.run(buf11, buf9, 16, 8, grid=grid(16, 8), stream=stream0)
del buf9
# Topologically Sorted Source Nodes: [mask], Original ATen: [aten.convolution]
buf12 = extern_kernels.convolution(buf11, primals_5, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 4, 4, 4), (64, 16, 4, 1))
buf13 = buf12; del buf12 # reuse
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.add]
triton_poi_fused_add_4.run(buf13, primals_1, 256, grid=grid(256), stream=stream0)
return (buf13, primals_1, primals_2, primals_3, primals_4, primals_5, buf7, buf11, buf14, buf15, reinterpret_tensor(buf0, (4, 8, 4), (32, 1, 8), 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((2, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 2, 1, 1), (2, 1, 1, 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._C
import torch.serialization
from torch import nn
from typing import *
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class NonLocal(nn.Module):
def __init__(self, in_channels):
super(NonLocal, self).__init__()
self.inter_channel = in_channels // 2
self.conv_phi = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_theta = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_g = nn.Conv2d(in_channels=in_channels, out_channels=self.
inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.softmax = nn.Softmax(dim=1)
self.conv_mask = nn.Conv2d(in_channels=self.inter_channel,
out_channels=in_channels, kernel_size=1, stride=1, padding=0,
bias=False)
def forward(self, x):
b, c, h, w = int_size(x)
x_phi = self.conv_phi(x).view(b, c, -1)
x_theta = self.conv_theta(x).view(b, c, -1).permute(0, 2, 1
).contiguous()
x_g = self.conv_g(x).view(b, c, -1).permute(0, 2, 1).contiguous()
mul_theta_phi = torch.matmul(x_theta, x_phi)
mul_theta_phi = self.softmax(mul_theta_phi)
mul_theta_phi_g = torch.matmul(mul_theta_phi, x_g)
mul_theta_phi_g = mul_theta_phi_g.permute(0, 2, 1).contiguous().view(b,
self.inter_channel, h, w)
mask = self.conv_mask(mul_theta_phi_g)
out = mask + x
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_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 torch._C
import torch.serialization
from torch import nn
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_clone_transpose_0(in_out_ptr0, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
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
x1 = xindex
y0 = yindex
y2 = yindex % 4
y3 = yindex // 4
tmp0 = tl.load(in_out_ptr0 + (x1 + 8 * y0), xmask & ymask,
eviction_policy='evict_last')
tl.debug_barrier()
tl.store(in_out_ptr0 + (x1 + 8 * y0), tmp0, xmask & ymask)
tl.store(out_ptr0 + (y2 + 4 * x1 + 32 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_per_fused__softmax_1(in_ptr0, out_ptr0, out_ptr1, xnumel, rnumel,
XBLOCK: tl.constexpr):
xnumel = 32
RBLOCK: tl.constexpr = 8
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
x0 = xindex % 8
x1 = xindex // 8
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 8 * r2 + 64 * x1), xmask, other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.where(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(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tl.store(out_ptr0 + x3, tmp4, xmask)
tl.store(out_ptr1 + x3, tmp10, xmask)
@triton.jit
def triton_poi_fused__softmax_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
x3 = xindex
x0 = xindex % 8
x2 = xindex // 64
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 8 * x2), xmask, eviction_policy='evict_last'
)
tmp4 = tl.load(in_ptr1 + (x0 + 8 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 - tmp1
tmp3 = tl_math.exp(tmp2)
tmp5 = tmp3 / tmp4
tl.store(in_out_ptr0 + x3, tmp5, xmask)
@triton.jit
def triton_poi_fused_clone_view_3(in_out_ptr0, in_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
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 % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 32 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.debug_barrier()
tl.store(in_out_ptr0 + (x2 + 8 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_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
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask)
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x0, tmp2, 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, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_3, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_4, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (4, 2, 1, 1), (2, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_1, 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, 2, 4, 4), (32, 16, 4, 1))
buf1 = extern_kernels.convolution(primals_1, primals_3, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 4, 4), (32, 16, 4, 1))
buf2 = extern_kernels.convolution(primals_1, primals_4, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 2, 4, 4), (32, 16, 4, 1))
buf3 = reinterpret_tensor(buf1, (4, 8, 4), (32, 1, 8), 0)
del buf1
buf15 = empty_strided_cuda((4, 4, 8), (32, 1, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_transpose_0[grid(16, 8)](buf3, buf15, 16, 8,
XBLOCK=8, YBLOCK=16, num_warps=4, num_stages=1)
buf4 = empty_strided_cuda((4, 8, 8), (64, 8, 1), torch.float32)
extern_kernels.bmm(buf3, reinterpret_tensor(buf0, (4, 4, 8), (32, 8,
1), 0), out=buf4)
buf5 = empty_strided_cuda((4, 1, 8), (8, 32, 1), torch.float32)
buf6 = empty_strided_cuda((4, 1, 8), (8, 32, 1), torch.float32)
triton_per_fused__softmax_1[grid(32)](buf4, buf5, buf6, 32, 8,
XBLOCK=32, num_warps=2, num_stages=1)
buf7 = buf4
del buf4
triton_poi_fused__softmax_2[grid(256)](buf7, buf5, buf6, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del buf5
del buf6
buf8 = reinterpret_tensor(buf2, (4, 8, 4), (32, 1, 8), 0)
del buf2
buf14 = reinterpret_tensor(buf3, (4, 4, 8), (32, 1, 4), 0)
del buf3
triton_poi_fused_clone_transpose_0[grid(16, 8)](buf8, buf14, 16, 8,
XBLOCK=8, YBLOCK=16, num_warps=4, num_stages=1)
buf9 = empty_strided_cuda((4, 8, 4), (32, 4, 1), torch.float32)
extern_kernels.bmm(buf7, buf8, out=buf9)
buf10 = reinterpret_tensor(buf8, (4, 4, 8), (32, 8, 1), 0)
del buf8
buf11 = reinterpret_tensor(buf10, (4, 2, 4, 4), (32, 16, 4, 1), 0)
del buf10
triton_poi_fused_clone_view_3[grid(16, 8)](buf11, buf9, 16, 8,
XBLOCK=8, YBLOCK=16, num_warps=4, num_stages=1)
del buf9
buf12 = extern_kernels.convolution(buf11, primals_5, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 4, 4, 4), (64, 16, 4, 1))
buf13 = buf12
del buf12
triton_poi_fused_add_4[grid(256)](buf13, primals_1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
return (buf13, primals_1, primals_2, primals_3, primals_4, primals_5,
buf7, buf11, buf14, buf15, reinterpret_tensor(buf0, (4, 8, 4), (32,
1, 8), 0))
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class NonLocalNew(nn.Module):
def __init__(self, in_channels):
super(NonLocalNew, self).__init__()
self.inter_channel = in_channels // 2
self.conv_phi = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_theta = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_g = nn.Conv2d(in_channels=in_channels, out_channels=self.
inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.softmax = nn.Softmax(dim=1)
self.conv_mask = nn.Conv2d(in_channels=self.inter_channel,
out_channels=in_channels, kernel_size=1, stride=1, padding=0,
bias=False)
def forward(self, input_0):
primals_2 = self.conv_phi.weight
primals_3 = self.conv_theta.weight
primals_4 = self.conv_g.weight
primals_5 = self.conv_mask.weight
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| shuaizzZ/mmsegmentation | NonLocal | false | 4,332 | [
"Apache-2.0"
] | 0 | a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | https://github.com/shuaizzZ/mmsegmentation/tree/a6c6b348dbf8c4a0a39ffbdb832a1e82309c533c | import torch
import torch._C
import torch.serialization
from torch import nn
from typing import *
def int_size(x):
size = tuple(int(s) for s in x.size())
return size
class Model(nn.Module):
def __init__(self, in_channels):
super().__init__()
self.inter_channel = in_channels // 2
self.conv_phi = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_theta = nn.Conv2d(in_channels=in_channels, out_channels=
self.inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.conv_g = nn.Conv2d(in_channels=in_channels, out_channels=self.
inter_channel, kernel_size=1, stride=1, padding=0, bias=False)
self.softmax = nn.Softmax(dim=1)
self.conv_mask = nn.Conv2d(in_channels=self.inter_channel,
out_channels=in_channels, kernel_size=1, stride=1, padding=0,
bias=False)
def forward(self, x):
b, c, h, w = int_size(x)
x_phi = self.conv_phi(x).view(b, c, -1)
x_theta = self.conv_theta(x).view(b, c, -1).permute(0, 2, 1
).contiguous()
x_g = self.conv_g(x).view(b, c, -1).permute(0, 2, 1).contiguous()
mul_theta_phi = torch.matmul(x_theta, x_phi)
mul_theta_phi = self.softmax(mul_theta_phi)
mul_theta_phi_g = torch.matmul(mul_theta_phi, x_g)
mul_theta_phi_g = mul_theta_phi_g.permute(0, 2, 1).contiguous().view(b,
self.inter_channel, h, w)
mask = self.conv_mask(mul_theta_phi_g)
out = mask + x
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
BertOutput | # 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_7/inductor_cache/ss/cssn3ayzwsxbizosd6ieezxafjef3fxscx57lbnlxbdiuph3p2je.py
# Topologically Sorted Source Nodes: [add, u], Original ATen: [aten.add, aten.mean]
# Source node to ATen node mapping:
# add => add
# u => mean
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_4), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add, [-1], True), kwargs = {})
triton_poi_fused_add_mean_0 = async_compile.triton('triton_poi_fused_add_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.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_0', '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_add_mean_0(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 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp4 = tl.load(in_ptr2 + (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))
tmp8 = tl.broadcast_to(tmp7, [XBLOCK])
tmp10 = tl.load(in_ptr2 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (2))
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp17 = tl.load(in_ptr2 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp21 = tl.load(in_ptr1 + (3))
tmp22 = tl.broadcast_to(tmp21, [XBLOCK])
tmp24 = tl.load(in_ptr2 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tmp0 + tmp2
tmp5 = tmp3 + tmp4
tmp9 = tmp6 + tmp8
tmp11 = tmp9 + tmp10
tmp12 = tmp5 + tmp11
tmp16 = tmp13 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp12 + tmp18
tmp23 = tmp20 + tmp22
tmp25 = tmp23 + tmp24
tmp26 = tmp19 + tmp25
tmp27 = 4.0
tmp28 = tmp26 / tmp27
tl.store(out_ptr0 + (x0), tmp28, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/l6/cl6vibrzoyykzmbhmvlsdaksh3k2diif7eg66z2ho46tjsy6emma.py
# Topologically Sorted Source Nodes: [add, sub], Original ATen: [aten.add, aten.sub]
# Source node to ATen node mapping:
# add => add
# sub => sub
# Graph fragment:
# %add : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_4), kwargs = {})
# %sub : [num_users=3] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %mean), kwargs = {})
triton_poi_fused_add_sub_1 = async_compile.triton('triton_poi_fused_add_sub_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_add_sub_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_add_sub_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
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')
tmp3 = tl.load(in_ptr1 + (x2), xmask)
tmp5 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/6d/c6dv445enwddy5hcfn3p4vvjynq4oa2vnla2acdsag6hrxww6urd.py
# Topologically Sorted Source Nodes: [pow_1, s, add_1, sqrt, x, mul, hidden_states_2], Original ATen: [aten.pow, aten.mean, aten.add, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# add_1 => add_1
# hidden_states_2 => add_2
# mul => mul
# pow_1 => pow_1
# s => mean_1
# sqrt => sqrt
# x => div
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [-1], True), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mean_1, 1e-05), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add_1,), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, %sqrt), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_5, %div), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_6), kwargs = {})
triton_poi_fused_add_div_mean_mul_pow_sqrt_2 = async_compile.triton('triton_poi_fused_add_div_mean_mul_pow_sqrt_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_add_div_mean_mul_pow_sqrt_2', '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_pow_sqrt_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
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')
tmp4 = tl.load(in_ptr1 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = 4.0
tmp14 = tmp12 / tmp13
tmp15 = 1e-05
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp1 / tmp17
tmp19 = tmp0 * tmp18
tmp21 = tmp19 + 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, primals_3, primals_4, primals_5, primals_6 = 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))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (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, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [add, u], Original ATen: [aten.add, aten.mean]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mean_0.run(buf0, primals_2, primals_4, buf1, 64, grid=grid(64), stream=stream0)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [add, sub], Original ATen: [aten.add, aten.sub]
triton_poi_fused_add_sub_1.run(buf2, primals_2, primals_4, buf1, 256, grid=grid(256), stream=stream0)
del buf1
del primals_2
del primals_4
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_1, s, add_1, sqrt, x, mul, hidden_states_2], Original ATen: [aten.pow, aten.mean, aten.add, aten.sqrt, aten.div, aten.mul]
triton_poi_fused_add_div_mean_mul_pow_sqrt_2.run(primals_5, buf2, primals_6, buf3, 256, grid=grid(256), stream=stream0)
del primals_6
return (buf3, primals_5, reinterpret_tensor(primals_3, (64, 4), (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), (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)
primals_5 = rand_strided((4, ), (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)
| from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
import torch.utils.data
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super(BertLayerNorm, self).__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class BertOutput(nn.Module):
def __init__(self, config):
super(BertOutput, self).__init__()
self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, hidden_states, input_tensor):
hidden_states = self.dense(hidden_states)
hidden_states = self.dropout(hidden_states)
hidden_states = self.LayerNorm(hidden_states + input_tensor)
return hidden_states
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'config': _mock_config(intermediate_size=4, hidden_size=4,
hidden_dropout_prob=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 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_add_mean_0(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 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp4 = tl.load(in_ptr2 + 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)
tmp8 = tl.broadcast_to(tmp7, [XBLOCK])
tmp10 = tl.load(in_ptr2 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp14 = tl.load(in_ptr1 + 2)
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp17 = tl.load(in_ptr2 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp21 = tl.load(in_ptr1 + 3)
tmp22 = tl.broadcast_to(tmp21, [XBLOCK])
tmp24 = tl.load(in_ptr2 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp3 = tmp0 + tmp2
tmp5 = tmp3 + tmp4
tmp9 = tmp6 + tmp8
tmp11 = tmp9 + tmp10
tmp12 = tmp5 + tmp11
tmp16 = tmp13 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp12 + tmp18
tmp23 = tmp20 + tmp22
tmp25 = tmp23 + tmp24
tmp26 = tmp19 + tmp25
tmp27 = 4.0
tmp28 = tmp26 / tmp27
tl.store(out_ptr0 + x0, tmp28, xmask)
@triton.jit
def triton_poi_fused_add_sub_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
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')
tmp3 = tl.load(in_ptr1 + x2, xmask)
tmp5 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tl.store(in_out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_div_mean_mul_pow_sqrt_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
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')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = 4.0
tmp14 = tmp12 / tmp13
tmp15 = 1e-05
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp1 / tmp17
tmp19 = tmp0 * tmp18
tmp21 = tmp19 + tmp20
tl.store(out_ptr0 + 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, (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))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (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, 1), (16, 4, 1, 64), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mean_0[grid(64)](buf0, primals_2, primals_4,
buf1, 64, XBLOCK=64, num_warps=1, num_stages=1)
buf2 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
triton_poi_fused_add_sub_1[grid(256)](buf2, primals_2, primals_4,
buf1, 256, XBLOCK=256, num_warps=4, num_stages=1)
del buf1
del primals_2
del primals_4
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_pow_sqrt_2[grid(256)](primals_5,
buf2, primals_6, buf3, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_6
return buf3, primals_5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf2
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super(BertLayerNorm, self).__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class BertOutputNew(nn.Module):
def __init__(self, config):
super(BertOutputNew, self).__init__()
self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, input_0, input_1):
primals_1 = self.dense.weight
primals_2 = self.dense.bias
primals_5 = self.LayerNorm.weight
primals_6 = self.LayerNorm.bias
primals_3 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| shubham-gupta-iitr/mmmlX | BertOutput | false | 4,333 | [
"Apache-2.0"
] | 0 | 3485e6191e0e45bf1c8168e4e928a36ab9264d22 | https://github.com/shubham-gupta-iitr/mmmlX/tree/3485e6191e0e45bf1c8168e4e928a36ab9264d22 | from _paritybench_helpers import _mock_config
import torch
import torch.nn as nn
import torch.utils.data
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super().__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class Model(nn.Module):
def __init__(self, config):
super().__init__()
self.dense = nn.Linear(config.intermediate_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, hidden_states, input_tensor):
hidden_states = self.dense(hidden_states)
hidden_states = self.dropout(hidden_states)
hidden_states = self.LayerNorm(hidden_states + input_tensor)
return hidden_states
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'config': _mock_config(intermediate_size=4, hidden_size=4,
hidden_dropout_prob=0.5)}]
|
GELU | # 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_7/inductor_cache/v7/cv7humnywkkqhrumbeetegqlkretdwtkj5pcanrbgxrolupvobzt.py
# Topologically Sorted Source Nodes: [mul_2, pow_1, mul, geluPow, mul_1, geluTanh, geluResult, mul_3], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
# Source node to ATen node mapping:
# geluPow => add
# geluResult => add_1
# geluTanh => tanh
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# pow_1 => pow_1
# Graph fragment:
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 0.5), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%arg0_1, 3), kwargs = {})
# %mul : [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 = (%arg0_1, %mul), kwargs = {})
# %mul_1 : [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_1,), 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_2, %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):
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: [mul_2, pow_1, mul, geluPow, mul_1, geluTanh, geluResult, mul_3], Original ATen: [aten.mul, aten.pow, aten.add, aten.tanh]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mul_pow_tanh_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 math
import torch
from torch import nn
class GELU(nn.Module):
def __init__(self):
super(GELU, self).__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
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
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_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):
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_mul_pow_tanh_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class GELUNew(nn.Module):
def __init__(self):
super(GELUNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| simonepreite/QABERT | GELU | false | 4,334 | [
"MIT"
] | 0 | ed3e49f6619f3ff660068291231909693cb8f5d5 | https://github.com/simonepreite/QABERT/tree/ed3e49f6619f3ff660068291231909693cb8f5d5 | import math
import torch
from torch import nn
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
RefModel1d | # 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_7/inductor_cache/q2/cq224fondqmxhhj34xqz6dnrl4uhzkcy5iwjlkcclhniwzabrr5u.py
# Topologically Sorted Source Nodes: [output, output_1, output_2], Original ATen: [aten.convolution, aten.repeat, aten._native_batch_norm_legit, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# output => convolution
# output_1 => add, repeat, rsqrt, var_mean
# output_2 => relu
# Graph fragment:
# %convolution : [num_users=2] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1], [0], [1], False, [0], 1), kwargs = {})
# %repeat : [num_users=2] = call_function[target=torch.ops.aten.repeat.default](args = (%primals_4, [4]), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view, [0, 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=2] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add,), kwargs = {})
# %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_per_fused__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_0 = async_compile.triton('triton_per_fused__native_batch_norm_legit_convolution_relu_repeat_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.persistent_reduction(
size_hints=[8, 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: '*i1', 8: '*fp32', 9: 'i32', 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, 10), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_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__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_0(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, out_ptr3, out_ptr4, out_ptr5, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 8
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)
x0 = xindex
r3 = rindex
x1 = xindex % 2
tmp0 = tl.load(in_ptr0 + (x0 % 2), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_out_ptr0 + (r3 + (64*x0)), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp28 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(xmask, tmp4, 0)
tmp7 = tl.broadcast_to(tmp4, [XBLOCK, RBLOCK])
tmp9 = tl.where(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tl.full([XBLOCK, 1], 64, tl.int32)
tmp12 = tmp11.to(tl.float32)
tmp13 = tmp10 / tmp12
tmp14 = tmp4 - tmp13
tmp15 = tmp14 * tmp14
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp18 = tl.where(xmask, tmp16, 0)
tmp19 = tl.sum(tmp18, 1)[:, None]
tmp20 = tmp3 - tmp13
tmp21 = 64.0
tmp22 = tmp19 / tmp21
tmp23 = 1e-05
tmp24 = tmp22 + tmp23
tmp25 = libdevice.rsqrt(tmp24)
tmp26 = tmp20 * tmp25
tmp27 = tmp26 * tmp0
tmp29 = tmp27 + tmp28
tmp30 = tl.full([1, 1], 0, tl.int32)
tmp31 = triton_helpers.maximum(tmp30, tmp29)
tmp32 = 0.0
tmp33 = tmp31 <= tmp32
tl.store(out_ptr0 + (x0), tmp0, xmask)
tl.store(in_out_ptr0 + (r3 + (64*x0)), tmp3, xmask)
tl.store(out_ptr3 + (r3 + (64*x0)), tmp31, xmask)
tl.store(out_ptr4 + (r3 + (64*x0)), tmp33, xmask)
tl.store(out_ptr5 + (x0), tmp25, xmask)
tl.store(out_ptr1 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ev/cevv6aa7t3gurbbr3vzw6nmgkpoyoyj5qjr5m57z6nbfkg42bgpf.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
# Source node to ATen node mapping:
# output_5 => add_2, clamp_min, convert_element_type, convert_element_type_1, iota, mul_2, sub_1
# Graph fragment:
# %iota : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (64,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%iota, torch.float32), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type, 0.5), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_2, 0.5), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_2, 0.5), kwargs = {})
# %clamp_min : [num_users=2] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_1, 0.0), kwargs = {})
# %convert_element_type_1 : [num_users=4] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%clamp_min, torch.int64), kwargs = {})
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_1 = async_compile.triton('triton_poi_fused__to_copy_add_arange_clamp_mul_sub_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: '*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, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_add_arange_clamp_mul_sub_1', '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__to_copy_add_arange_clamp_mul_sub_1(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ek/cektoo3xtedaewlh5uggdyf55krfjuty35h3vjq6vtyduxqrlkz4.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.add, aten.clamp]
# Source node to ATen node mapping:
# output_5 => add_3, clamp_max
# Graph fragment:
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type_1, 1), kwargs = {})
# %clamp_max : [num_users=2] = call_function[target=torch.ops.aten.clamp_max.default](args = (%add_3, 31), kwargs = {})
triton_poi_fused_add_clamp_2 = async_compile.triton('triton_poi_fused_add_clamp_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: '*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, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_clamp_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_add_clamp_2(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 31, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + (x0), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/a3/ca3np32wv5647cru4u4cskmo7z65jffrdabbplzceq4wcduwuwh7.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
# Source node to ATen node mapping:
# output_5 => add_2, clamp_max_1, clamp_min, clamp_min_1, convert_element_type, iota, mul_2, sub_1, sub_2
# Graph fragment:
# %iota : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (64,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%iota, torch.float32), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type, 0.5), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_2, 0.5), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_2, 0.5), kwargs = {})
# %clamp_min : [num_users=2] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_1, 0.0), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clamp_min, %convert_element_type_1), kwargs = {})
# %clamp_min_1 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_2, 0.0), kwargs = {})
# %clamp_max_1 : [num_users=2] = call_function[target=torch.ops.aten.clamp_max.default](args = (%clamp_min_1, 1.0), kwargs = {})
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_3 = async_compile.triton('triton_poi_fused__to_copy_add_arange_clamp_mul_sub_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: 'i32'}, 'device': 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__to_copy_add_arange_clamp_mul_sub_3', '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__to_copy_add_arange_clamp_mul_sub_3(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/2s/c2sh7tk7ss66nqj24rjxor3fsmcsc4smwnnopzbmlf3gm46qf7mk.py
# Topologically Sorted Source Nodes: [output_6], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# output_6 => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%unsqueeze_3, [-1, -2], True), kwargs = {})
triton_per_fused_mean_4 = async_compile.triton('triton_per_fused_mean_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=[8, 64],
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': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mean_4', '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_4(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 8
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
tmp22 = tl.load(in_ptr1 + (r1), None, eviction_policy='evict_last')
tmp0 = r1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + ((2*tmp8) + (64*x0)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (1 + (2*tmp8) + (64*x0)), xmask, eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tmp11 * tmp2
tmp13 = tl.full([1, 1], 1, tl.int64)
tmp14 = tmp8 + tmp13
tmp15 = tl.full([1, 1], 31, tl.int64)
tmp16 = triton_helpers.minimum(tmp14, tmp15)
tmp17 = tl.load(in_ptr0 + ((2*tmp16) + (64*x0)), xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (1 + (2*tmp16) + (64*x0)), xmask, eviction_policy='evict_last')
tmp19 = tmp18 + tmp17
tmp20 = tmp19 * tmp2
tmp21 = tmp20 - tmp12
tmp23 = tmp21 * tmp22
tmp24 = tmp12 + tmp23
tmp25 = tl.broadcast_to(tmp24, [XBLOCK, RBLOCK])
tmp27 = tl.where(xmask, tmp25, 0)
tmp28 = tl.sum(tmp27, 1)[:, None]
tmp29 = 64.0
tmp30 = tmp28 / tmp29
tl.debug_barrier()
tl.store(in_out_ptr0 + (x0), 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 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 1), (2, 1, 1))
assert_size_stride(primals_2, (2, ), (1, ))
assert_size_stride(primals_3, (4, 2, 64), (128, 64, 1))
assert_size_stride(primals_4, (2, ), (1, ))
assert_size_stride(primals_5, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf0, (4, 2, 64), (128, 64, 1))
buf2 = empty_strided_cuda((8, ), (1, ), torch.float32)
buf1 = buf0; del buf0 # reuse
buf3 = empty_strided_cuda((1, 8, 1), (8, 1, 8), torch.float32)
buf7 = empty_strided_cuda((4, 2, 64), (128, 64, 1), torch.float32)
buf13 = empty_strided_cuda((4, 2, 64), (128, 64, 1), torch.bool)
buf6 = empty_strided_cuda((1, 8, 1), (8, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [output, output_1, output_2], Original ATen: [aten.convolution, aten.repeat, aten._native_batch_norm_legit, aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_per_fused__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_0.run(buf1, primals_4, primals_2, primals_5, buf2, buf3, buf7, buf13, buf6, 8, 64, grid=grid(8), stream=stream0)
del primals_2
del primals_4
del primals_5
buf8 = empty_strided_cuda((64, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_1.run(buf8, 64, grid=grid(64), stream=stream0)
buf9 = empty_strided_cuda((64, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.add, aten.clamp]
triton_poi_fused_add_clamp_2.run(buf9, 64, grid=grid(64), stream=stream0)
buf10 = empty_strided_cuda((64, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_3.run(buf10, 64, grid=grid(64), stream=stream0)
buf11 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 8, 8), torch.float32)
buf12 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [output_6], Original ATen: [aten.mean]
triton_per_fused_mean_4.run(buf12, buf7, buf10, 8, 64, grid=grid(8), stream=stream0)
return (reinterpret_tensor(buf12, (4, 2, 1), (2, 1, 1), 0), primals_1, primals_3, buf1, buf2, reinterpret_tensor(buf6, (8, ), (1, ), 0), reinterpret_tensor(buf7, (4, 2, 1, 64), (128, 64, 64, 1), 0), buf8, buf9, buf10, buf13, reinterpret_tensor(buf3, (1, 8, 1), (8, 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((2, 2, 1), (2, 1, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 2, 64), (128, 64, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((2, ), (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
class RefModel1d(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv1d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm1d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout(0.2)
self.l5 = torch.nn.AvgPool1d(2)
self.l7 = torch.nn.AdaptiveAvgPool1d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='linear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 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 libdevice
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__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_0(
in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, out_ptr3,
out_ptr4, out_ptr5, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 8
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)
x0 = xindex
r3 = rindex
x1 = xindex % 2
tmp0 = tl.load(in_ptr0 + x0 % 2, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_out_ptr0 + (r3 + 64 * x0), xmask, other=0.0)
tmp2 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp28 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tl.where(xmask, tmp4, 0)
tmp7 = tl.broadcast_to(tmp4, [XBLOCK, RBLOCK])
tmp9 = tl.where(xmask, tmp7, 0)
tmp10 = tl.sum(tmp9, 1)[:, None]
tmp11 = tl.full([XBLOCK, 1], 64, tl.int32)
tmp12 = tmp11.to(tl.float32)
tmp13 = tmp10 / tmp12
tmp14 = tmp4 - tmp13
tmp15 = tmp14 * tmp14
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp18 = tl.where(xmask, tmp16, 0)
tmp19 = tl.sum(tmp18, 1)[:, None]
tmp20 = tmp3 - tmp13
tmp21 = 64.0
tmp22 = tmp19 / tmp21
tmp23 = 1e-05
tmp24 = tmp22 + tmp23
tmp25 = libdevice.rsqrt(tmp24)
tmp26 = tmp20 * tmp25
tmp27 = tmp26 * tmp0
tmp29 = tmp27 + tmp28
tmp30 = tl.full([1, 1], 0, tl.int32)
tmp31 = triton_helpers.maximum(tmp30, tmp29)
tmp32 = 0.0
tmp33 = tmp31 <= tmp32
tl.store(out_ptr0 + x0, tmp0, xmask)
tl.store(in_out_ptr0 + (r3 + 64 * x0), tmp3, xmask)
tl.store(out_ptr3 + (r3 + 64 * x0), tmp31, xmask)
tl.store(out_ptr4 + (r3 + 64 * x0), tmp33, xmask)
tl.store(out_ptr5 + x0, tmp25, xmask)
tl.store(out_ptr1 + x0, tmp13, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_clamp_mul_sub_1(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + x0, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_clamp_2(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 31, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + x0, tmp12, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_clamp_mul_sub_3(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + x0, tmp13, xmask)
@triton.jit
def triton_per_fused_mean_4(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel,
XBLOCK: tl.constexpr):
xnumel = 8
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
tmp22 = tl.load(in_ptr1 + r1, None, eviction_policy='evict_last')
tmp0 = r1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.load(in_ptr0 + (2 * tmp8 + 64 * x0), xmask, eviction_policy=
'evict_last')
tmp10 = tl.load(in_ptr0 + (1 + 2 * tmp8 + 64 * x0), xmask,
eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tmp11 * tmp2
tmp13 = tl.full([1, 1], 1, tl.int64)
tmp14 = tmp8 + tmp13
tmp15 = tl.full([1, 1], 31, tl.int64)
tmp16 = triton_helpers.minimum(tmp14, tmp15)
tmp17 = tl.load(in_ptr0 + (2 * tmp16 + 64 * x0), xmask, eviction_policy
='evict_last')
tmp18 = tl.load(in_ptr0 + (1 + 2 * tmp16 + 64 * x0), xmask,
eviction_policy='evict_last')
tmp19 = tmp18 + tmp17
tmp20 = tmp19 * tmp2
tmp21 = tmp20 - tmp12
tmp23 = tmp21 * tmp22
tmp24 = tmp12 + tmp23
tmp25 = tl.broadcast_to(tmp24, [XBLOCK, RBLOCK])
tmp27 = tl.where(xmask, tmp25, 0)
tmp28 = tl.sum(tmp27, 1)[:, None]
tmp29 = 64.0
tmp30 = tmp28 / tmp29
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp30, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 1), (2, 1, 1))
assert_size_stride(primals_2, (2,), (1,))
assert_size_stride(primals_3, (4, 2, 64), (128, 64, 1))
assert_size_stride(primals_4, (2,), (1,))
assert_size_stride(primals_5, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf0, (4, 2, 64), (128, 64, 1))
buf2 = empty_strided_cuda((8,), (1,), torch.float32)
buf1 = buf0
del buf0
buf3 = empty_strided_cuda((1, 8, 1), (8, 1, 8), torch.float32)
buf7 = empty_strided_cuda((4, 2, 64), (128, 64, 1), torch.float32)
buf13 = empty_strided_cuda((4, 2, 64), (128, 64, 1), torch.bool)
buf6 = empty_strided_cuda((1, 8, 1), (8, 1, 1), torch.float32)
get_raw_stream(0)
triton_per_fused__native_batch_norm_legit_convolution_relu_repeat_threshold_backward_0[
grid(8)](buf1, primals_4, primals_2, primals_5, buf2, buf3,
buf7, buf13, buf6, 8, 64, XBLOCK=1, num_warps=2, num_stages=1)
del primals_2
del primals_4
del primals_5
buf8 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_1[grid(64)](buf8,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf9 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_add_clamp_2[grid(64)](buf9, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf10 = empty_strided_cuda((64,), (1,), torch.float32)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_3[grid(64)](buf10,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf11 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 8, 8), torch.float32)
buf12 = buf11
del buf11
triton_per_fused_mean_4[grid(8)](buf12, buf7, buf10, 8, 64, XBLOCK=
1, num_warps=2, num_stages=1)
return reinterpret_tensor(buf12, (4, 2, 1), (2, 1, 1), 0
), primals_1, primals_3, buf1, buf2, reinterpret_tensor(buf6, (8,),
(1,), 0), reinterpret_tensor(buf7, (4, 2, 1, 64), (128, 64, 64, 1), 0
), buf8, buf9, buf10, buf13, reinterpret_tensor(buf3, (1, 8, 1), (8,
1, 1), 0)
class RefModel1dNew(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv1d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm1d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout(0.2)
self.l5 = torch.nn.AvgPool1d(2)
self.l7 = torch.nn.AdaptiveAvgPool1d(1)
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]
| shuohan/pytorch-layers | RefModel1d | false | 4,335 | [
"MIT"
] | 0 | 020846fd02d501cf477552179c19ba4b5e9a0695 | https://github.com/shuohan/pytorch-layers/tree/020846fd02d501cf477552179c19ba4b5e9a0695 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv1d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm1d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout(0.2)
self.l5 = torch.nn.AvgPool1d(2)
self.l7 = torch.nn.AdaptiveAvgPool1d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='linear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 64])]
def get_init_inputs():
return []
|
ScaledDotProduct | # 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_7/inductor_cache/6u/c6uffoadh2wyrqz23qy2vnhmakrv6j5onhv4qjgj37eyzzbw5wl6.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %constant_pad_nd_default : [num_users=1] = call_function[target=torch.ops.aten.constant_pad_nd.default](args = (%arg2_1, [0, 4], 0.0), 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=[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_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, xnumel, XBLOCK : tl.constexpr):
xnumel = 512
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], 4, tl.int64)
tmp2 = tmp0 < tmp1
tmp3 = tl.load(in_ptr0 + (x0 + (4*x1)), tmp2 & xmask, other=0.0)
tl.store(out_ptr0 + (x2), tmp3, xmask)
''', 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)
buf0 = empty_strided_cuda((4, 4, 4, 8), (128, 32, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(arg2_1, buf0, 512, grid=grid(512), stream=stream0)
del arg2_1
# Topologically Sorted Source Nodes: [], Original ATen: []
buf1 = torch.ops.aten._scaled_dot_product_efficient_attention.default(arg1_1, arg0_1, arg3_1, reinterpret_tensor(buf0, (4, 4, 4, 4), (128, 32, 8, 1), 0), False)
del arg0_1
del arg1_1
del arg3_1
del buf0
buf2 = buf1[0]
del buf1
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)
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 math
import torch
from torch import nn
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProduct, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
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 [[], {'attentionHeadSize': 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 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_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 512
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], 4, tl.int64)
tmp2 = tmp0 < tmp1
tmp3 = tl.load(in_ptr0 + (x0 + 4 * x1), tmp2 & xmask, other=0.0)
tl.store(out_ptr0 + x2, tmp3, xmask)
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)
buf0 = empty_strided_cuda((4, 4, 4, 8), (128, 32, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(512)](arg2_1, buf0, 512, XBLOCK=256,
num_warps=4, num_stages=1)
del arg2_1
buf1 = torch.ops.aten._scaled_dot_product_efficient_attention.default(
arg1_1, arg0_1, arg3_1, reinterpret_tensor(buf0, (4, 4, 4, 4),
(128, 32, 8, 1), 0), False)
del arg0_1
del arg1_1
del arg3_1
del buf0
buf2 = buf1[0]
del buf1
return buf2,
class ScaledDotProductNew(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProductNew, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
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]
| simonepreite/QABERT | ScaledDotProduct | false | 4,336 | [
"MIT"
] | 0 | ed3e49f6619f3ff660068291231909693cb8f5d5 | https://github.com/simonepreite/QABERT/tree/ed3e49f6619f3ff660068291231909693cb8f5d5 | import math
import torch
from torch import nn
class Model(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super().__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
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 [4]
|
FeedForward | # 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_7/inductor_cache/v7/cv7humnywkkqhrumbeetegqlkretdwtkj5pcanrbgxrolupvobzt.py
# Topologically Sorted Source Nodes: [pow_1, mul, geluPow, mul_1, geluTanh, geluResult, mul_2, intermediate], Original ATen: [aten.pow, aten.mul, aten.add, aten.tanh]
# Source node to ATen node mapping:
# geluPow => add
# geluResult => add_1
# geluTanh => tanh
# intermediate => mul_3
# mul => mul
# mul_1 => mul_1
# mul_2 => mul_2
# pow_1 => pow_1
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_1, 3), kwargs = {})
# %mul : [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), kwargs = {})
# %mul_1 : [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_1,), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%tanh, 1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, 0.5), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_2, %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: [pow_1, mul, geluPow, mul_1, geluTanh, geluResult, mul_2, intermediate], Original ATen: [aten.pow, aten.mul, 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: [linearOut], 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
from torch import nn
class GELU(nn.Module):
def __init__(self):
super(GELU, self).__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class FeedForward(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super(FeedForward, self).__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, tensor):
intermediate = self.activationFuncion(self.w1(tensor))
linearOut = self.w2(intermediate)
return self.dropout(linearOut)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'hiddenSize': 4, 'innerLayerDimension': 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
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_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 __init__(self):
super(GELU, self).__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class FeedForwardNew(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super(FeedForwardNew, self).__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, input_0):
primals_1 = self.w1.weight
primals_2 = self.w1.bias
primals_4 = self.w2.weight
primals_5 = self.w2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| simonepreite/QABERT | FeedForward | false | 4,337 | [
"MIT"
] | 0 | ed3e49f6619f3ff660068291231909693cb8f5d5 | https://github.com/simonepreite/QABERT/tree/ed3e49f6619f3ff660068291231909693cb8f5d5 | import math
import torch
from torch import nn
class GELU(nn.Module):
def __init__(self):
super().__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class Model(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super().__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, tensor):
intermediate = self.activationFuncion(self.w1(tensor))
linearOut = self.w2(intermediate)
return self.dropout(linearOut)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
RenormSoftmax | # 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_7/inductor_cache/wx/cwxwvlntewdrqi2r4caciy5ht4jdvafnhtiqncr4lo4aegcb4imz.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, [-1], 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
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_7/inductor_cache/hx/chxqytbjkrdreqavf4zgnnsibcuybgzguaaqoznho3mny2s2rcmp.py
# Topologically Sorted Source Nodes: [softmax, mul, mul_1], Original ATen: [aten._softmax, aten.mul]
# Source node to ATen node mapping:
# mul => mul
# mul_1 => mul_1
# softmax => 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 = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%div, 4), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, 0.07853981633974483), kwargs = {})
triton_poi_fused__softmax_mul_1 = async_compile.triton('triton_poi_fused__softmax_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=[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_mul_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_mul_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
tmp9 = 4.0
tmp10 = tmp8 * tmp9
tmp11 = 0.07853981633974483
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + (x2), 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, 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, mul, mul_1], Original ATen: [aten._softmax, aten.mul]
triton_poi_fused__softmax_mul_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 numpy as np
import torch.nn as nn
class RenormSoftmax(nn.Module):
def __init__(self, dim=-1, norm=np.pi / 40):
super().__init__()
self.softmax = nn.Softmax(dim=dim)
self.dim = dim
self.norm = norm
def forward(self, x):
N = x.shape[self.dim]
return self.softmax(x) * N * self.norm
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 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__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)
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_mul_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
tmp9 = 4.0
tmp10 = tmp8 * tmp9
tmp11 = 0.07853981633974483
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + x2, tmp12, 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=
256, 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_mul_1[grid(256)](buf0, buf1, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del buf0
return buf1,
class RenormSoftmaxNew(nn.Module):
def __init__(self, dim=-1, norm=np.pi / 40):
super().__init__()
self.softmax = nn.Softmax(dim=dim)
self.dim = dim
self.norm = norm
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| simonverret/deep_continuation | RenormSoftmax | false | 4,338 | [
"MIT"
] | 0 | 986bfba7f6806dc4869a023ff1fc1d0d18324b25 | https://github.com/simonverret/deep_continuation/tree/986bfba7f6806dc4869a023ff1fc1d0d18324b25 | import torch
import numpy as np
import torch.nn as nn
class Model(nn.Module):
def __init__(self, dim=-1, norm=np.pi / 40):
super().__init__()
self.softmax = nn.Softmax(dim=dim)
self.dim = dim
self.norm = norm
def forward(self, x):
N = x.shape[self.dim]
return self.softmax(x) * N * self.norm
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
BertAttention | # 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_7/inductor_cache/x2/cx2hdvwyo7m5jvhhvtugzxqvmy6z4nsfhkkjhvgzbbm3cb6dsum2.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %mul_scalar : [num_users=1] = call_function[target=torch.ops.aten.mul.Scalar](args = (%permute_default, 1.0), kwargs = {})
# %clone_default : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/iz/ciztqj6kop3hxov46yrmzprkzfir3eljcic4mkqznz2j5cfeaudr.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %sum_dim_int_list : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_default, [-1], True), kwargs = {})
# %eq_scalar : [num_users=1] = call_function[target=torch.ops.aten.eq.Scalar](args = (%add_tensor, -inf), kwargs = {})
# %logical_not_default : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%eq_scalar,), kwargs = {})
# %any_dim : [num_users=1] = call_function[target=torch.ops.aten.any.dim](args = (%logical_not_default, -1, True), 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=[64],
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + (4*x2), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x2)), 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*x2)), 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*x2)), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float("-inf")
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = (tmp29 != 0)
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = (tmp33 != 0)
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = (tmp38 != 0)
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = (tmp43 != 0)
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + (x2), tmp14, xmask)
tl.store(out_ptr1 + (x2), tmp25, xmask)
tl.store(out_ptr2 + (x2), tmp45, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/x5/cx5uvbfethxuwwkwxf3xaualzhlcwqsz4jxqpbhintggaypzjwqf.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_default, %sum_dim_int_list), kwargs = {})
# %logical_not_default_1 : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%any_dim,), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4, 4, 4], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_self : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%logical_not_default_1, %full_default, %div_tensor), kwargs = {})
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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 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_2', '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_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = (xindex // 4)
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last').to(tl.int1)
tmp2 = tl.load(in_out_ptr0 + (x4), xmask)
tmp3 = tl.load(in_ptr1 + (x5), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + (x3), xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/vv/cvvnhithjvmvhfjufxwwzclfobkrgbyyteg66hp24r675f7elw4c.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %clone_default_2 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default_3,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/6t/c6t5a5ere3lqjiu7zh3uu4oxmpdoujdaqqmeunxqapgzo4m74uav.py
# Topologically Sorted Source Nodes: [context_layer_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# context_layer_1 => 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')
# kernel path: runs/run_shard_7/inductor_cache/hk/chkirlrxzb52fxbrq2rynamgt7aligt77yn6j6ihfk46whjvd374.py
# Topologically Sorted Source Nodes: [add_1, u, sub, pow_1, s], Original ATen: [aten.add, aten.mean, aten.sub, aten.pow]
# Source node to ATen node mapping:
# add_1 => add_1
# pow_1 => pow_1
# s => mean_1
# sub => sub_1
# u => mean
# Graph fragment:
# %add_1 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_17, %primals_3), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_1, [-1], True), kwargs = {})
# %sub_1 : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_1, %mean), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub_1, 2), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [-1], True), kwargs = {})
triton_poi_fused_add_mean_pow_sub_5 = async_compile.triton('triton_poi_fused_add_mean_pow_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=[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_mean_pow_sub_5', '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_mean_pow_sub_5(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_7/inductor_cache/yq/cyqp2dtjka2kdwget3aoidhkk72sxga5xdfgoyaskxfpk55rnzdw.py
# Topologically Sorted Source Nodes: [add_1, u, sub, add_2, sqrt, x_3, mul, hidden_states_2], Original ATen: [aten.add, aten.mean, aten.sub, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# add_1 => add_1
# add_2 => add_2
# hidden_states_2 => add_3
# mul => mul
# sqrt => sqrt
# sub => sub_1
# u => mean
# x_3 => div_2
# Graph fragment:
# %add_1 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_17, %primals_3), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_1, [-1], True), kwargs = {})
# %sub_1 : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_1, %mean), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mean_1, 1e-05), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add_2,), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_1, %sqrt), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_11, %div_2), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_12), kwargs = {})
triton_poi_fused_add_div_mean_mul_sqrt_sub_6 = async_compile.triton('triton_poi_fused_add_div_mean_mul_sqrt_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.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_div_mean_mul_sqrt_sub_6', '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_sqrt_sub_6(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
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 = 1e-05
tmp8 = tmp6 + tmp7
tmp9 = libdevice.sqrt(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, primals_8, primals_9, primals_10, primals_11, primals_12 = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, ), (1, ))
assert_size_stride(primals_11, (4, ), (1, ))
assert_size_stride(primals_12, (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 = 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_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_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(buf0, primals_2, buf3, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_0.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: [], Original ATen: []
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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0); del buf1 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(buf5, primals_8, buf6, buf7, buf8, 64, grid=grid(64), stream=stream0)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(buf9, buf8, primals_8, buf6, buf7, 256, grid=grid(256), stream=stream0)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(buf2, primals_7, buf10, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [context_layer_1], Original ATen: [aten.clone]
triton_poi_fused_clone_4.run(buf11, buf12, 16, 4, grid=grid(16, 4), stream=stream0)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0); del buf11 # reuse
# Topologically Sorted Source Nodes: [hidden_states], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4), (4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf13)
del primals_10
buf14 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf15 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [add_1, u, sub, pow_1, s], Original ATen: [aten.add, aten.mean, aten.sub, aten.pow]
triton_poi_fused_add_mean_pow_sub_5.run(buf13, primals_3, buf14, buf15, 16, grid=grid(16), stream=stream0)
buf16 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [add_1, u, sub, add_2, sqrt, x_3, mul, hidden_states_2], Original ATen: [aten.add, aten.mean, aten.sub, aten.sqrt, aten.div, aten.mul]
triton_poi_fused_add_div_mean_mul_sqrt_sub_6.run(primals_11, buf13, primals_3, buf14, buf15, primals_12, buf16, 64, grid=grid(64), stream=stream0)
del buf14
del buf15
del primals_12
return (buf16, primals_3, primals_11, buf9, reinterpret_tensor(buf10, (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), reinterpret_tensor(buf12, (16, 4), (4, 1), 0), buf13, primals_9, )
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)
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), (16, 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, ), 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)
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])
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 math
import torch
import torch.nn as nn
import torch.utils.data
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super(BertLayerNorm, self).__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class BertSelfAttention(nn.Module):
def __init__(self, config):
super(BertSelfAttention, self).__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = x.view(*new_x_shape)
return x.permute(0, 2, 1, 3)
def forward(self, hidden_states, attention_mask, history_states=None):
if history_states is None:
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(hidden_states)
mixed_value_layer = self.value(hidden_states)
else:
x_states = torch.cat((history_states, hidden_states), dim=1)
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(x_states)
mixed_value_layer = self.value(x_states)
query_layer = self.transpose_for_scores(mixed_query_layer)
key_layer = self.transpose_for_scores(mixed_key_layer)
value_layer = self.transpose_for_scores(mixed_value_layer)
attention_scores = torch.matmul(query_layer, key_layer.transpose(-1,
-2))
attention_scores = attention_scores / math.sqrt(self.
attention_head_size)
attention_scores = attention_scores + attention_mask
attention_probs = nn.Softmax(dim=-1)(attention_scores)
attention_probs = self.dropout(attention_probs)
context_layer = torch.matmul(attention_probs, value_layer)
context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
new_context_layer_shape = context_layer.size()[:-2] + (self.
all_head_size,)
context_layer = context_layer.view(*new_context_layer_shape)
return context_layer
class BertSelfOutput(nn.Module):
def __init__(self, config):
super(BertSelfOutput, self).__init__()
self.dense = nn.Linear(config.hidden_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, hidden_states, input_tensor):
hidden_states = self.dense(hidden_states)
hidden_states = self.dropout(hidden_states)
hidden_states = self.LayerNorm(hidden_states + input_tensor)
return hidden_states
class BertAttention(nn.Module):
def __init__(self, config):
super(BertAttention, self).__init__()
self.self = BertSelfAttention(config)
self.output = BertSelfOutput(config)
def forward(self, input_tensor, attention_mask, history_states=None):
self_output = self.self(input_tensor, attention_mask,
history_states=history_states)
attention_output = self.output(self_output, input_tensor)
return attention_output
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'config': _mock_config(hidden_size=4, num_attention_heads=
4, attention_probs_dropout_prob=0.5, hidden_dropout_prob=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 import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import math
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + 4 * x2, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x2), 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 * x2), 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 * x2), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float('-inf')
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = tmp29 != 0
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = tmp33 != 0
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = tmp38 != 0
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = tmp43 != 0
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + x2, tmp14, xmask)
tl.store(out_ptr1 + x2, tmp25, xmask)
tl.store(out_ptr2 + x2, tmp45, xmask)
@triton.jit
def triton_poi_fused_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 4
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last').to(tl
.int1)
tmp2 = tl.load(in_out_ptr0 + x4, xmask)
tmp3 = tl.load(in_ptr1 + x5, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + x3, xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_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_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)
@triton.jit
def triton_poi_fused_add_mean_pow_sub_5(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_div_mean_mul_sqrt_sub_6(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
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 = 1e-05
tmp8 = tmp6 + tmp7
tmp9 = libdevice.sqrt(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, primals_8, primals_9, primals_10, primals_11, primals_12
) = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4,), (1,))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (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 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (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_3, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(16, 4)](buf0, primals_2, buf3, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_0[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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0)
del buf1
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
triton_poi_fused_1[grid(64)](buf5, primals_8, buf6, buf7, buf8, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused_2[grid(256)](buf9, buf8, primals_8, buf6, buf7,
256, XBLOCK=128, num_warps=4, num_stages=1)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf7
triton_poi_fused_3[grid(16, 4)](buf2, primals_7, buf10, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf6
triton_poi_fused_clone_4[grid(16, 4)](buf11, buf12, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0)
del buf11
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf13)
del primals_10
buf14 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf15 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused_add_mean_pow_sub_5[grid(16)](buf13, primals_3,
buf14, buf15, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf16 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_sqrt_sub_6[grid(64)](primals_11,
buf13, primals_3, buf14, buf15, primals_12, buf16, 64, XBLOCK=
64, num_warps=1, num_stages=1)
del buf14
del buf15
del primals_12
return buf16, primals_3, primals_11, buf9, reinterpret_tensor(buf10, (
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
), reinterpret_tensor(buf12, (16, 4), (4, 1), 0), buf13, primals_9
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super(BertLayerNorm, self).__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class BertSelfAttention(nn.Module):
def __init__(self, config):
super(BertSelfAttention, self).__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = x.view(*new_x_shape)
return x.permute(0, 2, 1, 3)
def forward(self, hidden_states, attention_mask, history_states=None):
if history_states is None:
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(hidden_states)
mixed_value_layer = self.value(hidden_states)
else:
x_states = torch.cat((history_states, hidden_states), dim=1)
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(x_states)
mixed_value_layer = self.value(x_states)
query_layer = self.transpose_for_scores(mixed_query_layer)
key_layer = self.transpose_for_scores(mixed_key_layer)
value_layer = self.transpose_for_scores(mixed_value_layer)
attention_scores = torch.matmul(query_layer, key_layer.transpose(-1,
-2))
attention_scores = attention_scores / math.sqrt(self.
attention_head_size)
attention_scores = attention_scores + attention_mask
attention_probs = nn.Softmax(dim=-1)(attention_scores)
attention_probs = self.dropout(attention_probs)
context_layer = torch.matmul(attention_probs, value_layer)
context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
new_context_layer_shape = context_layer.size()[:-2] + (self.
all_head_size,)
context_layer = context_layer.view(*new_context_layer_shape)
return context_layer
class BertSelfOutput(nn.Module):
def __init__(self, config):
super(BertSelfOutput, self).__init__()
self.dense = nn.Linear(config.hidden_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, hidden_states, input_tensor):
hidden_states = self.dense(hidden_states)
hidden_states = self.dropout(hidden_states)
hidden_states = self.LayerNorm(hidden_states + input_tensor)
return hidden_states
class BertAttentionNew(nn.Module):
def __init__(self, config):
super(BertAttentionNew, self).__init__()
self.self = BertSelfAttention(config)
self.output = BertSelfOutput(config)
def forward(self, input_0, input_1):
primals_1 = self.self.query.weight
primals_2 = self.self.query.bias
primals_4 = self.self.key.weight
primals_5 = self.self.key.bias
primals_6 = self.self.value.weight
primals_7 = self.self.value.bias
primals_9 = self.output.dense.weight
primals_10 = self.output.dense.bias
primals_11 = self.output.LayerNorm.weight
primals_12 = self.output.LayerNorm.bias
primals_3 = input_0
primals_8 = input_1
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])
return output[0]
| shubham-gupta-iitr/mmmlX | BertAttention | false | 4,339 | [
"Apache-2.0"
] | 0 | 3485e6191e0e45bf1c8168e4e928a36ab9264d22 | https://github.com/shubham-gupta-iitr/mmmlX/tree/3485e6191e0e45bf1c8168e4e928a36ab9264d22 | from _paritybench_helpers import _mock_config
import math
import torch
import torch.nn as nn
import torch.utils.data
class BertLayerNorm(nn.Module):
def __init__(self, hidden_size, eps=1e-05):
"""Construct a layernorm module in the TF style (epsilon inside the square root)."""
super().__init__()
self.weight = nn.Parameter(torch.ones(hidden_size))
self.bias = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = eps
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.weight * x + self.bias
class BertSelfAttention(nn.Module):
def __init__(self, config):
super().__init__()
if config.hidden_size % config.num_attention_heads != 0:
raise ValueError(
'The hidden size (%d) is not a multiple of the number of attention heads (%d)'
% (config.hidden_size, config.num_attention_heads))
self.num_attention_heads = config.num_attention_heads
self.attention_head_size = int(config.hidden_size / config.
num_attention_heads)
self.all_head_size = (self.num_attention_heads * self.
attention_head_size)
self.query = nn.Linear(config.hidden_size, self.all_head_size)
self.key = nn.Linear(config.hidden_size, self.all_head_size)
self.value = nn.Linear(config.hidden_size, self.all_head_size)
self.dropout = nn.Dropout(config.attention_probs_dropout_prob)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_attention_heads, self.
attention_head_size)
x = x.view(*new_x_shape)
return x.permute(0, 2, 1, 3)
def forward(self, hidden_states, attention_mask, history_states=None):
if history_states is None:
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(hidden_states)
mixed_value_layer = self.value(hidden_states)
else:
x_states = torch.cat((history_states, hidden_states), dim=1)
mixed_query_layer = self.query(hidden_states)
mixed_key_layer = self.key(x_states)
mixed_value_layer = self.value(x_states)
query_layer = self.transpose_for_scores(mixed_query_layer)
key_layer = self.transpose_for_scores(mixed_key_layer)
value_layer = self.transpose_for_scores(mixed_value_layer)
attention_scores = torch.matmul(query_layer, key_layer.transpose(-1,
-2))
attention_scores = attention_scores / math.sqrt(self.
attention_head_size)
attention_scores = attention_scores + attention_mask
attention_probs = nn.Softmax(dim=-1)(attention_scores)
attention_probs = self.dropout(attention_probs)
context_layer = torch.matmul(attention_probs, value_layer)
context_layer = context_layer.permute(0, 2, 1, 3).contiguous()
new_context_layer_shape = context_layer.size()[:-2] + (self.
all_head_size,)
context_layer = context_layer.view(*new_context_layer_shape)
return context_layer
class BertSelfOutput(nn.Module):
def __init__(self, config):
super().__init__()
self.dense = nn.Linear(config.hidden_size, config.hidden_size)
self.LayerNorm = BertLayerNorm(config.hidden_size, eps=1e-05)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
def forward(self, hidden_states, input_tensor):
hidden_states = self.dense(hidden_states)
hidden_states = self.dropout(hidden_states)
hidden_states = self.LayerNorm(hidden_states + input_tensor)
return hidden_states
class Model(nn.Module):
def __init__(self, config):
super().__init__()
self.self = BertSelfAttention(config)
self.output = BertSelfOutput(config)
def forward(self, input_tensor, atte
# ... truncated (>4000 chars) for memory efficiency |
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_7/inductor_cache/vb/cvbzgl2kevabwrh6e5ay2dasnq6gi3a7zeveknxks7jkbuq64ime.py
# Topologically Sorted Source Nodes: [mul, sigmoid, mul_1], Original ATen: [aten.mul, aten.sigmoid]
# Source node to ATen node mapping:
# mul => mul
# mul_1 => mul_1
# sigmoid => sigmoid
# Graph fragment:
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, 1.702), kwargs = {})
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%mul,), kwargs = {})
# %mul_1 : [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: 'i32'}, 'device': 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_sigmoid_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_mul_sigmoid_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 = 1.702
tmp2 = tmp0 * tmp1
tmp3 = tl.sigmoid(tmp2)
tmp4 = tmp0 * tmp3
tl.store(out_ptr0 + (x0), 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, (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, sigmoid, mul_1], Original ATen: [aten.mul, aten.sigmoid]
stream0 = get_raw_stream(0)
triton_poi_fused_mul_sigmoid_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 torch
from torch import Tensor
from torch import nn
class GELU(nn.Module):
"""Quick GELU"""
def forward(self, x: 'Tensor') ->Tensor:
return x * torch.sigmoid(1.702 * x)
class MLP(nn.Module):
def __init__(self, c1, ch, c2=None):
super().__init__()
self.c_fc = nn.Linear(c1, ch)
self.gelu = GELU()
self.c_proj = nn.Linear(ch, c2 or c1)
def forward(self, x: 'Tensor') ->Tensor:
return self.c_proj(self.gelu(self.c_fc(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'c1': 4, 'ch': 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 import Tensor
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_mul_sigmoid_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 = 1.702
tmp2 = tmp0 * tmp1
tmp3 = tl.sigmoid(tmp2)
tmp4 = tmp0 * tmp3
tl.store(out_ptr0 + x0, 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, (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_mul_sigmoid_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):
"""Quick GELU"""
def forward(self, x: 'Tensor') ->Tensor:
return x * torch.sigmoid(1.702 * x)
class MLPNew(nn.Module):
def __init__(self, c1, ch, c2=None):
super().__init__()
self.c_fc = nn.Linear(c1, ch)
self.gelu = GELU()
self.c_proj = nn.Linear(ch, c2 or c1)
def forward(self, input_0):
primals_1 = self.c_fc.weight
primals_2 = self.c_fc.bias
primals_4 = self.c_proj.weight
primals_5 = self.c_proj.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| sithu31296/multimodal | MLP | false | 4,340 | [
"MIT"
] | 0 | 78f57956cc84273579eb9e2e2be2a58fa1f38814 | https://github.com/sithu31296/multimodal/tree/78f57956cc84273579eb9e2e2be2a58fa1f38814 | import torch
from torch import Tensor
from torch import nn
class GELU(nn.Module):
"""Quick GELU"""
def forward(self, x: 'Tensor') ->Tensor:
return x * torch.sigmoid(1.702 * x)
class Model(nn.Module):
def __init__(self, c1, ch, c2=None):
super().__init__()
self.c_fc = nn.Linear(c1, ch)
self.gelu = GELU()
self.c_proj = nn.Linear(ch, c2 or c1)
def forward(self, x: 'Tensor') ->Tensor:
return self.c_proj(self.gelu(self.c_fc(x)))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
RefModel2d | # 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_7/inductor_cache/eq/ceqcduq3mwis6hwhx3k2hdrhuh5karhoe7hcdbfinghbpx2xm6ph.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 = (%primals_2, [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_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=[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_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 = 288
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')
# kernel path: runs/run_shard_7/inductor_cache/me/cmeiz3aiel7lbg3fmbh7fgzrkcu4bqoriig554gtrpnvtsxhg45h.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten._native_batch_norm_legit]
# Source node to ATen node mapping:
# output_1 => add, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%convolution, [0, 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 = {})
triton_per_fused__native_batch_norm_legit_1 = async_compile.triton('triton_per_fused__native_batch_norm_legit_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=[2, 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': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__native_batch_norm_legit_1', '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_1(in_out_ptr0, in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 2
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 % 4
r2 = (rindex // 4)
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (4*x0) + (8*r2)), 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_7/inductor_cache/sk/cskxoge3n4emzvlb5ys73jxzprcr3gymcp2m4o2xt4glitanjpsn.py
# Topologically Sorted Source Nodes: [output_1, output_2], Original ATen: [aten._native_batch_norm_legit, aten.leaky_relu]
# Source node to ATen node mapping:
# output_1 => add_1, mul, mul_1, sub_4
# output_2 => gt, mul_2, where
# Graph fragment:
# %sub_4 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%convolution, %getitem_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_4, %rsqrt), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul, %unsqueeze_1), kwargs = {})
# %add_1 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %unsqueeze_3), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%add_1, 0), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_1, 0.02), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %add_1, %mul_2), kwargs = {})
triton_poi_fused__native_batch_norm_legit_leaky_relu_2 = async_compile.triton('triton_poi_fused__native_batch_norm_legit_leaky_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=[32],
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_batch_norm_legit_leaky_relu_2', '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__native_batch_norm_legit_leaky_relu_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 4) % 2
tmp0 = tl.load(in_ptr0 + (x3), 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 + (x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tmp9 = 0.0
tmp10 = tmp8 > tmp9
tmp11 = 0.02
tmp12 = tmp8 * tmp11
tmp13 = tl.where(tmp10, tmp8, tmp12)
tl.store(in_out_ptr0 + (x3), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/6q/c6qif56bwzpszgw5ut3oe77mzt3mf34ct3csishbbvmctu4z3lr5.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten._to_copy]
# Source node to ATen node mapping:
# output_5 => convert_element_type_1
# Graph fragment:
# %convert_element_type_1 : [num_users=5] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%view, torch.int64), kwargs = {})
triton_poi_fused__to_copy_3 = async_compile.triton('triton_poi_fused__to_copy_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=[2],
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__to_copy_3', '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__to_copy_3(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/sr/csro45kj2viw32d3iykp6lm5bmdxalfad5ija3eglyjmmcfzntz3.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.add, aten.clamp]
# Source node to ATen node mapping:
# output_5 => add_3, clamp_max
# Graph fragment:
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type_1, 1), kwargs = {})
# %clamp_max : [num_users=3] = call_function[target=torch.ops.aten.clamp_max.default](args = (%add_3, 0), kwargs = {})
triton_poi_fused_add_clamp_4 = async_compile.triton('triton_poi_fused_add_clamp_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=[2],
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_add_clamp_4', '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_add_clamp_4(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 0, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + (x0), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/uf/cuf4x43hb53yhj2wmgdljkybgjpp5igcyodhz6fmzuvkntljhi3e.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
# Source node to ATen node mapping:
# output_5 => add_2, clamp_max_2, clamp_min, clamp_min_2, convert_element_type, iota_2, mul_3, sub_5, sub_7
# Graph fragment:
# %iota_2 : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (2,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%iota_2, torch.float32), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type, 0.5), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_2, 0.5), kwargs = {})
# %sub_5 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_3, 0.5), kwargs = {})
# %clamp_min : [num_users=3] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_5, 0.0), kwargs = {})
# %sub_7 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clamp_min, %convert_element_type_3), kwargs = {})
# %clamp_min_2 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_7, 0.0), kwargs = {})
# %clamp_max_2 : [num_users=3] = call_function[target=torch.ops.aten.clamp_max.default](args = (%clamp_min_2, 1.0), kwargs = {})
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5 = async_compile.triton('triton_poi_fused__to_copy_add_arange_clamp_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=[2],
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,), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5', '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__to_copy_add_arange_clamp_mul_sub_5(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/e5/ce5uk72pxjyulegzkdkszgmjfgevtrf7j7pmd6m5aelcjmxgzzxm.py
# Topologically Sorted Source Nodes: [output_4, output_5], Original ATen: [aten.avg_pool2d, aten._unsafe_index, aten.sub, aten.mul, aten.add]
# Source node to ATen node mapping:
# output_4 => avg_pool2d
# output_5 => _unsafe_index_2, _unsafe_index_3, _unsafe_index_4, _unsafe_index_5, add_6, add_7, mul_5, mul_6, sub_11, sub_8, sub_9
# Graph fragment:
# %avg_pool2d : [num_users=4] = call_function[target=torch.ops.aten.avg_pool2d.default](args = (%where, [2, 2], [4, 4]), kwargs = {})
# %_unsafe_index_2 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %convert_element_type_1, %convert_element_type_3]), kwargs = {})
# %_unsafe_index_3 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %convert_element_type_1, %clamp_max_1]), kwargs = {})
# %_unsafe_index_4 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %clamp_max, %convert_element_type_3]), kwargs = {})
# %_unsafe_index_5 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %clamp_max, %clamp_max_1]), kwargs = {})
# %sub_8 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%_unsafe_index_3, %_unsafe_index_2), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_8, %clamp_max_2), kwargs = {})
# %add_6 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%_unsafe_index_2, %mul_5), kwargs = {})
# %sub_9 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%_unsafe_index_5, %_unsafe_index_4), kwargs = {})
# %mul_6 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_9, %clamp_max_2), kwargs = {})
# %add_7 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%_unsafe_index_4, %mul_6), kwargs = {})
# %sub_11 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_7, %add_6), kwargs = {})
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6 = async_compile.triton('triton_poi_fused__unsafe_index_add_avg_pool2d_mul_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.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i64', 3: '*i64', 4: '*fp32', 5: '*i64', 6: '*i64', 7: '*fp32', 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_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1'], '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__unsafe_index_add_avg_pool2d_mul_sub_6(in_out_ptr0, in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 2) % 2
x0 = xindex % 2
x2 = (xindex // 4)
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr3 + (x1), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 1, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr2 + ((4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + (1 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tl.load(in_ptr2 + (2 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp13 = tmp12 + tmp11
tmp14 = tl.load(in_ptr2 + (3 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = 0.25
tmp17 = tmp15 * tmp16
tmp19 = tmp18 + tmp1
tmp20 = tmp18 < 0
tmp21 = tl.where(tmp20, tmp19, tmp18)
tmp22 = tl.load(in_ptr2 + ((4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr2 + (1 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp24 = tmp23 + tmp22
tmp25 = tl.load(in_ptr2 + (2 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp26 = tmp25 + tmp24
tmp27 = tl.load(in_ptr2 + (3 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp28 = tmp27 + tmp26
tmp29 = tmp28 * tmp16
tmp31 = tmp30 + tmp1
tmp32 = tmp30 < 0
tmp33 = tl.where(tmp32, tmp31, tmp30)
tmp34 = tl.load(in_ptr2 + ((4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr2 + (1 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp36 = tmp35 + tmp34
tmp37 = tl.load(in_ptr2 + (2 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp38 = tmp37 + tmp36
tmp39 = tl.load(in_ptr2 + (3 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp40 = tmp39 + tmp38
tmp41 = tmp40 * tmp16
tmp42 = tmp41 - tmp29
tmp44 = tmp42 * tmp43
tmp45 = tmp29 + tmp44
tmp46 = tl.load(in_ptr2 + ((4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp47 = tl.load(in_ptr2 + (1 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp48 = tmp47 + tmp46
tmp49 = tl.load(in_ptr2 + (2 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp50 = tmp49 + tmp48
tmp51 = tl.load(in_ptr2 + (3 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp52 = tmp51 + tmp50
tmp53 = tmp52 * tmp16
tmp54 = tmp53 - tmp17
tmp55 = tmp54 * tmp43
tmp56 = tmp17 + tmp55
tmp57 = tmp56 - tmp45
tl.store(in_out_ptr0 + (x4), tmp45, xmask)
tl.store(in_out_ptr1 + (x4), tmp57, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/2v/c2vqbwuq7yxh5e3rn6u4lxtfa4ca5a5ijl2omrl6jba6yluz3b2x.py
# Topologically Sorted Source Nodes: [output_5, output_6], Original ATen: [aten.mul, aten.add, aten.mean]
# Source node to ATen node mapping:
# output_5 => add_8, mul_7
# output_6 => mean
# Graph fragment:
# %mul_7 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_11, %clamp_max_3), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_6, %mul_7), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_8, [-1, -2], True), kwargs = {})
triton_poi_fused_add_mean_mul_7 = async_compile.triton('triton_poi_fused_add_mean_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=[8],
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_mean_mul_7', '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_add_mean_mul_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8
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')
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
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')
tmp11 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr2 + (1))
tmp14 = tl.broadcast_to(tmp13, [XBLOCK])
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')
tmp4 = tmp1 * tmp3
tmp5 = tmp0 + tmp4
tmp8 = tmp7 * tmp3
tmp9 = tmp6 + tmp8
tmp10 = tmp5 + tmp9
tmp15 = tmp12 * tmp14
tmp16 = tmp11 + tmp15
tmp17 = tmp10 + tmp16
tmp20 = tmp19 * tmp14
tmp21 = tmp18 + tmp20
tmp22 = tmp17 + tmp21
tmp23 = 4.0
tmp24 = tmp22 / tmp23
tl.store(out_ptr0 + (x0), tmp24, 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, (2, 2, 3, 3), (18, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 4, 4), (32, 16, 4, 1))
assert_size_stride(primals_3, (2, ), (1, ))
assert_size_stride(primals_4, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 6, 6), (72, 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(primals_2, buf0, 288, grid=grid(288), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_1, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 2, 2), (8, 4, 2, 1))
buf2 = empty_strided_cuda((1, 2, 1, 1), (2, 1, 1, 1), torch.float32)
buf3 = empty_strided_cuda((1, 2, 1, 1), (2, 1, 2, 2), torch.float32)
buf5 = reinterpret_tensor(buf3, (1, 2, 1, 1), (2, 1, 1, 1), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten._native_batch_norm_legit]
triton_per_fused__native_batch_norm_legit_1.run(buf5, buf1, buf2, 2, 16, grid=grid(2), stream=stream0)
buf6 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf7 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [output_1, output_2], Original ATen: [aten._native_batch_norm_legit, aten.leaky_relu]
triton_poi_fused__native_batch_norm_legit_leaky_relu_2.run(buf7, buf1, buf2, buf5, primals_3, primals_4, 32, grid=grid(32), stream=stream0)
buf8 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten._to_copy]
triton_poi_fused__to_copy_3.run(buf8, 2, grid=grid(2), stream=stream0)
buf9 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.add, aten.clamp]
triton_poi_fused_add_clamp_4.run(buf9, 2, grid=grid(2), stream=stream0)
buf10 = empty_strided_cuda((2, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_3.run(buf10, 2, grid=grid(2), stream=stream0)
buf11 = empty_strided_cuda((2, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.add, aten.clamp]
triton_poi_fused_add_clamp_4.run(buf11, 2, grid=grid(2), stream=stream0)
buf14 = empty_strided_cuda((2, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5.run(buf14, 2, grid=grid(2), stream=stream0)
buf13 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf12 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf15 = buf12; del buf12 # reuse
buf17 = buf13; del buf13 # reuse
# Topologically Sorted Source Nodes: [output_4, output_5], Original ATen: [aten.avg_pool2d, aten._unsafe_index, aten.sub, aten.mul, aten.add]
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6.run(buf15, buf17, buf9, buf10, buf7, buf8, buf11, buf14, 32, grid=grid(32), stream=stream0)
buf16 = empty_strided_cuda((2, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5.run(buf16, 2, grid=grid(2), stream=stream0)
buf18 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_5, output_6], Original ATen: [aten.mul, aten.add, aten.mean]
triton_poi_fused_add_mean_mul_7.run(buf15, buf17, buf16, buf18, 8, grid=grid(8), stream=stream0)
del buf15
del buf17
return (buf18, primals_1, primals_3, primals_4, buf0, buf1, buf2, buf5, buf7, buf8, buf9, buf10, buf11, buf14, buf16, )
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((2, 2, 3, 3), (18, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 2, 4, 4), (32, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, ), (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
class RefModel2d(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, stride=2, bias=False, padding=1,
padding_mode='reflect')
self.l2 = torch.nn.BatchNorm2d(2, track_running_stats=False)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Dropout2d(0.5)
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='bilinear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 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
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(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 288
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)
@triton.jit
def triton_per_fused__native_batch_norm_legit_1(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 2
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 % 4
r2 = rindex // 4
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 4 * x0 + 8 * r2), 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__native_batch_norm_legit_leaky_relu_2(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, xnumel, XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 2
tmp0 = tl.load(in_ptr0 + x3, 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 + x1, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tmp9 = 0.0
tmp10 = tmp8 > tmp9
tmp11 = 0.02
tmp12 = tmp8 * tmp11
tmp13 = tl.where(tmp10, tmp8, tmp12)
tl.store(in_out_ptr0 + x3, tmp13, xmask)
@triton.jit
def triton_poi_fused__to_copy_3(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + x0, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_clamp_4(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 0, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + x0, tmp12, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + x0, tmp13, xmask)
@triton.jit
def triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6(in_out_ptr0,
in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5,
xnumel, XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 2 % 2
x0 = xindex % 2
x2 = xindex // 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 1, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr2 + (4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + (1 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tl.load(in_ptr2 + (2 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp13 = tmp12 + tmp11
tmp14 = tl.load(in_ptr2 + (3 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = 0.25
tmp17 = tmp15 * tmp16
tmp19 = tmp18 + tmp1
tmp20 = tmp18 < 0
tmp21 = tl.where(tmp20, tmp19, tmp18)
tmp22 = tl.load(in_ptr2 + (4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp23 = tl.load(in_ptr2 + (1 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp24 = tmp23 + tmp22
tmp25 = tl.load(in_ptr2 + (2 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp26 = tmp25 + tmp24
tmp27 = tl.load(in_ptr2 + (3 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp28 = tmp27 + tmp26
tmp29 = tmp28 * tmp16
tmp31 = tmp30 + tmp1
tmp32 = tmp30 < 0
tmp33 = tl.where(tmp32, tmp31, tmp30)
tmp34 = tl.load(in_ptr2 + (4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp35 = tl.load(in_ptr2 + (1 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp36 = tmp35 + tmp34
tmp37 = tl.load(in_ptr2 + (2 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp38 = tmp37 + tmp36
tmp39 = tl.load(in_ptr2 + (3 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp40 = tmp39 + tmp38
tmp41 = tmp40 * tmp16
tmp42 = tmp41 - tmp29
tmp44 = tmp42 * tmp43
tmp45 = tmp29 + tmp44
tmp46 = tl.load(in_ptr2 + (4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp47 = tl.load(in_ptr2 + (1 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp48 = tmp47 + tmp46
tmp49 = tl.load(in_ptr2 + (2 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp50 = tmp49 + tmp48
tmp51 = tl.load(in_ptr2 + (3 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp52 = tmp51 + tmp50
tmp53 = tmp52 * tmp16
tmp54 = tmp53 - tmp17
tmp55 = tmp54 * tmp43
tmp56 = tmp17 + tmp55
tmp57 = tmp56 - tmp45
tl.store(in_out_ptr0 + x4, tmp45, xmask)
tl.store(in_out_ptr1 + x4, tmp57, xmask)
@triton.jit
def triton_poi_fused_add_mean_mul_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 8
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')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
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')
tmp11 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr2 + 1)
tmp14 = tl.broadcast_to(tmp13, [XBLOCK])
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'
)
tmp4 = tmp1 * tmp3
tmp5 = tmp0 + tmp4
tmp8 = tmp7 * tmp3
tmp9 = tmp6 + tmp8
tmp10 = tmp5 + tmp9
tmp15 = tmp12 * tmp14
tmp16 = tmp11 + tmp15
tmp17 = tmp10 + tmp16
tmp20 = tmp19 * tmp14
tmp21 = tmp18 + tmp20
tmp22 = tmp17 + tmp21
tmp23 = 4.0
tmp24 = tmp22 / tmp23
tl.store(out_ptr0 + x0, tmp24, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 3, 3), (18, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 4, 4), (32, 16, 4, 1))
assert_size_stride(primals_3, (2,), (1,))
assert_size_stride(primals_4, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 6, 6), (72, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0[grid(288)](primals_2, buf0, 288,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf1 = extern_kernels.convolution(buf0, primals_1, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 2, 2), (8, 4, 2, 1))
buf2 = empty_strided_cuda((1, 2, 1, 1), (2, 1, 1, 1), torch.float32)
buf3 = empty_strided_cuda((1, 2, 1, 1), (2, 1, 2, 2), torch.float32)
buf5 = reinterpret_tensor(buf3, (1, 2, 1, 1), (2, 1, 1, 1), 0)
del buf3
triton_per_fused__native_batch_norm_legit_1[grid(2)](buf5, buf1,
buf2, 2, 16, XBLOCK=1, num_warps=2, num_stages=1)
buf6 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf7 = buf6
del buf6
triton_poi_fused__native_batch_norm_legit_leaky_relu_2[grid(32)](buf7,
buf1, buf2, buf5, primals_3, primals_4, 32, XBLOCK=32,
num_warps=1, num_stages=1)
buf8 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_3[grid(2)](buf8, 2, XBLOCK=2, num_warps=1,
num_stages=1)
buf9 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_4[grid(2)](buf9, 2, XBLOCK=2, num_warps=
1, num_stages=1)
buf10 = empty_strided_cuda((2,), (1,), torch.int64)
triton_poi_fused__to_copy_3[grid(2)](buf10, 2, XBLOCK=2, num_warps=
1, num_stages=1)
buf11 = empty_strided_cuda((2,), (1,), torch.int64)
triton_poi_fused_add_clamp_4[grid(2)](buf11, 2, XBLOCK=2, num_warps
=1, num_stages=1)
buf14 = empty_strided_cuda((2,), (1,), torch.float32)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5[grid(2)](buf14,
2, XBLOCK=2, num_warps=1, num_stages=1)
buf13 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf12 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf15 = buf12
del buf12
buf17 = buf13
del buf13
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6[grid(32)](buf15
, buf17, buf9, buf10, buf7, buf8, buf11, buf14, 32, XBLOCK=32,
num_warps=1, num_stages=1)
buf16 = empty_strided_cuda((2, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5[grid(2)](buf16,
2, XBLOCK=2, num_warps=1, num_stages=1)
buf18 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
triton_poi_fused_add_mean_mul_7[grid(8)](buf15, buf17, buf16, buf18,
8, XBLOCK=8, num_warps=1, num_stages=1)
del buf15
del buf17
return (buf18, primals_1, primals_3, primals_4, buf0, buf1, buf2, buf5,
buf7, buf8, buf9, buf10, buf11, buf14, buf16)
class RefModel2dNew(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, stride=2, bias=False, padding=1,
padding_mode='reflect')
self.l2 = torch.nn.BatchNorm2d(2, track_running_stats=False)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Dropout2d(0.5)
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, input_0):
primals_1 = self.l1.weight
primals_3 = self.l2.weight
primals_4 = self.l2.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| shuohan/pytorch-layers | RefModel2d | false | 4,341 | [
"MIT"
] | 0 | 020846fd02d501cf477552179c19ba4b5e9a0695 | https://github.com/shuohan/pytorch-layers/tree/020846fd02d501cf477552179c19ba4b5e9a0695 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, stride=2, bias=False, padding=1,
padding_mode='reflect')
self.l2 = torch.nn.BatchNorm2d(2, track_running_stats=False)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Dropout2d(0.5)
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='bilinear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 4, 4])]
def get_init_inputs():
return []
|
TripletLoss | # 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_7/inductor_cache/nq/cnqxadd2nqiglmqaw4vqaddkj5bet76gl7wd4ucmiz7rls7hcdpu.py
# Topologically Sorted Source Nodes: [add, mul, dist], Original ATen: [aten.add, aten.mul, aten.sub]
# Source node to ATen node mapping:
# add => add
# dist => sub
# mul => mul
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%expand, %expand_1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mm, 2), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %mul), kwargs = {})
triton_poi_fused_add_mul_sub_0 = async_compile.triton('triton_poi_fused_add_mul_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=[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_add_mul_sub_0', 'mutated_arg_names': ['in_out_ptr0'], '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_sub_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
x1 = (xindex // 4)
x0 = 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')
tmp11 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp16 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp23 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp12 = tmp11 * tmp11
tmp14 = tmp13 * tmp13
tmp15 = tmp12 + tmp14
tmp17 = tmp16 * tmp16
tmp18 = tmp15 + tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp10 + tmp21
tmp24 = 2.0
tmp25 = tmp23 * tmp24
tmp26 = tmp22 - tmp25
tl.store(in_out_ptr0 + (x2), tmp26, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ka/ckaxq6y5mhx62yabsjzhpn6yjyeenx7523rr6eunxttnkjz43mmm.py
# Topologically Sorted Source Nodes: [is_pos, mul_1, max_1, is_neg, mul_2, mul_3, add_1, min_1, loss], Original ATen: [aten.eq, aten.mul, aten.max, aten.bitwise_not, aten.add, aten.min, aten.neg, aten.sub, aten.clamp_min, aten.mean]
# Source node to ATen node mapping:
# add_1 => add_1
# is_neg => bitwise_not
# is_pos => eq
# loss => add_2, clamp_min, full_default, mean, mul_4, sub_1
# max_1 => max_1
# min_1 => min_1
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# Graph fragment:
# %eq : [num_users=3] = call_function[target=torch.ops.aten.eq.Tensor](args = (%unsqueeze, %unsqueeze_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %eq), kwargs = {})
# %max_1 : [num_users=1] = call_function[target=torch.ops.aten.max.dim](args = (%mul_1, 1), kwargs = {})
# %bitwise_not : [num_users=1] = call_function[target=torch.ops.aten.bitwise_not.default](args = (%eq,), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %bitwise_not), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%eq, 1000000000.0), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %mul_3), kwargs = {})
# %min_1 : [num_users=1] = call_function[target=torch.ops.aten.min.dim](args = (%add_1, 1), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4], -1.0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%getitem_2, %getitem), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%full_default, %sub_1), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Scalar](args = (%mul_4, 0.3), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%add_2, 0), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%clamp_min,), kwargs = {})
triton_per_fused_add_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1 = async_compile.triton('triton_per_fused_add_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_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, 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': {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_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 9, '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_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1(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 % 4
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (r0), None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr1 + (r2), None)
tmp6 = tl.load(in_ptr0 + (4 + r0), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (4 + r0), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (8 + r0), None, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (8 + r0), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (12 + r0), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (12 + r0), None, eviction_policy='evict_last')
tmp3 = tmp1 == tmp2
tmp4 = tmp3.to(tl.float32)
tmp5 = tmp0 * tmp4
tmp8 = tmp7 == tmp2
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp6 * tmp9
tmp11 = triton_helpers.maximum(tmp5, tmp10)
tmp14 = tmp13 == tmp2
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp12 * tmp15
tmp17 = triton_helpers.maximum(tmp11, tmp16)
tmp20 = tmp19 == tmp2
tmp21 = tmp20.to(tl.float32)
tmp22 = tmp18 * tmp21
tmp23 = triton_helpers.maximum(tmp17, tmp22)
tmp24 = tmp3 == 0
tmp25 = tmp24.to(tl.float32)
tmp26 = tmp0 * tmp25
tmp27 = 1000000000.0
tmp28 = tmp4 * tmp27
tmp29 = tmp26 + tmp28
tmp30 = tmp8 == 0
tmp31 = tmp30.to(tl.float32)
tmp32 = tmp6 * tmp31
tmp33 = tmp9 * tmp27
tmp34 = tmp32 + tmp33
tmp35 = triton_helpers.minimum(tmp29, tmp34)
tmp36 = tmp14 == 0
tmp37 = tmp36.to(tl.float32)
tmp38 = tmp12 * tmp37
tmp39 = tmp15 * tmp27
tmp40 = tmp38 + tmp39
tmp41 = triton_helpers.minimum(tmp35, tmp40)
tmp42 = tmp20 == 0
tmp43 = tmp42.to(tl.float32)
tmp44 = tmp18 * tmp43
tmp45 = tmp21 * tmp27
tmp46 = tmp44 + tmp45
tmp47 = triton_helpers.minimum(tmp41, tmp46)
tmp48 = tmp47 - tmp23
tmp49 = -1.0
tmp50 = tmp49 * tmp48
tmp51 = 0.3
tmp52 = tmp50 + tmp51
tmp53 = 0.0
tmp54 = triton_helpers.maximum(tmp52, tmp53)
tmp55 = tl.broadcast_to(tmp54, [XBLOCK, RBLOCK])
tmp57 = tl.sum(tmp55, 1)[:, None]
tmp58 = 16.0
tmp59 = tmp57 / tmp58
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([XBLOCK, 1], 0, tl.int32)), tmp59, 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, 1))
assert_size_stride(arg1_1, (4, 4), (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: [matmul], Original ATen: [aten.mm]
extern_kernels.mm(arg0_1, reinterpret_tensor(arg0_1, (4, 4), (1, 4), 0), out=buf0)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [add, mul, dist], Original ATen: [aten.add, aten.mul, aten.sub]
stream0 = get_raw_stream(0)
triton_poi_fused_add_mul_sub_0.run(buf1, arg0_1, 16, grid=grid(16), stream=stream0)
del arg0_1
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [is_pos, mul_1, max_1, is_neg, mul_2, mul_3, add_1, min_1, loss], Original ATen: [aten.eq, aten.mul, aten.max, aten.bitwise_not, aten.add, aten.min, aten.neg, aten.sub, aten.clamp_min, aten.mean]
triton_per_fused_add_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1.run(buf5, buf1, arg1_1, 1, 16, 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, 1), device='cuda:0', dtype=torch.float32)
arg1_1 = rand_strided((4, 4), (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 Tensor
from torch import nn
from torch.nn import functional as F
def euclidean_dist(x: 'Tensor', y: 'Tensor') ->Tensor:
xx, yy = torch.meshgrid((x ** 2).sum(1), (y ** 2).sum(1))
return xx + yy - 2 * (x @ y.t())
class TripletLoss(nn.Module):
"""
Modified from Tong Xiao's open-reid (https://github.com/Cysu/open-reid).
Related Triplet Loss theory can be found in paper 'In Defense of the Triplet
Loss for Person Re-Identification'
"""
def __init__(self, margin: 'float'=0.3) ->None:
super().__init__()
self.margin = margin
def forward(self, features: 'Tensor', targets: 'Tensor') ->Tensor:
dist = euclidean_dist(features, features)
is_pos = targets.unsqueeze(0) == targets.unsqueeze(1)
is_neg = ~is_pos
dist_ap, dist_an = self.hard_example_mining(dist, is_pos, is_neg)
y = torch.ones_like(dist_an)
if self.margin > 0:
loss = F.margin_ranking_loss(dist_an, dist_ap, y, self.margin)
else:
loss = F.soft_margin_loss(dist_an - dist_ap, y)
if loss == float('inf'):
loss = F.margin_ranking_loss(dist_an, dist_ap, y, 0.3)
return loss
def hard_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the hardest positive and negative sample.
Args:
dist_mat: pair wise distance between samples, shape [N, M]
is_pos: positive index with shape [N, M]
is_neg: negative index with shape [N, M]
Returns:
dist_ap: Tensor, distance(anchor, positive); shape [N]
dist_an: Tensor, distance(anchor, negative); shape [N]
NOTE: Only consider the case in which all labels have same num of samples,
thus we can cope with all anchors in parallel.
"""
assert len(dist.size()) == 2
dist_ap = torch.max(dist * is_pos, dim=1)[0]
dist_an = torch.min(dist * is_neg + is_pos * 1000000000.0, dim=1)[0]
return dist_ap, dist_an
def weighted_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the weighted positive and negative sample.
"""
assert len(dist.size()) == 2
dist_ap = dist * is_pos
dist_an = dist * is_neg
weights_ap = self.softmax_weights(dist_ap, is_pos)
weights_an = self.softmax_weights(-dist_an, is_neg)
dist_ap = torch.sum(dist_ap * weights_ap, dim=1)
dist_an = torch.sum(dist_an * weights_an, dim=1)
return dist_ap, dist_an
def softmax_weights(self, dist, mask):
max_v = torch.max(dist * mask, dim=1, keepdim=True)[0]
diff = dist - max_v
Z = torch.sum(torch.exp(diff) * mask, dim=1, keepdim=True) + 1e-06
W = torch.exp(diff) * mask / Z
return W
def get_inputs():
return [torch.rand([4, 4]), torch.rand([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 import Tensor
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_add_mul_sub_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
x1 = xindex // 4
x0 = 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')
tmp11 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp16 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp23 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp12 = tmp11 * tmp11
tmp14 = tmp13 * tmp13
tmp15 = tmp12 + tmp14
tmp17 = tmp16 * tmp16
tmp18 = tmp15 + tmp17
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp10 + tmp21
tmp24 = 2.0
tmp25 = tmp23 * tmp24
tmp26 = tmp22 - tmp25
tl.store(in_out_ptr0 + x2, tmp26, xmask)
@triton.jit
def triton_per_fused_add_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1(
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 % 4
r2 = rindex
tmp0 = tl.load(in_ptr0 + r0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + r0, None, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr1 + r2, None)
tmp6 = tl.load(in_ptr0 + (4 + r0), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (4 + r0), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (8 + r0), None, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr1 + (8 + r0), None, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr0 + (12 + r0), None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr1 + (12 + r0), None, eviction_policy='evict_last')
tmp3 = tmp1 == tmp2
tmp4 = tmp3.to(tl.float32)
tmp5 = tmp0 * tmp4
tmp8 = tmp7 == tmp2
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp6 * tmp9
tmp11 = triton_helpers.maximum(tmp5, tmp10)
tmp14 = tmp13 == tmp2
tmp15 = tmp14.to(tl.float32)
tmp16 = tmp12 * tmp15
tmp17 = triton_helpers.maximum(tmp11, tmp16)
tmp20 = tmp19 == tmp2
tmp21 = tmp20.to(tl.float32)
tmp22 = tmp18 * tmp21
tmp23 = triton_helpers.maximum(tmp17, tmp22)
tmp24 = tmp3 == 0
tmp25 = tmp24.to(tl.float32)
tmp26 = tmp0 * tmp25
tmp27 = 1000000000.0
tmp28 = tmp4 * tmp27
tmp29 = tmp26 + tmp28
tmp30 = tmp8 == 0
tmp31 = tmp30.to(tl.float32)
tmp32 = tmp6 * tmp31
tmp33 = tmp9 * tmp27
tmp34 = tmp32 + tmp33
tmp35 = triton_helpers.minimum(tmp29, tmp34)
tmp36 = tmp14 == 0
tmp37 = tmp36.to(tl.float32)
tmp38 = tmp12 * tmp37
tmp39 = tmp15 * tmp27
tmp40 = tmp38 + tmp39
tmp41 = triton_helpers.minimum(tmp35, tmp40)
tmp42 = tmp20 == 0
tmp43 = tmp42.to(tl.float32)
tmp44 = tmp18 * tmp43
tmp45 = tmp21 * tmp27
tmp46 = tmp44 + tmp45
tmp47 = triton_helpers.minimum(tmp41, tmp46)
tmp48 = tmp47 - tmp23
tmp49 = -1.0
tmp50 = tmp49 * tmp48
tmp51 = 0.3
tmp52 = tmp50 + tmp51
tmp53 = 0.0
tmp54 = triton_helpers.maximum(tmp52, tmp53)
tmp55 = tl.broadcast_to(tmp54, [XBLOCK, RBLOCK])
tmp57 = tl.sum(tmp55, 1)[:, None]
tmp58 = 16.0
tmp59 = tmp57 / tmp58
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp59, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (4, 1))
assert_size_stride(arg1_1, (4, 4), (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(arg0_1, reinterpret_tensor(arg0_1, (4, 4), (1, 4),
0), out=buf0)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_add_mul_sub_0[grid(16)](buf1, arg0_1, 16, XBLOCK=
16, num_warps=1, num_stages=1)
del arg0_1
buf4 = empty_strided_cuda((), (), torch.float32)
buf5 = buf4
del buf4
triton_per_fused_add_bitwise_not_clamp_min_eq_max_mean_min_mul_neg_sub_1[
grid(1)](buf5, buf1, arg1_1, 1, 16, XBLOCK=1, num_warps=2,
num_stages=1)
del arg1_1
del buf1
return buf5,
def euclidean_dist(x: 'Tensor', y: 'Tensor') ->Tensor:
xx, yy = torch.meshgrid((x ** 2).sum(1), (y ** 2).sum(1))
return xx + yy - 2 * (x @ y.t())
class TripletLossNew(nn.Module):
"""
Modified from Tong Xiao's open-reid (https://github.com/Cysu/open-reid).
Related Triplet Loss theory can be found in paper 'In Defense of the Triplet
Loss for Person Re-Identification'
"""
def __init__(self, margin: 'float'=0.3) ->None:
super().__init__()
self.margin = margin
def hard_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the hardest positive and negative sample.
Args:
dist_mat: pair wise distance between samples, shape [N, M]
is_pos: positive index with shape [N, M]
is_neg: negative index with shape [N, M]
Returns:
dist_ap: Tensor, distance(anchor, positive); shape [N]
dist_an: Tensor, distance(anchor, negative); shape [N]
NOTE: Only consider the case in which all labels have same num of samples,
thus we can cope with all anchors in parallel.
"""
assert len(dist.size()) == 2
dist_ap = torch.max(dist * is_pos, dim=1)[0]
dist_an = torch.min(dist * is_neg + is_pos * 1000000000.0, dim=1)[0]
return dist_ap, dist_an
def weighted_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the weighted positive and negative sample.
"""
assert len(dist.size()) == 2
dist_ap = dist * is_pos
dist_an = dist * is_neg
weights_ap = self.softmax_weights(dist_ap, is_pos)
weights_an = self.softmax_weights(-dist_an, is_neg)
dist_ap = torch.sum(dist_ap * weights_ap, dim=1)
dist_an = torch.sum(dist_an * weights_an, dim=1)
return dist_ap, dist_an
def softmax_weights(self, dist, mask):
max_v = torch.max(dist * mask, dim=1, keepdim=True)[0]
diff = dist - max_v
Z = torch.sum(torch.exp(diff) * mask, dim=1, keepdim=True) + 1e-06
W = torch.exp(diff) * mask / Z
return W
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| sithu31296/re_identification | TripletLoss | false | 4,342 | [
"MIT"
] | 0 | 28c2cf32c6c8c9d79330e1419a7156fe10d8ac95 | https://github.com/sithu31296/re_identification/tree/28c2cf32c6c8c9d79330e1419a7156fe10d8ac95 | import torch
from torch import Tensor
from torch import nn
from torch.nn import functional as F
def euclidean_dist(x: 'Tensor', y: 'Tensor') ->Tensor:
xx, yy = torch.meshgrid((x ** 2).sum(1), (y ** 2).sum(1))
return xx + yy - 2 * (x @ y.t())
class Model(nn.Module):
"""
Modified from Tong Xiao's open-reid (https://github.com/Cysu/open-reid).
Related Triplet Loss theory can be found in paper 'In Defense of the Triplet
Loss for Person Re-Identification'
"""
def __init__(self, margin: 'float'=0.3) ->None:
super().__init__()
self.margin = margin
def forward(self, features: 'Tensor', targets: 'Tensor') ->Tensor:
dist = euclidean_dist(features, features)
is_pos = targets.unsqueeze(0) == targets.unsqueeze(1)
is_neg = ~is_pos
dist_ap, dist_an = self.hard_example_mining(dist, is_pos, is_neg)
y = torch.ones_like(dist_an)
if self.margin > 0:
loss = F.margin_ranking_loss(dist_an, dist_ap, y, self.margin)
else:
loss = F.soft_margin_loss(dist_an - dist_ap, y)
if loss == float('inf'):
loss = F.margin_ranking_loss(dist_an, dist_ap, y, 0.3)
return loss
def hard_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the hardest positive and negative sample.
Args:
dist_mat: pair wise distance between samples, shape [N, M]
is_pos: positive index with shape [N, M]
is_neg: negative index with shape [N, M]
Returns:
dist_ap: Tensor, distance(anchor, positive); shape [N]
dist_an: Tensor, distance(anchor, negative); shape [N]
NOTE: Only consider the case in which all labels have same num of samples,
thus we can cope with all anchors in parallel.
"""
assert len(dist.size()) == 2
dist_ap = torch.max(dist * is_pos, dim=1)[0]
dist_an = torch.min(dist * is_neg + is_pos * 1000000000.0, dim=1)[0]
return dist_ap, dist_an
def weighted_example_mining(self, dist, is_pos, is_neg):
"""For each anchor, find the weighted positive and negative sample.
"""
assert len(dist.size()) == 2
dist_ap = dist * is_pos
dist_an = dist * is_neg
weights_ap = self.softmax_weights(dist_ap, is_pos)
weights_an = self.softmax_weights(-dist_an, is_neg)
dist_ap = torch.sum(dist_ap * weights_ap, dim=1)
dist_an = torch.sum(dist_an * weights_an, dim=1)
return dist_ap, dist_an
def softmax_weights(self, dist, mask):
max_v = torch.max(dist * mask, dim=1, keepdim=True)[0]
diff = dist - max_v
Z = torch.sum(torch.exp(diff) * mask, dim=1, keepdim=True) + 1e-06
W = torch.exp(diff) * mask / Z
return W
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return []
|
RefModel2d2 | # 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_7/inductor_cache/bx/cbxrkpfkiamdpst4hgky5kjgudxp5zbt7qfrxz2x7dxwithhie2t.py
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.copy]
# Source node to ATen node mapping:
# pad => copy
# Graph fragment:
# %copy : [num_users=1] = call_function[target=torch.ops.aten.copy.default](args = (%slice_3, %slice_4), kwargs = {})
# %slice_scatter_default : [num_users=1] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%slice_tensor, %copy, 2, 1, 5), kwargs = {})
# %slice_scatter_default_1 : [num_users=2] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%empty, %slice_scatter_default, 3, 1, 5), kwargs = {})
# %slice_scatter_default_2 : [num_users=2] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%slice_scatter_default_1, %slice_11, 3, 0, 1), kwargs = {})
# %slice_scatter_default_3 : [num_users=2] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%slice_scatter_default_2, %slice_16, 3, 5, 6), kwargs = {})
triton_poi_fused_copy_0 = async_compile.triton('triton_poi_fused_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=[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_copy_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_copy_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 288
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)
x4 = xindex
tmp0 = x0
tmp1 = tl.full([1], 5, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = (-4) + x0
tmp4 = tl.full([1], 1, tl.int64)
tmp5 = tmp3 < tmp4
tmp6 = tmp5 & tmp2
tmp7 = tmp0 >= tmp4
tmp8 = tmp0 < tmp1
tmp9 = tmp7 & tmp8
tmp10 = tmp9 & tmp6
tmp11 = x1
tmp12 = tmp11 >= tmp4
tmp13 = tmp11 < tmp1
tmp14 = tmp12 & tmp13
tmp15 = tmp14 & tmp10
tmp16 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp15 & xmask, other=0.0)
tmp17 = tl.load(in_ptr1 + (x4), tmp10 & xmask, other=0.0)
tmp18 = tl.where(tmp14, tmp16, tmp17)
tmp19 = tl.full(tmp18.shape, 0.0, tmp18.dtype)
tmp20 = tl.where(tmp10, tmp18, tmp19)
tmp21 = float("nan")
tmp22 = tl.where(tmp9, tmp20, tmp21)
tmp23 = tl.full(tmp22.shape, 0.0, tmp22.dtype)
tmp24 = tl.where(tmp6, tmp22, tmp23)
tmp25 = tmp3 >= tmp4
tmp26 = tmp3 < tmp1
tmp27 = tmp25 & tmp26
tmp28 = tmp27 & tmp2
tmp29 = tmp14 & tmp28
tmp30 = tl.load(in_ptr0 + ((-9) + x0 + (4*x1) + (16*x2)), tmp29 & xmask, other=0.0)
tmp31 = tl.load(in_ptr1 + ((-4) + x4), tmp28 & xmask, other=0.0)
tmp32 = tl.where(tmp14, tmp30, tmp31)
tmp33 = tl.full(tmp32.shape, 0.0, tmp32.dtype)
tmp34 = tl.where(tmp28, tmp32, tmp33)
tmp35 = tl.where(tmp27, tmp34, tmp21)
tmp36 = tl.where(tmp5, tmp24, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp2, tmp36, tmp37)
tmp39 = tmp0 < tmp4
tmp40 = 4 + x0
tmp41 = tmp40 >= tmp4
tmp42 = tmp40 < tmp1
tmp43 = tmp41 & tmp42
tmp44 = tmp43 & tmp39
tmp45 = tmp14 & tmp44
tmp46 = tl.load(in_ptr0 + ((-1) + x0 + (4*x1) + (16*x2)), tmp45 & xmask, other=0.0)
tmp47 = tl.load(in_ptr1 + (4 + x4), tmp44 & xmask, other=0.0)
tmp48 = tl.where(tmp14, tmp46, tmp47)
tmp49 = tl.full(tmp48.shape, 0.0, tmp48.dtype)
tmp50 = tl.where(tmp44, tmp48, tmp49)
tmp51 = tl.where(tmp43, tmp50, tmp21)
tmp52 = tl.full(tmp51.shape, 0.0, tmp51.dtype)
tmp53 = tl.where(tmp39, tmp51, tmp52)
tmp54 = tmp14 & tmp9
tmp55 = tl.load(in_ptr0 + ((-5) + x0 + (4*x1) + (16*x2)), tmp54 & xmask, other=0.0)
tmp56 = tl.load(in_ptr1 + (x4), tmp9 & xmask, other=0.0)
tmp57 = tl.where(tmp14, tmp55, tmp56)
tmp58 = tl.full(tmp57.shape, 0.0, tmp57.dtype)
tmp59 = tl.where(tmp9, tmp57, tmp58)
tmp60 = tl.where(tmp9, tmp59, tmp21)
tmp61 = tl.where(tmp39, tmp53, tmp60)
tmp62 = tl.where(tmp2, tmp38, tmp61)
tl.store(out_ptr0 + (x4), tmp62, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/yk/cykeyceeourcl5atvsccijudgei6xeuokqdobmnl4444bhysybfj.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %slice_scatter_default_4 : [num_users=2] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%slice_scatter_default_3, %slice_21, 2, 0, 1), kwargs = {})
# %slice_scatter_default_5 : [num_users=2] = call_function[target=torch.ops.aten.slice_scatter.default](args = (%slice_scatter_default_4, %slice_26, 2, 5, 6), 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=[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_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_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 288
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)
x3 = xindex
tmp14 = tl.load(in_ptr0 + (x3), xmask)
tmp0 = x1
tmp1 = tl.full([1], 5, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = (-4) + x1
tmp4 = tl.full([1], 1, tl.int64)
tmp5 = tmp3 < tmp4
tmp6 = tmp5 & tmp2
tmp7 = tl.load(in_ptr0 + (24 + x0 + (36*x2)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp8 = tl.load(in_ptr0 + ((-24) + x3), tmp2 & xmask, other=0.0)
tmp9 = tl.where(tmp5, tmp7, tmp8)
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp2, tmp9, tmp10)
tmp12 = tmp0 < tmp4
tmp13 = tl.load(in_ptr0 + (24 + x0 + (36*x2)), tmp12 & xmask, eviction_policy='evict_last', other=0.0)
tmp15 = tl.where(tmp12, tmp13, tmp14)
tmp16 = tl.where(tmp2, tmp11, tmp15)
tl.store(out_ptr0 + (x3), tmp16, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rc/crcmvfnctwaepxvquentb6xnoqlcmm2mrbcdvvz4ltgk3nnrc77w.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.leaky_relu]
# Source node to ATen node mapping:
# output_1 => gt, mul, where
# Graph fragment:
# %gt : [num_users=2] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution, 0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution, 0.02), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %convolution, %mul), kwargs = {})
triton_poi_fused_leaky_relu_2 = async_compile.triton('triton_poi_fused_leaky_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=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 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_leaky_relu_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_leaky_relu_2(in_ptr0, out_ptr0, out_ptr1, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
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 = 0.02
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + (x0), tmp2, xmask)
tl.store(out_ptr1 + (x0), tmp5, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/6q/c6qif56bwzpszgw5ut3oe77mzt3mf34ct3csishbbvmctu4z3lr5.py
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten._to_copy]
# Source node to ATen node mapping:
# output_3 => convert_element_type_1
# Graph fragment:
# %convert_element_type_1 : [num_users=5] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%view, torch.int64), kwargs = {})
triton_poi_fused__to_copy_3 = async_compile.triton('triton_poi_fused__to_copy_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=[2],
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__to_copy_3', '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__to_copy_3(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + (x0), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/sr/csro45kj2viw32d3iykp6lm5bmdxalfad5ija3eglyjmmcfzntz3.py
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.add, aten.clamp]
# Source node to ATen node mapping:
# output_3 => add_1, clamp_max
# Graph fragment:
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type_1, 1), kwargs = {})
# %clamp_max : [num_users=3] = call_function[target=torch.ops.aten.clamp_max.default](args = (%add_1, 0), kwargs = {})
triton_poi_fused_add_clamp_4 = async_compile.triton('triton_poi_fused_add_clamp_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=[2],
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_add_clamp_4', '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_add_clamp_4(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 0, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + (x0), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/uf/cuf4x43hb53yhj2wmgdljkybgjpp5igcyodhz6fmzuvkntljhi3e.py
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
# Source node to ATen node mapping:
# output_3 => add, clamp_max_2, clamp_min, clamp_min_2, convert_element_type, iota, mul_1, sub, sub_2
# Graph fragment:
# %iota : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (2,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%iota, torch.float32), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type, 0.5), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add, 0.5), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul_1, 0.5), kwargs = {})
# %clamp_min : [num_users=3] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub, 0.0), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clamp_min, %convert_element_type_3), kwargs = {})
# %clamp_min_2 : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%sub_2, 0.0), kwargs = {})
# %clamp_max_2 : [num_users=3] = call_function[target=torch.ops.aten.clamp_max.default](args = (%clamp_min_2, 1.0), kwargs = {})
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5 = async_compile.triton('triton_poi_fused__to_copy_add_arange_clamp_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=[2],
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,), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5', '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__to_copy_add_arange_clamp_mul_sub_5(out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/e5/ce5uk72pxjyulegzkdkszgmjfgevtrf7j7pmd6m5aelcjmxgzzxm.py
# Topologically Sorted Source Nodes: [output_2, output_3], Original ATen: [aten.avg_pool2d, aten._unsafe_index, aten.sub, aten.mul, aten.add]
# Source node to ATen node mapping:
# output_2 => avg_pool2d
# output_3 => _unsafe_index, _unsafe_index_1, _unsafe_index_2, _unsafe_index_3, add_4, add_5, mul_3, mul_4, sub_3, sub_4, sub_6
# Graph fragment:
# %avg_pool2d : [num_users=4] = call_function[target=torch.ops.aten.avg_pool2d.default](args = (%where, [2, 2], [4, 4]), kwargs = {})
# %_unsafe_index : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %convert_element_type_1, %convert_element_type_3]), kwargs = {})
# %_unsafe_index_1 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %convert_element_type_1, %clamp_max_1]), kwargs = {})
# %_unsafe_index_2 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %clamp_max, %convert_element_type_3]), kwargs = {})
# %_unsafe_index_3 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool2d, [None, None, %clamp_max, %clamp_max_1]), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%_unsafe_index_1, %_unsafe_index), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_3, %clamp_max_2), kwargs = {})
# %add_4 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%_unsafe_index, %mul_3), kwargs = {})
# %sub_4 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%_unsafe_index_3, %_unsafe_index_2), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_4, %clamp_max_2), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%_unsafe_index_2, %mul_4), kwargs = {})
# %sub_6 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_5, %add_4), kwargs = {})
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6 = async_compile.triton('triton_poi_fused__unsafe_index_add_avg_pool2d_mul_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.pointwise(
size_hints=[32],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*i64', 3: '*i64', 4: '*fp32', 5: '*i64', 6: '*i64', 7: '*fp32', 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_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6', 'mutated_arg_names': ['in_out_ptr0', 'in_out_ptr1'], '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__unsafe_index_add_avg_pool2d_mul_sub_6(in_out_ptr0, in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, xnumel, XBLOCK : tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 2) % 2
x0 = xindex % 2
x2 = (xindex // 4)
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (x0), xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr3 + (x1), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr4 + (x0), xmask, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr5 + (x0), xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 1, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr2 + ((4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + (1 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tl.load(in_ptr2 + (2 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp13 = tmp12 + tmp11
tmp14 = tl.load(in_ptr2 + (3 + (4*tmp8) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = 0.25
tmp17 = tmp15 * tmp16
tmp19 = tmp18 + tmp1
tmp20 = tmp18 < 0
tmp21 = tl.where(tmp20, tmp19, tmp18)
tmp22 = tl.load(in_ptr2 + ((4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr2 + (1 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp24 = tmp23 + tmp22
tmp25 = tl.load(in_ptr2 + (2 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp26 = tmp25 + tmp24
tmp27 = tl.load(in_ptr2 + (3 + (4*tmp8) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp28 = tmp27 + tmp26
tmp29 = tmp28 * tmp16
tmp31 = tmp30 + tmp1
tmp32 = tmp30 < 0
tmp33 = tl.where(tmp32, tmp31, tmp30)
tmp34 = tl.load(in_ptr2 + ((4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr2 + (1 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp36 = tmp35 + tmp34
tmp37 = tl.load(in_ptr2 + (2 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp38 = tmp37 + tmp36
tmp39 = tl.load(in_ptr2 + (3 + (4*tmp33) + (4*x2) + (8*tmp21)), xmask, eviction_policy='evict_last')
tmp40 = tmp39 + tmp38
tmp41 = tmp40 * tmp16
tmp42 = tmp41 - tmp29
tmp44 = tmp42 * tmp43
tmp45 = tmp29 + tmp44
tmp46 = tl.load(in_ptr2 + ((4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp47 = tl.load(in_ptr2 + (1 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp48 = tmp47 + tmp46
tmp49 = tl.load(in_ptr2 + (2 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp50 = tmp49 + tmp48
tmp51 = tl.load(in_ptr2 + (3 + (4*tmp33) + (4*x2) + (8*tmp4)), xmask, eviction_policy='evict_last')
tmp52 = tmp51 + tmp50
tmp53 = tmp52 * tmp16
tmp54 = tmp53 - tmp17
tmp55 = tmp54 * tmp43
tmp56 = tmp17 + tmp55
tmp57 = tmp56 - tmp45
tl.store(in_out_ptr0 + (x4), tmp45, xmask)
tl.store(in_out_ptr1 + (x4), tmp57, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/2v/c2vqbwuq7yxh5e3rn6u4lxtfa4ca5a5ijl2omrl6jba6yluz3b2x.py
# Topologically Sorted Source Nodes: [output_3, output_4], Original ATen: [aten.mul, aten.add, aten.mean]
# Source node to ATen node mapping:
# output_3 => add_6, mul_5
# output_4 => mean
# Graph fragment:
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_6, %clamp_max_3), kwargs = {})
# %add_6 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_4, %mul_5), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_6, [-1, -2], True), kwargs = {})
triton_poi_fused_add_mean_mul_7 = async_compile.triton('triton_poi_fused_add_mean_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=[8],
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_mean_mul_7', '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_add_mean_mul_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8
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')
tmp2 = tl.load(in_ptr2 + (0))
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
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')
tmp11 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr2 + (1))
tmp14 = tl.broadcast_to(tmp13, [XBLOCK])
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')
tmp4 = tmp1 * tmp3
tmp5 = tmp0 + tmp4
tmp8 = tmp7 * tmp3
tmp9 = tmp6 + tmp8
tmp10 = tmp5 + tmp9
tmp15 = tmp12 * tmp14
tmp16 = tmp11 + tmp15
tmp17 = tmp10 + tmp16
tmp20 = tmp19 * tmp14
tmp21 = tmp18 + tmp20
tmp22 = tmp17 + tmp21
tmp23 = 4.0
tmp24 = tmp22 / tmp23
tl.store(out_ptr0 + (x0), tmp24, 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, (2, 2, 3, 3), (18, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 4, 4), (32, 16, 4, 1))
buf0 = empty_strided_cuda((4, 2, 6, 6), (72, 36, 6, 1), torch.float32)
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((4, 2, 6, 6), (72, 36, 6, 1), torch.float32)
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.copy]
stream0 = get_raw_stream(0)
triton_poi_fused_copy_0.run(primals_2, buf0, buf1, 288, grid=grid(288), stream=stream0)
del primals_2
buf2 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(buf1, buf2, 288, grid=grid(288), stream=stream0)
del buf1
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.convolution]
buf3 = extern_kernels.convolution(buf2, primals_1, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 2, 2, 2), (8, 4, 2, 1))
buf4 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.bool)
buf5 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.leaky_relu]
triton_poi_fused_leaky_relu_2.run(buf3, buf4, buf5, 32, grid=grid(32), stream=stream0)
buf6 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten._to_copy]
triton_poi_fused__to_copy_3.run(buf6, 2, grid=grid(2), stream=stream0)
buf7 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.add, aten.clamp]
triton_poi_fused_add_clamp_4.run(buf7, 2, grid=grid(2), stream=stream0)
buf8 = empty_strided_cuda((2, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_3.run(buf8, 2, grid=grid(2), stream=stream0)
buf9 = empty_strided_cuda((2, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.add, aten.clamp]
triton_poi_fused_add_clamp_4.run(buf9, 2, grid=grid(2), stream=stream0)
buf12 = empty_strided_cuda((2, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.arange, aten._to_copy, aten.add, aten.mul, aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5.run(buf12, 2, grid=grid(2), stream=stream0)
buf11 = buf3; del buf3 # reuse
buf10 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf13 = buf10; del buf10 # reuse
buf15 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [output_2, output_3], Original ATen: [aten.avg_pool2d, aten._unsafe_index, aten.sub, aten.mul, aten.add]
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6.run(buf13, buf15, buf7, buf8, buf5, buf6, buf9, buf12, 32, grid=grid(32), stream=stream0)
buf14 = empty_strided_cuda((2, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.sub, aten.clamp]
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5.run(buf14, 2, grid=grid(2), stream=stream0)
buf16 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_3, output_4], Original ATen: [aten.mul, aten.add, aten.mean]
triton_poi_fused_add_mean_mul_7.run(buf13, buf15, buf14, buf16, 8, grid=grid(8), stream=stream0)
del buf13
del buf15
return (buf16, primals_1, buf2, buf4, buf5, buf6, buf7, buf8, buf9, buf12, 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((2, 2, 3, 3), (18, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 2, 4, 4), (32, 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.functional as F
class RefModel2d2(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, padding=1, stride=2,
padding_mode='circular', bias=False)
self.l2 = torch.nn.Identity()
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='bilinear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 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
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_copy_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 288
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
x4 = xindex
tmp0 = x0
tmp1 = tl.full([1], 5, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = -4 + x0
tmp4 = tl.full([1], 1, tl.int64)
tmp5 = tmp3 < tmp4
tmp6 = tmp5 & tmp2
tmp7 = tmp0 >= tmp4
tmp8 = tmp0 < tmp1
tmp9 = tmp7 & tmp8
tmp10 = tmp9 & tmp6
tmp11 = x1
tmp12 = tmp11 >= tmp4
tmp13 = tmp11 < tmp1
tmp14 = tmp12 & tmp13
tmp15 = tmp14 & tmp10
tmp16 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp15 & xmask,
other=0.0)
tmp17 = tl.load(in_ptr1 + x4, tmp10 & xmask, other=0.0)
tmp18 = tl.where(tmp14, tmp16, tmp17)
tmp19 = tl.full(tmp18.shape, 0.0, tmp18.dtype)
tmp20 = tl.where(tmp10, tmp18, tmp19)
tmp21 = float('nan')
tmp22 = tl.where(tmp9, tmp20, tmp21)
tmp23 = tl.full(tmp22.shape, 0.0, tmp22.dtype)
tmp24 = tl.where(tmp6, tmp22, tmp23)
tmp25 = tmp3 >= tmp4
tmp26 = tmp3 < tmp1
tmp27 = tmp25 & tmp26
tmp28 = tmp27 & tmp2
tmp29 = tmp14 & tmp28
tmp30 = tl.load(in_ptr0 + (-9 + x0 + 4 * x1 + 16 * x2), tmp29 & xmask,
other=0.0)
tmp31 = tl.load(in_ptr1 + (-4 + x4), tmp28 & xmask, other=0.0)
tmp32 = tl.where(tmp14, tmp30, tmp31)
tmp33 = tl.full(tmp32.shape, 0.0, tmp32.dtype)
tmp34 = tl.where(tmp28, tmp32, tmp33)
tmp35 = tl.where(tmp27, tmp34, tmp21)
tmp36 = tl.where(tmp5, tmp24, tmp35)
tmp37 = tl.full(tmp36.shape, 0.0, tmp36.dtype)
tmp38 = tl.where(tmp2, tmp36, tmp37)
tmp39 = tmp0 < tmp4
tmp40 = 4 + x0
tmp41 = tmp40 >= tmp4
tmp42 = tmp40 < tmp1
tmp43 = tmp41 & tmp42
tmp44 = tmp43 & tmp39
tmp45 = tmp14 & tmp44
tmp46 = tl.load(in_ptr0 + (-1 + x0 + 4 * x1 + 16 * x2), tmp45 & xmask,
other=0.0)
tmp47 = tl.load(in_ptr1 + (4 + x4), tmp44 & xmask, other=0.0)
tmp48 = tl.where(tmp14, tmp46, tmp47)
tmp49 = tl.full(tmp48.shape, 0.0, tmp48.dtype)
tmp50 = tl.where(tmp44, tmp48, tmp49)
tmp51 = tl.where(tmp43, tmp50, tmp21)
tmp52 = tl.full(tmp51.shape, 0.0, tmp51.dtype)
tmp53 = tl.where(tmp39, tmp51, tmp52)
tmp54 = tmp14 & tmp9
tmp55 = tl.load(in_ptr0 + (-5 + x0 + 4 * x1 + 16 * x2), tmp54 & xmask,
other=0.0)
tmp56 = tl.load(in_ptr1 + x4, tmp9 & xmask, other=0.0)
tmp57 = tl.where(tmp14, tmp55, tmp56)
tmp58 = tl.full(tmp57.shape, 0.0, tmp57.dtype)
tmp59 = tl.where(tmp9, tmp57, tmp58)
tmp60 = tl.where(tmp9, tmp59, tmp21)
tmp61 = tl.where(tmp39, tmp53, tmp60)
tmp62 = tl.where(tmp2, tmp38, tmp61)
tl.store(out_ptr0 + x4, tmp62, xmask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 288
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
x3 = xindex
tmp14 = tl.load(in_ptr0 + x3, xmask)
tmp0 = x1
tmp1 = tl.full([1], 5, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = -4 + x1
tmp4 = tl.full([1], 1, tl.int64)
tmp5 = tmp3 < tmp4
tmp6 = tmp5 & tmp2
tmp7 = tl.load(in_ptr0 + (24 + x0 + 36 * x2), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp8 = tl.load(in_ptr0 + (-24 + x3), tmp2 & xmask, other=0.0)
tmp9 = tl.where(tmp5, tmp7, tmp8)
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp2, tmp9, tmp10)
tmp12 = tmp0 < tmp4
tmp13 = tl.load(in_ptr0 + (24 + x0 + 36 * x2), tmp12 & xmask,
eviction_policy='evict_last', other=0.0)
tmp15 = tl.where(tmp12, tmp13, tmp14)
tmp16 = tl.where(tmp2, tmp11, tmp15)
tl.store(out_ptr0 + x3, tmp16, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_2(in_ptr0, out_ptr0, out_ptr1, xnumel,
XBLOCK: tl.constexpr):
xnumel = 32
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 = 0.02
tmp4 = tmp0 * tmp3
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp2, xmask)
tl.store(out_ptr1 + x0, tmp5, xmask)
@triton.jit
def triton_poi_fused__to_copy_3(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tl.store(out_ptr0 + x0, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_clamp_4(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 + tmp9
tmp11 = tl.full([1], 0, tl.int64)
tmp12 = triton_helpers.minimum(tmp10, tmp11)
tl.store(out_ptr0 + x0, tmp12, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 2
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = tmp3 * tmp2
tmp5 = tmp4 - tmp2
tmp6 = 0.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tmp8.to(tl.float32)
tmp10 = tmp7 - tmp9
tmp11 = triton_helpers.maximum(tmp10, tmp6)
tmp12 = 1.0
tmp13 = triton_helpers.minimum(tmp11, tmp12)
tl.store(out_ptr0 + x0, tmp13, xmask)
@triton.jit
def triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6(in_out_ptr0,
in_out_ptr1, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5,
xnumel, XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 2 % 2
x0 = xindex % 2
x2 = xindex // 4
x4 = xindex
tmp0 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr5 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 1, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr2 + (4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + (1 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp11 = tmp10 + tmp9
tmp12 = tl.load(in_ptr2 + (2 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp13 = tmp12 + tmp11
tmp14 = tl.load(in_ptr2 + (3 + 4 * tmp8 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = 0.25
tmp17 = tmp15 * tmp16
tmp19 = tmp18 + tmp1
tmp20 = tmp18 < 0
tmp21 = tl.where(tmp20, tmp19, tmp18)
tmp22 = tl.load(in_ptr2 + (4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp23 = tl.load(in_ptr2 + (1 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp24 = tmp23 + tmp22
tmp25 = tl.load(in_ptr2 + (2 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp26 = tmp25 + tmp24
tmp27 = tl.load(in_ptr2 + (3 + 4 * tmp8 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp28 = tmp27 + tmp26
tmp29 = tmp28 * tmp16
tmp31 = tmp30 + tmp1
tmp32 = tmp30 < 0
tmp33 = tl.where(tmp32, tmp31, tmp30)
tmp34 = tl.load(in_ptr2 + (4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp35 = tl.load(in_ptr2 + (1 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp36 = tmp35 + tmp34
tmp37 = tl.load(in_ptr2 + (2 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp38 = tmp37 + tmp36
tmp39 = tl.load(in_ptr2 + (3 + 4 * tmp33 + 4 * x2 + 8 * tmp21), xmask,
eviction_policy='evict_last')
tmp40 = tmp39 + tmp38
tmp41 = tmp40 * tmp16
tmp42 = tmp41 - tmp29
tmp44 = tmp42 * tmp43
tmp45 = tmp29 + tmp44
tmp46 = tl.load(in_ptr2 + (4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp47 = tl.load(in_ptr2 + (1 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp48 = tmp47 + tmp46
tmp49 = tl.load(in_ptr2 + (2 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp50 = tmp49 + tmp48
tmp51 = tl.load(in_ptr2 + (3 + 4 * tmp33 + 4 * x2 + 8 * tmp4), xmask,
eviction_policy='evict_last')
tmp52 = tmp51 + tmp50
tmp53 = tmp52 * tmp16
tmp54 = tmp53 - tmp17
tmp55 = tmp54 * tmp43
tmp56 = tmp17 + tmp55
tmp57 = tmp56 - tmp45
tl.store(in_out_ptr0 + x4, tmp45, xmask)
tl.store(in_out_ptr1 + x4, tmp57, xmask)
@triton.jit
def triton_poi_fused_add_mean_mul_7(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 8
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')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
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')
tmp11 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr2 + 1)
tmp14 = tl.broadcast_to(tmp13, [XBLOCK])
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'
)
tmp4 = tmp1 * tmp3
tmp5 = tmp0 + tmp4
tmp8 = tmp7 * tmp3
tmp9 = tmp6 + tmp8
tmp10 = tmp5 + tmp9
tmp15 = tmp12 * tmp14
tmp16 = tmp11 + tmp15
tmp17 = tmp10 + tmp16
tmp20 = tmp19 * tmp14
tmp21 = tmp18 + tmp20
tmp22 = tmp17 + tmp21
tmp23 = 4.0
tmp24 = tmp22 / tmp23
tl.store(out_ptr0 + x0, tmp24, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 3, 3), (18, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 4, 4), (32, 16, 4, 1))
buf0 = empty_strided_cuda((4, 2, 6, 6), (72, 36, 6, 1), torch.float32)
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((4, 2, 6, 6), (72, 36, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_copy_0[grid(288)](primals_2, buf0, buf1, 288,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = buf0
del buf0
triton_poi_fused_1[grid(288)](buf1, buf2, 288, XBLOCK=256,
num_warps=4, num_stages=1)
del buf1
buf3 = extern_kernels.convolution(buf2, primals_1, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 2, 2, 2), (8, 4, 2, 1))
buf4 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.bool)
buf5 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
triton_poi_fused_leaky_relu_2[grid(32)](buf3, buf4, buf5, 32,
XBLOCK=32, num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_3[grid(2)](buf6, 2, XBLOCK=2, num_warps=1,
num_stages=1)
buf7 = empty_strided_cuda((2, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_4[grid(2)](buf7, 2, XBLOCK=2, num_warps=
1, num_stages=1)
buf8 = empty_strided_cuda((2,), (1,), torch.int64)
triton_poi_fused__to_copy_3[grid(2)](buf8, 2, XBLOCK=2, num_warps=1,
num_stages=1)
buf9 = empty_strided_cuda((2,), (1,), torch.int64)
triton_poi_fused_add_clamp_4[grid(2)](buf9, 2, XBLOCK=2, num_warps=
1, num_stages=1)
buf12 = empty_strided_cuda((2,), (1,), torch.float32)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5[grid(2)](buf12,
2, XBLOCK=2, num_warps=1, num_stages=1)
buf11 = buf3
del buf3
buf10 = empty_strided_cuda((4, 2, 2, 2), (8, 4, 2, 1), torch.float32)
buf13 = buf10
del buf10
buf15 = buf11
del buf11
triton_poi_fused__unsafe_index_add_avg_pool2d_mul_sub_6[grid(32)](buf13
, buf15, buf7, buf8, buf5, buf6, buf9, buf12, 32, XBLOCK=32,
num_warps=1, num_stages=1)
buf14 = empty_strided_cuda((2, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_add_arange_clamp_mul_sub_5[grid(2)](buf14,
2, XBLOCK=2, num_warps=1, num_stages=1)
buf16 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
triton_poi_fused_add_mean_mul_7[grid(8)](buf13, buf15, buf14, buf16,
8, XBLOCK=8, num_warps=1, num_stages=1)
del buf13
del buf15
return (buf16, primals_1, buf2, buf4, buf5, buf6, buf7, buf8, buf9,
buf12, buf14)
class RefModel2d2New(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, padding=1, stride=2,
padding_mode='circular', bias=False)
self.l2 = torch.nn.Identity()
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, input_0):
primals_1 = self.l1.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
| shuohan/pytorch-layers | RefModel2d2 | false | 4,343 | [
"MIT"
] | 0 | 020846fd02d501cf477552179c19ba4b5e9a0695 | https://github.com/shuohan/pytorch-layers/tree/020846fd02d501cf477552179c19ba4b5e9a0695 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""The 2D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv2d(2, 2, 3, padding=1, stride=2,
padding_mode='circular', bias=False)
self.l2 = torch.nn.Identity()
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool2d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool2d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='bilinear', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 4, 4])]
def get_init_inputs():
return []
|
PositionAttentionModule | # 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_7/inductor_cache/u3/cu3litezfpnwhpnfnfuj6dtimz6ml42wmcwnwxlnovd4p5lvyin4.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=[2048, 4096], 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_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 = 2048
xnumel = 4096
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 = tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = (yindex // 512)
tmp0 = tl.load(in_ptr0 + (x2 + (4096*y3)), None, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (512*x2) + (2097152*y1)), tmp0, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/oc/cochsno6wpkwamgsqz5legelnxxchuje5twfzhozvusus3e5bzmo.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=[262144, 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_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 = 262144
xnumel = 9
yoffset = (tl.program_id(1) + tl.program_id(2) * tl.num_programs(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 % 512
y1 = (yindex // 512)
tmp0 = tl.load(in_ptr0 + (x2 + (9*y3)), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (512*x2) + (4608*y1)), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/a3/ca3b7scxnvwtz5kednjbz6knhbbsrdcpcqxmv2rtdrxdr73frape.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 = (%permute,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_2 = async_compile.triton('triton_poi_fused_clone_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=[8388608],
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_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_clone_2(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8388608
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 512
tmp0 = tl.load(in_out_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr0 + (x0), None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ub/cubtzfxsytxlbhhtcnbpcrd7btnmqhljo6a5wycfz3ohm45phqdm.py
# Topologically Sorted Source Nodes: [wrapped_sqrt, att_1], Original ATen: [aten.sqrt, aten._softmax]
# Source node to ATen node mapping:
# att_1 => div_1, exp, sum_1
# wrapped_sqrt => full_default
# Graph fragment:
# %full_default : [num_users=2] = call_function[target=torch.ops.aten.full.default](args = ([], 22.62741699796952), kwargs = {dtype: torch.float64, layout: torch.strided, device: cpu, pin_memory: False})
# %scalar_tensor_default : [num_users=2] = call_function[target=torch.ops.aten.scalar_tensor.default](args = (1,), kwargs = {dtype: torch.float32, device: cuda:0, pin_memory: False})
# %ge_scalar : [num_users=1] = call_function[target=torch.ops.aten.ge.Scalar](args = (%full_default, 0), kwargs = {})
# %neg_default : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%scalar_tensor_default,), kwargs = {})
# %where_self : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%ge_scalar, %scalar_tensor_default, %neg_default), kwargs = {})
# %mul_tensor : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_12, %where_self), 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 = (%where_self, %full_default), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_tensor, %mul_tensor_1), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%div_tensor,), kwargs = {})
# %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_red_fused__softmax_sqrt_3 = async_compile.triton('triton_red_fused__softmax_sqrt_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.reduction(
size_hints=[16384, 4096],
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_red_fused__softmax_sqrt_3', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, '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_red_fused__softmax_sqrt_3(in_ptr0, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 16384
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
_tmp9 = tl.full([XBLOCK, RBLOCK], float("-inf"), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask, eviction_policy='evict_last', other=0.0)
tmp1 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp2 = tl.full([1, 1], 0.0, tl.float64)
tmp3 = tmp1 >= tmp2
tmp4 = 1.0
tmp5 = -1.0
tmp6 = tl.where(tmp3, tmp4, tmp5)
tmp7 = tmp0 * tmp6
tmp8 = tl.broadcast_to(tmp7, [XBLOCK, RBLOCK])
tmp10 = triton_helpers.maximum(_tmp9, tmp8)
_tmp9 = tl.where(rmask, tmp10, _tmp9)
tmp9 = triton_helpers.max2(_tmp9, 1)[:, None]
_tmp26 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp11 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask, eviction_policy='evict_last', other=0.0)
tmp12 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp13 = tl.full([1, 1], 0.0, tl.float64)
tmp14 = tmp12 >= tmp13
tmp15 = 1.0
tmp16 = -1.0
tmp17 = tl.where(tmp14, tmp15, tmp16)
tmp18 = tmp11 * tmp17
tmp19 = tmp18 - tmp9
tmp20 = tmp17.to(tl.float64)
tmp21 = tmp20 * tmp12
tmp22 = tmp21.to(tl.float32)
tmp23 = tmp19 / tmp22
tmp24 = tl_math.exp(tmp23)
tmp25 = tl.broadcast_to(tmp24, [XBLOCK, RBLOCK])
tmp27 = _tmp26 + tmp25
_tmp26 = tl.where(rmask, tmp27, _tmp26)
tmp26 = tl.sum(_tmp26, 1)[:, None]
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp28 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask, eviction_policy='evict_first', other=0.0)
tmp29 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp30 = tl.full([1, 1], 0.0, tl.float64)
tmp31 = tmp29 >= tmp30
tmp32 = 1.0
tmp33 = -1.0
tmp34 = tl.where(tmp31, tmp32, tmp33)
tmp35 = tmp28 * tmp34
tmp36 = tmp35 - tmp9
tmp37 = tmp34.to(tl.float64)
tmp38 = tmp37 * tmp29
tmp39 = tmp38.to(tl.float32)
tmp40 = tmp36 / tmp39
tmp41 = tl_math.exp(tmp40)
tmp42 = tmp41 / tmp26
tl.store(out_ptr2 + (r1 + (4096*x0)), tmp42, rmask)
''', 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, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (512, 512, 3, 3), (4608, 9, 3, 1))
assert_size_stride(primals_3, (512, ), (1, ))
assert_size_stride(primals_4, (512, 512), (512, 1))
assert_size_stride(primals_5, (512, ), (1, ))
assert_size_stride(primals_6, (512, 512), (512, 1))
assert_size_stride(primals_7, (512, ), (1, ))
assert_size_stride(primals_8, (512, 512), (512, 1))
assert_size_stride(primals_9, (512, ), (1, ))
assert_size_stride(primals_10, (512, 512), (512, 1))
assert_size_stride(primals_11, (512, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(primals_1, buf0, 2048, 4096, grid=grid(2048, 4096), stream=stream0)
del primals_1
buf1 = empty_strided_cuda((512, 512, 3, 3), (4608, 1, 1536, 512), torch.float32)
# Unsorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(primals_2, buf1, 262144, 9, grid=grid(262144, 9), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [y], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf0, buf1, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 512, 64, 64), (2097152, 1, 32768, 512))
buf3 = reinterpret_tensor(buf2, (4, 4096, 512), (2097152, 512, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.clone]
triton_poi_fused_clone_2.run(buf3, primals_3, 8388608, grid=grid(8388608), stream=stream0)
del primals_3
buf4 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1), 0), reinterpret_tensor(primals_4, (512, 512), (1, 512), 0), out=buf4)
buf5 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1), 0), reinterpret_tensor(primals_6, (512, 512), (1, 512), 0), out=buf5)
buf6 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_2], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1), 0), reinterpret_tensor(primals_8, (512, 512), (1, 512), 0), out=buf6)
buf7 = reinterpret_tensor(buf4, (4, 4096, 512), (2097152, 512, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [linear], Original ATen: [aten.add]
triton_poi_fused_clone_2.run(buf7, primals_5, 8388608, grid=grid(8388608), stream=stream0)
del primals_5
buf8 = reinterpret_tensor(buf5, (4, 4096, 512), (2097152, 512, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.add]
triton_poi_fused_clone_2.run(buf8, primals_7, 8388608, grid=grid(8388608), stream=stream0)
del primals_7
buf9 = empty_strided_cuda((4, 4096, 4096), (16777216, 4096, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul], Original ATen: [aten.bmm]
extern_kernels.bmm(buf7, reinterpret_tensor(buf8, (4, 512, 4096), (2097152, 1, 512), 0), out=buf9)
buf12 = empty_strided_cuda((4, 1, 4096, 4096), (16777216, 1, 4096, 1), torch.float32)
# Topologically Sorted Source Nodes: [wrapped_sqrt, att_1], Original ATen: [aten.sqrt, aten._softmax]
triton_red_fused__softmax_sqrt_3.run(buf9, buf12, 16384, 4096, grid=grid(16384), stream=stream0)
del buf9
buf13 = reinterpret_tensor(buf6, (4, 4096, 512), (2097152, 512, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [linear_2], Original ATen: [aten.add]
triton_poi_fused_clone_2.run(buf13, primals_9, 8388608, grid=grid(8388608), stream=stream0)
del primals_9
buf14 = empty_strided_cuda((4, 4096, 512), (2097152, 512, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul_1], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf12, (4, 4096, 4096), (16777216, 4096, 1), 0), buf13, out=buf14)
buf15 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf14, (16384, 512), (512, 1), 0), reinterpret_tensor(primals_10, (512, 512), (1, 512), 0), alpha=1, beta=1, out=buf15)
del primals_11
return (reinterpret_tensor(buf15, (4, 4096, 512), (2097152, 512, 1), 0), buf0, buf1, reinterpret_tensor(buf3, (16384, 512), (512, 1), 0), buf12, reinterpret_tensor(buf14, (16384, 512), (512, 1), 0), primals_10, reinterpret_tensor(buf13, (4, 512, 4096), (2097152, 1, 512), 0), reinterpret_tensor(buf7, (4, 512, 4096), (2097152, 1, 512), 0), buf8, 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, 512, 64, 64), (2097152, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((512, 512, 3, 3), (4608, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((512, 512), (512, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((512, 512), (512, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((512, 512), (512, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((512, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((512, 512), (512, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((512, ), (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 numpy as np
from torch import nn
from torch.nn import init
class ScaledDotProductAttention(nn.Module):
"""
Scaled dot-product attention
"""
def __init__(self, d_model, d_k, d_v, h, dropout=0.1):
"""
:param d_model: Output dimensionality of the model
:param d_k: Dimensionality of queries and keys
:param d_v: Dimensionality of values
:param h: Number of heads
"""
super(ScaledDotProductAttention, self).__init__()
self.fc_q = nn.Linear(d_model, h * d_k)
self.fc_k = nn.Linear(d_model, h * d_k)
self.fc_v = nn.Linear(d_model, h * d_v)
self.fc_o = nn.Linear(h * d_v, d_model)
self.dropout = nn.Dropout(dropout)
self.d_model = d_model
self.d_k = d_k
self.d_v = d_v
self.h = h
self.init_weights()
def init_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
init.kaiming_normal_(m.weight, mode='fan_out')
if m.bias is not None:
init.constant_(m.bias, 0)
elif isinstance(m, nn.BatchNorm2d):
init.constant_(m.weight, 1)
init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
init.normal_(m.weight, std=0.001)
if m.bias is not None:
init.constant_(m.bias, 0)
def forward(self, queries, keys, values, attention_mask=None,
attention_weights=None):
"""
Computes
:param queries: Queries (b_s, nq, d_model)
:param keys: Keys (b_s, nk, d_model)
:param values: Values (b_s, nk, d_model)
:param attention_mask: Mask over attention values (b_s, h, nq, nk). True indicates masking.
:param attention_weights: Multiplicative weights for attention values (b_s, h, nq, nk).
:return:
"""
b_s, nq = queries.shape[:2]
nk = keys.shape[1]
q = self.fc_q(queries).view(b_s, nq, self.h, self.d_k).permute(0, 2,
1, 3)
k = self.fc_k(keys).view(b_s, nk, self.h, self.d_k).permute(0, 2, 3, 1)
v = self.fc_v(values).view(b_s, nk, self.h, self.d_v).permute(0, 2,
1, 3)
att = torch.matmul(q, k) / np.sqrt(self.d_k)
if attention_weights is not None:
att = att * attention_weights
if attention_mask is not None:
att = att.masked_fill(attention_mask, -np.inf)
att = torch.softmax(att, -1)
att = self.dropout(att)
out = torch.matmul(att, v).permute(0, 2, 1, 3).contiguous().view(b_s,
nq, self.h * self.d_v)
out = self.fc_o(out)
return out
class PositionAttentionModule(nn.Module):
def __init__(self, d_model=512, kernel_size=3, H=7, W=7):
super().__init__()
self.cnn = nn.Conv2d(d_model, d_model, kernel_size=kernel_size,
padding=(kernel_size - 1) // 2)
self.pa = ScaledDotProductAttention(d_model, d_k=d_model, d_v=
d_model, h=1)
def forward(self, x):
bs, c, _h, _w = x.shape
y = self.cnn(x)
y = y.view(bs, c, -1).permute(0, 2, 1)
y = self.pa(y, y, y)
return y
def get_inputs():
return [torch.rand([4, 512, 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 numpy as np
from torch import nn
from torch.nn import 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_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
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]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 512
y1 = yindex // 512
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y3), None, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 512 * x2 + 2097152 * y1), tmp0, None)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = (tl.program_id(1) + tl.program_id(2) * tl.num_programs(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 % 512
y1 = yindex // 512
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 512 * x2 + 4608 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_clone_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)
x2 = xindex
x0 = xindex % 512
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, None)
@triton.jit
def triton_red_fused__softmax_sqrt_3(in_ptr0, out_ptr2, xnumel, rnumel,
XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
_tmp9 = tl.full([XBLOCK, RBLOCK], float('-inf'), tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask, eviction_policy=
'evict_last', other=0.0)
tmp1 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp2 = tl.full([1, 1], 0.0, tl.float64)
tmp3 = tmp1 >= tmp2
tmp4 = 1.0
tmp5 = -1.0
tmp6 = tl.where(tmp3, tmp4, tmp5)
tmp7 = tmp0 * tmp6
tmp8 = tl.broadcast_to(tmp7, [XBLOCK, RBLOCK])
tmp10 = triton_helpers.maximum(_tmp9, tmp8)
_tmp9 = tl.where(rmask, tmp10, _tmp9)
tmp9 = triton_helpers.max2(_tmp9, 1)[:, None]
_tmp26 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp11 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask, eviction_policy=
'evict_last', other=0.0)
tmp12 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp13 = tl.full([1, 1], 0.0, tl.float64)
tmp14 = tmp12 >= tmp13
tmp15 = 1.0
tmp16 = -1.0
tmp17 = tl.where(tmp14, tmp15, tmp16)
tmp18 = tmp11 * tmp17
tmp19 = tmp18 - tmp9
tmp20 = tmp17.to(tl.float64)
tmp21 = tmp20 * tmp12
tmp22 = tmp21.to(tl.float32)
tmp23 = tmp19 / tmp22
tmp24 = tl_math.exp(tmp23)
tmp25 = tl.broadcast_to(tmp24, [XBLOCK, RBLOCK])
tmp27 = _tmp26 + tmp25
_tmp26 = tl.where(rmask, tmp27, _tmp26)
tmp26 = tl.sum(_tmp26, 1)[:, None]
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp28 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask, eviction_policy=
'evict_first', other=0.0)
tmp29 = tl.full([1, 1], 22.62741699796952, tl.float64)
tmp30 = tl.full([1, 1], 0.0, tl.float64)
tmp31 = tmp29 >= tmp30
tmp32 = 1.0
tmp33 = -1.0
tmp34 = tl.where(tmp31, tmp32, tmp33)
tmp35 = tmp28 * tmp34
tmp36 = tmp35 - tmp9
tmp37 = tmp34.to(tl.float64)
tmp38 = tmp37 * tmp29
tmp39 = tmp38.to(tl.float32)
tmp40 = tmp36 / tmp39
tmp41 = tl_math.exp(tmp40)
tmp42 = tmp41 / tmp26
tl.store(out_ptr2 + (r1 + 4096 * x0), tmp42, rmask)
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, 512, 64, 64), (2097152, 4096, 64, 1))
assert_size_stride(primals_2, (512, 512, 3, 3), (4608, 9, 3, 1))
assert_size_stride(primals_3, (512,), (1,))
assert_size_stride(primals_4, (512, 512), (512, 1))
assert_size_stride(primals_5, (512,), (1,))
assert_size_stride(primals_6, (512, 512), (512, 1))
assert_size_stride(primals_7, (512,), (1,))
assert_size_stride(primals_8, (512, 512), (512, 1))
assert_size_stride(primals_9, (512,), (1,))
assert_size_stride(primals_10, (512, 512), (512, 1))
assert_size_stride(primals_11, (512,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 512, 64, 64), (2097152, 1, 32768, 512
), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(2048, 4096)](primals_1, buf0, 2048, 4096,
XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((512, 512, 3, 3), (4608, 1, 1536, 512),
torch.float32)
triton_poi_fused_1[grid(262144, 9)](primals_2, buf1, 262144, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf0, buf1, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 512, 64, 64), (2097152, 1, 32768, 512))
buf3 = reinterpret_tensor(buf2, (4, 4096, 512), (2097152, 512, 1), 0)
del buf2
triton_poi_fused_clone_2[grid(8388608)](buf3, primals_3, 8388608,
XBLOCK=1024, num_warps=4, num_stages=1)
del primals_3
buf4 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1),
0), reinterpret_tensor(primals_4, (512, 512), (1, 512), 0), out
=buf4)
buf5 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1),
0), reinterpret_tensor(primals_6, (512, 512), (1, 512), 0), out
=buf5)
buf6 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (16384, 512), (512, 1),
0), reinterpret_tensor(primals_8, (512, 512), (1, 512), 0), out
=buf6)
buf7 = reinterpret_tensor(buf4, (4, 4096, 512), (2097152, 512, 1), 0)
del buf4
triton_poi_fused_clone_2[grid(8388608)](buf7, primals_5, 8388608,
XBLOCK=1024, num_warps=4, num_stages=1)
del primals_5
buf8 = reinterpret_tensor(buf5, (4, 4096, 512), (2097152, 512, 1), 0)
del buf5
triton_poi_fused_clone_2[grid(8388608)](buf8, primals_7, 8388608,
XBLOCK=1024, num_warps=4, num_stages=1)
del primals_7
buf9 = empty_strided_cuda((4, 4096, 4096), (16777216, 4096, 1),
torch.float32)
extern_kernels.bmm(buf7, reinterpret_tensor(buf8, (4, 512, 4096), (
2097152, 1, 512), 0), out=buf9)
buf12 = empty_strided_cuda((4, 1, 4096, 4096), (16777216, 1, 4096,
1), torch.float32)
triton_red_fused__softmax_sqrt_3[grid(16384)](buf9, buf12, 16384,
4096, XBLOCK=1, RBLOCK=2048, num_warps=16, num_stages=1)
del buf9
buf13 = reinterpret_tensor(buf6, (4, 4096, 512), (2097152, 512, 1), 0)
del buf6
triton_poi_fused_clone_2[grid(8388608)](buf13, primals_9, 8388608,
XBLOCK=1024, num_warps=4, num_stages=1)
del primals_9
buf14 = empty_strided_cuda((4, 4096, 512), (2097152, 512, 1), torch
.float32)
extern_kernels.bmm(reinterpret_tensor(buf12, (4, 4096, 4096), (
16777216, 4096, 1), 0), buf13, out=buf14)
buf15 = empty_strided_cuda((16384, 512), (512, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf14, (16384,
512), (512, 1), 0), reinterpret_tensor(primals_10, (512, 512),
(1, 512), 0), alpha=1, beta=1, out=buf15)
del primals_11
return reinterpret_tensor(buf15, (4, 4096, 512), (2097152, 512, 1), 0
), buf0, buf1, reinterpret_tensor(buf3, (16384, 512), (512, 1), 0
), buf12, reinterpret_tensor(buf14, (16384, 512), (512, 1), 0
), primals_10, reinterpret_tensor(buf13, (4, 512, 4096), (2097152,
1, 512), 0), reinterpret_tensor(buf7, (4, 512, 4096), (2097152, 1,
512), 0), buf8, primals_8, primals_6, primals_4
class ScaledDotProductAttention(nn.Module):
"""
Scaled dot-product attention
"""
def __init__(self, d_model, d_k, d_v, h, dropout=0.1):
"""
:param d_model: Output dimensionality of the model
:param d_k: Dimensionality of queries and keys
:param d_v: Dimensionality of values
:param h: Number of heads
"""
super(ScaledDotProductAttention, self).__init__()
self.fc_q = nn.Linear(d_model, h * d_k)
self.fc_k = nn.Linear(d_model, h * d_k)
self.fc_v = nn.Linear(d_model, h * d_v)
self.fc_o = nn.Linear(h * d_v, d_model)
self.dropout = nn.Dropout(dropout)
self.d_model = d_model
self.d_k = d_k
self.d_v = d_v
self.h = h
self.init_weights()
def init_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
init.kaiming_normal_(m.weight, mode='fan_out')
if m.bias is not None:
init.constant_(m.bias, 0)
elif isinstance(m, nn.BatchNorm2d):
init.constant_(m.weight, 1)
init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
init.normal_(m.weight, std=0.001)
if m.bias is not None:
init.constant_(m.bias, 0)
def forward(self, queries, keys, values, attention_mask=None,
attention_weights=None):
"""
Computes
:param queries: Queries (b_s, nq, d_model)
:param keys: Keys (b_s, nk, d_model)
:param values: Values (b_s, nk, d_model)
:param attention_mask: Mask over attention values (b_s, h, nq, nk). True indicates masking.
:param attention_weights: Multiplicative weights for attention values (b_s, h, nq, nk).
:return:
"""
b_s, nq = queries.shape[:2]
nk = keys.shape[1]
q = self.fc_q(queries).view(b_s, nq, self.h, self.d_k).permute(0, 2,
1, 3)
k = self.fc_k(keys).view(b_s, nk, self.h, self.d_k).permute(0, 2, 3, 1)
v = self.fc_v(values).view(b_s, nk, self.h, self.d_v).permute(0, 2,
1, 3)
att = torch.matmul(q, k) / np.sqrt(self.d_k)
if attention_weights is not None:
att = att * attention_weights
if attention_mask is not None:
att = att.masked_fill(attention_mask, -np.inf)
att = torch.softmax(att, -1)
att = self.dropout(att)
out = torch.matmul(att, v).permute(0, 2, 1, 3).contiguous().view(b_s,
nq, self.h * self.d_v)
out = self.fc_o(out)
return out
class PositionAttentionModuleNew(nn.Module):
def __init__(self, d_model=512, kernel_size=3, H=7, W=7):
super().__init__()
self.cnn = nn.Conv2d(d_model, d_model, kernel_size=kernel_size,
padding=(kernel_size - 1) // 2)
self.pa = ScaledDotProductAttention(d_model, d_k=d_model, d_v=
d_model, h=1)
def forward(self, input_0):
primals_2 = self.cnn.weight
primals_3 = self.cnn.bias
primals_4 = self.pa.fc_q.weight
primals_5 = self.pa.fc_q.bias
primals_6 = self.pa.fc_k.weight
primals_7 = self.pa.fc_k.bias
primals_8 = self.pa.fc_v.weight
primals_9 = self.pa.fc_v.bias
primals_10 = self.pa.fc_o.weight
primals_11 = self.pa.fc_o.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,
primals_10, primals_11])
return output[0]
| rushirajsherlocked/External-Attention-pytorch | PositionAttentionModule | false | 4,344 | [
"MIT"
] | 0 | 7d6814b2d90909adf81c62f3f8a89e30a59d6481 | https://github.com/rushirajsherlocked/External-Attention-pytorch/tree/7d6814b2d90909adf81c62f3f8a89e30a59d6481 | import torch
import numpy as np
from torch import nn
from torch.nn import init
class ScaledDotProductAttention(nn.Module):
"""
Scaled dot-product attention
"""
def __init__(self, d_model, d_k, d_v, h, dropout=0.1):
"""
:param d_model: Output dimensionality of the model
:param d_k: Dimensionality of queries and keys
:param d_v: Dimensionality of values
:param h: Number of heads
"""
super().__init__()
self.fc_q = nn.Linear(d_model, h * d_k)
self.fc_k = nn.Linear(d_model, h * d_k)
self.fc_v = nn.Linear(d_model, h * d_v)
self.fc_o = nn.Linear(h * d_v, d_model)
self.dropout = nn.Dropout(dropout)
self.d_model = d_model
self.d_k = d_k
self.d_v = d_v
self.h = h
self.init_weights()
def init_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
init.kaiming_normal_(m.weight, mode='fan_out')
if m.bias is not None:
init.constant_(m.bias, 0)
elif isinstance(m, nn.BatchNorm2d):
init.constant_(m.weight, 1)
init.constant_(m.bias, 0)
elif isinstance(m, nn.Linear):
init.normal_(m.weight, std=0.001)
if m.bias is not None:
init.constant_(m.bias, 0)
def forward(self, queries, keys, values, attention_mask=None,
attention_weights=None):
"""
Computes
:param queries: Queries (b_s, nq, d_model)
:param keys: Keys (b_s, nk, d_model)
:param values: Values (b_s, nk, d_model)
:param attention_mask: Mask over attention values (b_s, h, nq, nk). True indicates masking.
:param attention_weights: Multiplicative weights for attention values (b_s, h, nq, nk).
:return:
"""
b_s, nq = queries.shape[:2]
nk = keys.shape[1]
q = self.fc_q(queries).view(b_s, nq, self.h, self.d_k).permute(0, 2,
1, 3)
k = self.fc_k(keys).view(b_s, nk, self.h, self.d_k).permute(0, 2, 3, 1)
v = self.fc_v(values).view(b_s, nk, self.h, self.d_v).permute(0, 2,
1, 3)
att = torch.matmul(q, k) / np.sqrt(self.d_k)
if attention_weights is not None:
att = att * attention_weights
if attention_mask is not None:
att = att.masked_fill(attention_mask, -np.inf)
att = torch.softmax(att, -1)
att = self.dropout(att)
out = torch.matmul(att, v).permute(0, 2, 1, 3).contiguous().view(b_s,
nq, self.h * self.d_v)
out = self.fc_o(out)
return out
class Model(nn.Module):
def __init__(self, d_model=512, kernel_size=3, H=7, W=7):
super().__init__()
self.cnn = nn.Conv2d(d_model, d_model, kernel_size=kernel_size,
padding=(kernel_size - 1) // 2)
self.pa = ScaledDotProductAttention(d_model, d_k=d_model, d_v=
d_model, h=1)
def forward(self, x):
bs, c, _h, _w = x.shape
y = self.cnn(x)
y = y.view(bs, c, -1).permute(0, 2, 1)
y = self.pa(y, y, y)
return y
def get_inputs():
return [torch.rand([4, 512, 64, 64])]
def get_init_inputs():
return []
|
Actor | # 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_7/inductor_cache/r3/cr3febcwm3t44fuoitsx3ou2p6xg4sk4f7unagmmrvffasxf47te.py
# Topologically Sorted Source Nodes: [a], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# a => 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=[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_7/inductor_cache/cp/ccp5m5apf7ka2skqyfxhf2df54c52qocprpycry7jrzoptyjvbti.py
# Topologically Sorted Source Nodes: [tanh, mul], Original ATen: [aten.tanh, aten.mul]
# Source node to ATen node mapping:
# mul => mul
# tanh => tanh
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%view_5,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%tanh, 4), kwargs = {})
triton_poi_fused_mul_tanh_1 = async_compile.triton('triton_poi_fused_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=[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_mul_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_mul_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 = libdevice.tanh(tmp0)
tmp2 = 4.0
tmp3 = tmp1 * 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, 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: [], 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
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [a], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, primals_2, buf7, 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
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [a_1], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_0.run(buf3, primals_5, buf6, 256, grid=grid(256), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 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: [tanh, mul], Original ATen: [aten.tanh, aten.mul]
triton_poi_fused_mul_tanh_1.run(buf4, buf5, 256, grid=grid(256), stream=stream0)
return (buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(buf3, (64, 4), (4, 1), 0), buf4, primals_6, buf6, 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, 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.functional as F
import torch.nn as nn
class Actor(nn.Module):
def __init__(self, state_dim, action_dim, max_action, nhid):
super(Actor, self).__init__()
self.l1 = nn.Linear(state_dim, nhid)
self.l2 = nn.Linear(nhid, nhid)
self.l3 = nn.Linear(nhid, action_dim)
self.max_action = max_action
def forward(self, state):
a = F.relu(self.l1(state))
a = F.relu(self.l2(a))
return self.max_action * torch.tanh(self.l3(a))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'state_dim': 4, 'action_dim': 4, 'max_action': 4, 'nhid': 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
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_mul_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 = libdevice.tanh(tmp0)
tmp2 = 4.0
tmp3 = tmp1 * tmp2
tl.store(out_ptr0 + x0, tmp3, 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,), (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.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
buf7 = 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, buf7, 256, XBLOCK=256, 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
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf3,
primals_5, buf6, 256, XBLOCK=256, 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, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 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_mul_tanh_1[grid(256)](buf4, buf5, 256, XBLOCK=256,
num_warps=4, num_stages=1)
return buf5, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(
buf3, (64, 4), (4, 1), 0), buf4, primals_6, buf6, primals_4, buf7
class ActorNew(nn.Module):
def __init__(self, state_dim, action_dim, max_action, nhid):
super(ActorNew, self).__init__()
self.l1 = nn.Linear(state_dim, nhid)
self.l2 = nn.Linear(nhid, nhid)
self.l3 = nn.Linear(nhid, action_dim)
self.max_action = max_action
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_6 = self.l3.weight
primals_7 = self.l3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| simondlevy/pytorch-drl | Actor | false | 4,345 | [
"MIT"
] | 0 | b197bb93c2cc698971f98095d4e0180811c52042 | https://github.com/simondlevy/pytorch-drl/tree/b197bb93c2cc698971f98095d4e0180811c52042 | import torch
import torch.nn.functional as F
import torch.nn as nn
class Model(nn.Module):
def __init__(self, state_dim, action_dim, max_action, nhid):
super().__init__()
self.l1 = nn.Linear(state_dim, nhid)
self.l2 = nn.Linear(nhid, nhid)
self.l3 = nn.Linear(nhid, action_dim)
self.max_action = max_action
def forward(self, state):
a = F.relu(self.l1(state))
a = F.relu(self.l2(a))
return self.max_action * torch.tanh(self.l3(a))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4, 4]
|
DeepContinuor | # 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_7/inductor_cache/qj/cqjsidtha4xc6oe2z5r3nxm2bxnc6t7i5uthrvcme2g4pfq66x4w.py
# Topologically Sorted Source Nodes: [add, x], Original ATen: [aten.add, aten.relu]
# Source node to ATen node mapping:
# add => add
# x => relu
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_3), kwargs = {})
# %relu : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%add,), kwargs = {})
triton_poi_fused_add_relu_0 = async_compile.triton('triton_poi_fused_add_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: '*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_relu_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_relu_0(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_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/hy/chykub5chypkkwpcun3kkoriqs32lvjj2vg5ehmdbfv3upkc6htn.py
# Topologically Sorted Source Nodes: [add_1, x_1], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# add_1 => add_1
# x_1 => relu_1
# Graph fragment:
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %relu), kwargs = {})
# %relu_1 : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%add_1,), kwargs = {})
# %le_2 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_add_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=[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_relu_threshold_backward_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_relu_threshold_backward_1(in_out_ptr0, 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
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = 0.0
tmp8 = tmp3 <= tmp7
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/bo/cboxyeuaw2yuwsi37eq7n7nfiyfjkj5gdtv2va5xkun6rhovazq7.py
# Topologically Sorted Source Nodes: [add_2, x_2], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# add_2 => add_2
# x_2 => relu_2
# Graph fragment:
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_5, %relu_1), kwargs = {})
# %relu_2 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_2,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_2, 0), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_1, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_2 = async_compile.triton('triton_poi_fused_add_relu_threshold_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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*i1', 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_relu_threshold_backward_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_relu_threshold_backward_2(in_out_ptr0, 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_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (x2), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = 0.0
tmp8 = tmp6 <= tmp7
tmp9 = tmp3 <= tmp7
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
tl.store(out_ptr0 + (x2), tmp8, xmask)
tl.store(out_ptr1 + (x2), 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, primals_7, primals_8, primals_9 = 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, ))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (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: [add, x], Original ATen: [aten.add, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_add_relu_0.run(buf1, primals_2, primals_3, 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
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [add_1, x_1], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_1.run(buf3, primals_5, buf1, buf9, 256, grid=grid(256), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf4 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [add_2, x_2], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_2.run(buf5, primals_7, buf3, buf7, buf8, 256, grid=grid(256), stream=stream0)
del primals_7
buf6 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_9, reinterpret_tensor(buf5, (64, 4), (4, 1), 0), reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf6)
del primals_9
return (reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(buf5, (64, 4), (4, 1), 0), primals_8, buf7, primals_6, buf8, primals_4, 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, 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)
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)
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
import torch.nn.functional as F
class DeepContinuor(nn.Module):
def __init__(self, x_dim, h_dim, y_dim):
super().__init__()
self.layer1 = nn.Linear(x_dim, h_dim)
self.layer2 = nn.Linear(h_dim, h_dim)
self.layer3 = nn.Linear(h_dim, h_dim)
self.layer4 = nn.Linear(h_dim, y_dim)
def forward(self, x):
x = F.relu(self.layer1(x) + x)
x = F.relu(self.layer2(x) + x)
x = F.relu(self.layer3(x) + x)
x = self.layer4(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'x_dim': 4, 'h_dim': 4, 'y_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
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_relu_0(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_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tl.store(in_out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_relu_threshold_backward_1(in_out_ptr0, 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
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = 0.0
tmp8 = tmp3 <= tmp7
tl.store(in_out_ptr0 + x2, tmp6, xmask)
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_relu_threshold_backward_2(in_out_ptr0, 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_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = 0.0
tmp8 = tmp6 <= tmp7
tmp9 = tmp3 <= tmp7
tl.store(in_out_ptr0 + x2, tmp6, xmask)
tl.store(out_ptr0 + x2, tmp8, xmask)
tl.store(out_ptr1 + x2, tmp9, 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, 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,))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (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_add_relu_0[grid(256)](buf1, primals_2, primals_3,
256, XBLOCK=256, 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
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_relu_threshold_backward_1[grid(256)](buf3,
primals_5, buf1, buf9, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf4
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_relu_threshold_backward_2[grid(256)](buf5,
primals_7, buf3, buf7, buf8, 256, XBLOCK=256, num_warps=4,
num_stages=1)
del primals_7
buf6 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_9, reinterpret_tensor(buf5, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_8, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf6)
del primals_9
return reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(
buf3, (64, 4), (4, 1), 0), reinterpret_tensor(buf5, (64, 4), (4, 1), 0
), primals_8, buf7, primals_6, buf8, primals_4, buf9
class DeepContinuorNew(nn.Module):
def __init__(self, x_dim, h_dim, y_dim):
super().__init__()
self.layer1 = nn.Linear(x_dim, h_dim)
self.layer2 = nn.Linear(h_dim, h_dim)
self.layer3 = nn.Linear(h_dim, h_dim)
self.layer4 = nn.Linear(h_dim, y_dim)
def forward(self, input_0):
primals_1 = self.layer1.weight
primals_2 = self.layer1.bias
primals_4 = self.layer2.weight
primals_5 = self.layer2.bias
primals_6 = self.layer3.weight
primals_7 = self.layer3.bias
primals_8 = self.layer4.weight
primals_9 = self.layer4.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]
| simonverret/deep_continuation | DeepContinuor | false | 4,346 | [
"MIT"
] | 0 | 986bfba7f6806dc4869a023ff1fc1d0d18324b25 | https://github.com/simonverret/deep_continuation/tree/986bfba7f6806dc4869a023ff1fc1d0d18324b25 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, x_dim, h_dim, y_dim):
super().__init__()
self.layer1 = nn.Linear(x_dim, h_dim)
self.layer2 = nn.Linear(h_dim, h_dim)
self.layer3 = nn.Linear(h_dim, h_dim)
self.layer4 = nn.Linear(h_dim, y_dim)
def forward(self, x):
x = F.relu(self.layer1(x) + x)
x = F.relu(self.layer2(x) + x)
x = F.relu(self.layer3(x) + x)
x = self.layer4(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
Normalizer | # 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_7/inductor_cache/y4/cy45ytne4h6d5ry2wgij2ncscyqgbquxaann7s7us7wnizufhipz.py
# Topologically Sorted Source Nodes: [out, abs_1, sum_1], Original ATen: [aten.softplus, aten.abs, aten.sum]
# Source node to ATen node mapping:
# abs_1 => abs_1
# out => div, exp, gt, log1p, mul, where
# sum_1 => sum_1
# Graph fragment:
# %mul : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 1.0), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%mul, 20.0), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%mul,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%log1p, 1.0), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %arg0_1, %div), kwargs = {})
# %abs_1 : [num_users=1] = call_function[target=torch.ops.aten.abs.default](args = (%where,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%abs_1, [-1], True), kwargs = {})
triton_poi_fused_abs_softplus_sum_0 = async_compile.triton('triton_poi_fused_abs_softplus_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=[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_abs_softplus_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_abs_softplus_sum_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
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp28 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp3 = 20.0
tmp4 = tmp2 > tmp3
tmp5 = tl_math.exp(tmp2)
tmp6 = libdevice.log1p(tmp5)
tmp7 = tmp6 * tmp1
tmp8 = tl.where(tmp4, tmp0, tmp7)
tmp9 = tl_math.abs(tmp8)
tmp11 = tmp10 * tmp1
tmp12 = tmp11 > tmp3
tmp13 = tl_math.exp(tmp11)
tmp14 = libdevice.log1p(tmp13)
tmp15 = tmp14 * tmp1
tmp16 = tl.where(tmp12, tmp10, tmp15)
tmp17 = tl_math.abs(tmp16)
tmp18 = tmp9 + tmp17
tmp20 = tmp19 * tmp1
tmp21 = tmp20 > tmp3
tmp22 = tl_math.exp(tmp20)
tmp23 = libdevice.log1p(tmp22)
tmp24 = tmp23 * tmp1
tmp25 = tl.where(tmp21, tmp19, tmp24)
tmp26 = tl_math.abs(tmp25)
tmp27 = tmp18 + tmp26
tmp29 = tmp28 * tmp1
tmp30 = tmp29 > tmp3
tmp31 = tl_math.exp(tmp29)
tmp32 = libdevice.log1p(tmp31)
tmp33 = tmp32 * tmp1
tmp34 = tl.where(tmp30, tmp28, tmp33)
tmp35 = tl_math.abs(tmp34)
tmp36 = tmp27 + tmp35
tl.store(out_ptr0 + (x0), tmp36, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/fj/cfjxxzvthcyqhvahzmjzfcdrluxcvds52hxovgoxnu2lcjwr5gpk.py
# Topologically Sorted Source Nodes: [out, truediv], Original ATen: [aten.softplus, aten.div]
# Source node to ATen node mapping:
# out => div, exp, gt, log1p, mul, where
# truediv => div_1
# Graph fragment:
# %mul : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, 1.0), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%mul, 20.0), kwargs = {})
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%mul,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%log1p, 1.0), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %arg0_1, %div), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%where, %sum_1), kwargs = {})
triton_poi_fused_div_softplus_1 = async_compile.triton('triton_poi_fused_div_softplus_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: 'i32'}, 'device': 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_softplus_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_div_softplus_1(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
x2 = xindex
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp9 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp3 = 20.0
tmp4 = tmp2 > tmp3
tmp5 = tl_math.exp(tmp2)
tmp6 = libdevice.log1p(tmp5)
tmp7 = tmp6 * tmp1
tmp8 = tl.where(tmp4, tmp0, tmp7)
tmp10 = tmp8 / tmp9
tl.store(out_ptr0 + (x2), tmp10, 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, 1), (16, 4, 1, 64), torch.float32)
# Topologically Sorted Source Nodes: [out, abs_1, sum_1], Original ATen: [aten.softplus, aten.abs, aten.sum]
stream0 = get_raw_stream(0)
triton_poi_fused_abs_softplus_sum_0.run(arg0_1, buf0, 64, grid=grid(64), stream=stream0)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [out, truediv], Original ATen: [aten.softplus, aten.div]
triton_poi_fused_div_softplus_1.run(arg0_1, buf0, buf1, 256, grid=grid(256), stream=stream0)
del arg0_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, 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 Normalizer(nn.Module):
def __init__(self, dim=-1, norm=1.0):
super().__init__()
self.dim = dim
self.norm = norm
self.softplus = nn.Softplus()
def forward(self, x):
out = self.softplus(x)
return out / torch.abs(out.detach()).sum(dim=self.dim, keepdims=True)
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, 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_abs_softplus_sum_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
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp19 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp28 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp3 = 20.0
tmp4 = tmp2 > tmp3
tmp5 = tl_math.exp(tmp2)
tmp6 = libdevice.log1p(tmp5)
tmp7 = tmp6 * tmp1
tmp8 = tl.where(tmp4, tmp0, tmp7)
tmp9 = tl_math.abs(tmp8)
tmp11 = tmp10 * tmp1
tmp12 = tmp11 > tmp3
tmp13 = tl_math.exp(tmp11)
tmp14 = libdevice.log1p(tmp13)
tmp15 = tmp14 * tmp1
tmp16 = tl.where(tmp12, tmp10, tmp15)
tmp17 = tl_math.abs(tmp16)
tmp18 = tmp9 + tmp17
tmp20 = tmp19 * tmp1
tmp21 = tmp20 > tmp3
tmp22 = tl_math.exp(tmp20)
tmp23 = libdevice.log1p(tmp22)
tmp24 = tmp23 * tmp1
tmp25 = tl.where(tmp21, tmp19, tmp24)
tmp26 = tl_math.abs(tmp25)
tmp27 = tmp18 + tmp26
tmp29 = tmp28 * tmp1
tmp30 = tmp29 > tmp3
tmp31 = tl_math.exp(tmp29)
tmp32 = libdevice.log1p(tmp31)
tmp33 = tmp32 * tmp1
tmp34 = tl.where(tmp30, tmp28, tmp33)
tmp35 = tl_math.abs(tmp34)
tmp36 = tmp27 + tmp35
tl.store(out_ptr0 + x0, tmp36, xmask)
@triton.jit
def triton_poi_fused_div_softplus_1(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
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp9 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tmp3 = 20.0
tmp4 = tmp2 > tmp3
tmp5 = tl_math.exp(tmp2)
tmp6 = libdevice.log1p(tmp5)
tmp7 = tmp6 * tmp1
tmp8 = tl.where(tmp4, tmp0, tmp7)
tmp10 = tmp8 / tmp9
tl.store(out_ptr0 + x2, tmp10, 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, 1), (16, 4, 1, 64), torch.float32)
get_raw_stream(0)
triton_poi_fused_abs_softplus_sum_0[grid(64)](arg0_1, buf0, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_div_softplus_1[grid(256)](arg0_1, buf0, buf1, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
del buf0
return buf1,
class NormalizerNew(nn.Module):
def __init__(self, dim=-1, norm=1.0):
super().__init__()
self.dim = dim
self.norm = norm
self.softplus = nn.Softplus()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| simonverret/deep_continuation | Normalizer | false | 4,347 | [
"MIT"
] | 0 | 986bfba7f6806dc4869a023ff1fc1d0d18324b25 | https://github.com/simonverret/deep_continuation/tree/986bfba7f6806dc4869a023ff1fc1d0d18324b25 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, dim=-1, norm=1.0):
super().__init__()
self.dim = dim
self.norm = norm
self.softplus = nn.Softplus()
def forward(self, x):
out = self.softplus(x)
return out / torch.abs(out.detach()).sum(dim=self.dim, keepdims=True)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
BasicModel_MaxPool_ReLU | # 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_7/inductor_cache/3n/c3nau4mdumoqeackxybyc6pc37ouhkpfmzyxkbbiawq24byar4ds.py
# Topologically Sorted Source Nodes: [relu, sum_1], Original ATen: [aten.relu, aten.sum]
# Source node to ATen node mapping:
# relu => relu
# sum_1 => sum_1
# Graph fragment:
# %relu : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%squeeze,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%relu, [1]), kwargs = {})
triton_poi_fused_relu_sum_0 = async_compile.triton('triton_poi_fused_relu_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_relu_sum_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_relu_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 + (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')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tl.store(out_ptr0 + (x0), tmp6, 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, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [relu, sum_1], Original ATen: [aten.relu, aten.sum]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_sum_0.run(arg0_1, buf0, 4, grid=grid(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, 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 BasicModel_MaxPool_ReLU(nn.Module):
def __init__(self, inplace=False) ->None:
super().__init__()
self.maxpool = nn.MaxPool1d(3)
self.relu = nn.ReLU(inplace=inplace)
def forward(self, x):
return self.relu(self.maxpool(x)).sum(dim=1)
def get_inputs():
return [torch.rand([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_relu_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 + 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')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (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_relu_sum_0[grid(4)](arg0_1, buf0, 4, XBLOCK=4,
num_warps=1, num_stages=1)
del arg0_1
return buf0,
class BasicModel_MaxPool_ReLUNew(nn.Module):
def __init__(self, inplace=False) ->None:
super().__init__()
self.maxpool = nn.MaxPool1d(3)
self.relu = nn.ReLU(inplace=inplace)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| sagnik/captum | BasicModel_MaxPool_ReLU | false | 4,348 | [
"BSD-3-Clause"
] | 0 | d6b663745ee6c01f072a4358233dec381324c283 | https://github.com/sagnik/captum/tree/d6b663745ee6c01f072a4358233dec381324c283 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, inplace=False) ->None:
super().__init__()
self.maxpool = nn.MaxPool1d(3)
self.relu = nn.ReLU(inplace=inplace)
def forward(self, x):
return self.relu(self.maxpool(x)).sum(dim=1)
def get_inputs():
return [torch.rand([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_7/inductor_cache/x2/cx2hdvwyo7m5jvhhvtugzxqvmy6z4nsfhkkjhvgzbbm3cb6dsum2.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %mul_scalar : [num_users=1] = call_function[target=torch.ops.aten.mul.Scalar](args = (%permute_default, 1.0), kwargs = {})
# %clone_default : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/iz/ciztqj6kop3hxov46yrmzprkzfir3eljcic4mkqznz2j5cfeaudr.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %sum_dim_int_list : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_default, [-1], True), kwargs = {})
# %eq_scalar : [num_users=1] = call_function[target=torch.ops.aten.eq.Scalar](args = (%add_tensor, -inf), kwargs = {})
# %logical_not_default : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%eq_scalar,), kwargs = {})
# %any_dim : [num_users=1] = call_function[target=torch.ops.aten.any.dim](args = (%logical_not_default, -1, True), 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=[64],
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + (4*x2), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x2)), 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*x2)), 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*x2)), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float("-inf")
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = (tmp29 != 0)
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = (tmp33 != 0)
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = (tmp38 != 0)
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = (tmp43 != 0)
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + (x2), tmp14, xmask)
tl.store(out_ptr1 + (x2), tmp25, xmask)
tl.store(out_ptr2 + (x2), tmp45, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/x5/cx5uvbfethxuwwkwxf3xaualzhlcwqsz4jxqpbhintggaypzjwqf.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_default, %sum_dim_int_list), kwargs = {})
# %logical_not_default_1 : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%any_dim,), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4, 4, 4], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_self : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%logical_not_default_1, %full_default, %div_tensor), kwargs = {})
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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 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_2', '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_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = (xindex // 4)
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last').to(tl.int1)
tmp2 = tl.load(in_out_ptr0 + (x4), xmask)
tmp3 = tl.load(in_ptr1 + (x5), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + (x3), xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/vv/cvvnhithjvmvhfjufxwwzclfobkrgbyyteg66hp24r675f7elw4c.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %clone_default_2 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default_3,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/6t/c6t5a5ere3lqjiu7zh3uu4oxmpdoujdaqqmeunxqapgzo4m74uav.py
# Topologically Sorted Source Nodes: [attentionScores_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# attentionScores_1 => 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 = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (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 = 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_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_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(buf0, primals_2, buf3, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_0.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: [], Original ATen: []
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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0); del buf1 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(buf5, primals_8, buf6, buf7, buf8, 64, grid=grid(64), stream=stream0)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(buf9, buf8, primals_8, buf6, buf7, 256, grid=grid(256), stream=stream0)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(buf2, primals_7, buf10, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [attentionScores_1], Original ATen: [aten.clone]
triton_poi_fused_clone_4.run(buf11, buf12, 16, 4, grid=grid(16, 4), stream=stream0)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0); del buf11 # reuse
# Topologically Sorted Source Nodes: [linear_3], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4), (4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf13)
del primals_10
return (reinterpret_tensor(buf13, (4, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), buf9, reinterpret_tensor(buf10, (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), reinterpret_tensor(buf12, (16, 4), (4, 1), 0), primals_9, )
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)
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), (16, 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, ), 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])
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 nn
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProduct, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class MultiHeadAttention(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super(MultiHeadAttention, self).__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, hiddenStates, attentionMask):
projQ = self.queriesLinear(hiddenStates)
projK = self.keysLinear(hiddenStates)
projV = self.valuesLinear(hiddenStates)
queries = self.prepareForScores(projQ)
keys = self.prepareForScores(projK)
values = self.prepareForScores(projV)
attentionScores = self.sdp(queries, keys, values, attentionMask)
attentionScores = attentionScores.permute(0, 2, 1, 3).contiguous()
newShape = attentionScores.size()[:-2] + (self.numAttentionHeads *
self.attentionHeadSize,)
attentionScores = attentionScores.view(*newShape)
return self.outputLinear(attentionScores)
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'hiddenSize': 4, 'numAttentionHeads': 4, 'dropoutProb': 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 import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import math
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + 4 * x2, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x2), 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 * x2), 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 * x2), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float('-inf')
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = tmp29 != 0
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = tmp33 != 0
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = tmp38 != 0
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = tmp43 != 0
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + x2, tmp14, xmask)
tl.store(out_ptr1 + x2, tmp25, xmask)
tl.store(out_ptr2 + x2, tmp45, xmask)
@triton.jit
def triton_poi_fused_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 4
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last').to(tl
.int1)
tmp2 = tl.load(in_out_ptr0 + x4, xmask)
tmp3 = tl.load(in_ptr1 + x5, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + x3, xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_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_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) = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (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 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (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_3, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(16, 4)](buf0, primals_2, buf3, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_0[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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0)
del buf1
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
triton_poi_fused_1[grid(64)](buf5, primals_8, buf6, buf7, buf8, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused_2[grid(256)](buf9, buf8, primals_8, buf6, buf7,
256, XBLOCK=128, num_warps=4, num_stages=1)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf7
triton_poi_fused_3[grid(16, 4)](buf2, primals_7, buf10, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf6
triton_poi_fused_clone_4[grid(16, 4)](buf11, buf12, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0)
del buf11
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf13)
del primals_10
return reinterpret_tensor(buf13, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_3, (16, 4), (4, 1), 0
), buf9, reinterpret_tensor(buf10, (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
), reinterpret_tensor(buf12, (16, 4), (4, 1), 0), primals_9
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProduct, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class MultiHeadAttentionNew(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super(MultiHeadAttentionNew, self).__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, input_0, input_1):
primals_1 = self.queriesLinear.weight
primals_2 = self.queriesLinear.bias
primals_4 = self.keysLinear.weight
primals_5 = self.keysLinear.bias
primals_6 = self.valuesLinear.weight
primals_7 = self.valuesLinear.bias
primals_9 = self.outputLinear.weight
primals_10 = self.outputLinear.bias
primals_3 = input_0
primals_8 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9, primals_10])
return output[0]
| simonepreite/QABERT | MultiHeadAttention | false | 4,349 | [
"MIT"
] | 0 | ed3e49f6619f3ff660068291231909693cb8f5d5 | https://github.com/simonepreite/QABERT/tree/ed3e49f6619f3ff660068291231909693cb8f5d5 | import math
import torch
from torch import nn
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super().__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class Model(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super().__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, hiddenStates, attentionMask):
projQ = self.queriesLinear(hiddenStates)
projK = self.keysLinear(hiddenStates)
projV = self.valuesLinear(hiddenStates)
queries = self.prepareForScores(projQ)
keys = self.prepareForScores(projK)
values = self.prepareForScores(projV)
attentionScores = self.sdp(queries, keys, values, attentionMask)
attentionScores = attentionScores.permute(0, 2, 1, 3).contiguous()
newShape = attentionScores.size()[:-2] + (self.numAttentionHeads *
self.attentionHeadSize,)
attentionScores = attentionScores.view(*newShape)
return self.outputLinear(attentionScores)
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4, 0.5]
|
NormLayer | # 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_7/inductor_cache/kr/ckrxqy227ju5oy4cidmyb35esvbq3qjjqbbpiqrdjo6j5hkwzhi4.py
# Topologically Sorted Source Nodes: [sub, truediv], Original ATen: [aten.sub, aten.div]
# Source node to ATen node mapping:
# sub => sub
# truediv => div
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, 4), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub, 4.00000001), kwargs = {})
triton_poi_fused_div_sub_0 = async_compile.triton('triton_poi_fused_div_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: 'i32'}, 'device': 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_sub_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_div_sub_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 = 4.0
tmp2 = tmp0 - tmp1
tmp3 = 0.249999999375
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + (x0), tmp4, 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: [sub, truediv], Original ATen: [aten.sub, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_div_sub_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 NormLayer(nn.Module):
def __init__(self, mean, std, n=None, eps=1e-08) ->None:
super().__init__()
self.mean = mean
self.std = std
self.eps = eps
def forward(self, x):
return (x - self.mean) / (self.std + self.eps)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'mean': 4, 'std': 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
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_sub_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 = 4.0
tmp2 = tmp0 - tmp1
tmp3 = 0.249999999375
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x0, tmp4, 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_div_sub_0[grid(256)](arg0_1, buf0, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class NormLayerNew(nn.Module):
def __init__(self, mean, std, n=None, eps=1e-08) ->None:
super().__init__()
self.mean = mean
self.std = std
self.eps = eps
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| sagnik/captum | NormLayer | false | 4,350 | [
"BSD-3-Clause"
] | 0 | d6b663745ee6c01f072a4358233dec381324c283 | https://github.com/sagnik/captum/tree/d6b663745ee6c01f072a4358233dec381324c283 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, mean, std, n=None, eps=1e-08) ->None:
super().__init__()
self.mean = mean
self.std = std
self.eps = eps
def forward(self, x):
return (x - self.mean) / (self.std + self.eps)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
LinearMaxPoolLinearModel | # 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_7/inductor_cache/l7/cl7et626a3hygrp6632jwvtzl7kamkrqyyuwaki5lrfkro7ftlke.py
# Topologically Sorted Source Nodes: [max_pool1d], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# max_pool1d => _low_memory_max_pool2d_with_offsets, getitem_1
# 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 = (%unsqueeze_1, [1, 4], [1, 4], [0, 0], [1, 1], False), 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_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=[4],
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), 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': 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_0(in_ptr0, out_ptr0, out_ptr1, 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 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (3 + (4*x0)), 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 + (x0), tmp15, xmask)
tl.store(out_ptr1 + (x0), tmp16, 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, 4), (4, 1))
assert_size_stride(primals_3, (1, 1), (1, 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: [linear], Original ATen: [aten.mm]
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf0)
del primals_2
buf1 = empty_strided_cuda((4, 1, 1, 1), (1, 1, 1, 1), torch.int8)
buf2 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
# Topologically Sorted Source Nodes: [max_pool1d], Original ATen: [aten.max_pool2d_with_indices]
stream0 = get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0.run(buf0, buf1, buf2, 4, grid=grid(4), stream=stream0)
buf3 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf2, (4, 1), (1, 0), 0), primals_3, out=buf3)
return (buf3, primals_1, reinterpret_tensor(buf0, (4, 1, 1, 4), (4, 4, 4, 1), 0), buf1, reinterpret_tensor(buf2, (4, 1), (1, 1), 0), 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, 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((1, 1), (1, 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 LinearMaxPoolLinearModel(nn.Module):
def __init__(self) ->None:
super().__init__()
self.lin1 = nn.Linear(4, 4, bias=False)
self.lin1.weight = nn.Parameter(torch.eye(4, 4))
self.pool1 = nn.MaxPool1d(4)
self.lin2 = nn.Linear(1, 1, bias=False)
self.lin2.weight = nn.Parameter(torch.ones(1, 1))
def forward(self, x):
x = x.unsqueeze(1)
return self.lin2(self.pool1(self.lin1(x))[:, 0, :])
def get_inputs():
return [torch.rand([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
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_max_pool2d_with_indices_0(in_ptr0, out_ptr0, out_ptr1,
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 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (3 + 4 * x0), 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 + x0, tmp15, xmask)
tl.store(out_ptr1 + x0, tmp16, 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, 4), (4, 1))
assert_size_stride(primals_3, (1, 1), (1, 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 = empty_strided_cuda((4, 1, 1, 1), (1, 1, 1, 1), torch.int8)
buf2 = empty_strided_cuda((4, 1, 1, 1), (1, 4, 4, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0[grid(4)](buf0, buf1,
buf2, 4, XBLOCK=4, num_warps=1, num_stages=1)
buf3 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (4, 1), (1, 0), 0),
primals_3, out=buf3)
return buf3, primals_1, reinterpret_tensor(buf0, (4, 1, 1, 4), (4, 4, 4,
1), 0), buf1, reinterpret_tensor(buf2, (4, 1), (1, 1), 0), primals_3
class LinearMaxPoolLinearModelNew(nn.Module):
def __init__(self) ->None:
super().__init__()
self.lin1 = nn.Linear(4, 4, bias=False)
self.lin1.weight = nn.Parameter(torch.eye(4, 4))
self.pool1 = nn.MaxPool1d(4)
self.lin2 = nn.Linear(1, 1, bias=False)
self.lin2.weight = nn.Parameter(torch.ones(1, 1))
def forward(self, input_0):
primals_1 = self.lin1.weight
primals_3 = self.lin2.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| sagnik/captum | LinearMaxPoolLinearModel | false | 4,351 | [
"BSD-3-Clause"
] | 0 | d6b663745ee6c01f072a4358233dec381324c283 | https://github.com/sagnik/captum/tree/d6b663745ee6c01f072a4358233dec381324c283 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self) ->None:
super().__init__()
self.lin1 = nn.Linear(4, 4, bias=False)
self.lin1.weight = nn.Parameter(torch.eye(4, 4))
self.pool1 = nn.MaxPool1d(4)
self.lin2 = nn.Linear(1, 1, bias=False)
self.lin2.weight = nn.Parameter(torch.ones(1, 1))
def forward(self, x):
x = x.unsqueeze(1)
return self.lin2(self.pool1(self.lin1(x))[:, 0, :])
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return []
|
BasicLinearReLULinear | # 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_7/inductor_cache/lt/cltavbgc7xs5l5xlmkkbsf3rbmv3siu3f7s4h2uiv4x6gnjhbvbg.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 : [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: '*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_relu_threshold_backward_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_threshold_backward_0(in_out_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 320
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 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp3 = 0.0
tmp4 = tmp2 <= tmp3
tl.store(in_out_ptr0 + (x0), tmp2, xmask)
tl.store(out_ptr0 + (x0), tmp4, 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, (5, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (1, 5), (5, 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: [x], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_2, (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: [x_1], Original ATen: [aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0.run(buf1, buf3, 320, grid=grid(320), stream=stream0)
buf2 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf1, (64, 5), (5, 1), 0), reinterpret_tensor(primals_3, (5, 1), (1, 5), 0), out=buf2)
return (reinterpret_tensor(buf2, (4, 4, 4, 1), (16, 4, 1, 1), 0), reinterpret_tensor(primals_2, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 5), (5, 1), 0), primals_3, 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((4, 4, 4, 4), (64, 16, 4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((1, 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 BasicLinearReLULinear(nn.Module):
def __init__(self, in_features, out_features=5, bias=False):
super().__init__()
self.fc1 = nn.Linear(in_features, out_features, bias=bias)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(out_features, 1, bias=bias)
def forward(self, x):
x = self.fc1(x)
x = self.relu1(x)
x = self.fc2(x)
return x
def get_inputs():
return [torch.rand([4, 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
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, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 320
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 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp3 = 0.0
tmp4 = tmp2 <= tmp3
tl.store(in_out_ptr0 + x0, tmp2, xmask)
tl.store(out_ptr0 + x0, tmp4, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (5, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (1, 5), (5, 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_2, (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, buf3,
320, XBLOCK=256, num_warps=4, num_stages=1)
buf2 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 5), (5, 1), 0),
reinterpret_tensor(primals_3, (5, 1), (1, 5), 0), out=buf2)
return reinterpret_tensor(buf2, (4, 4, 4, 1), (16, 4, 1, 1), 0
), reinterpret_tensor(primals_2, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 5), (5, 1), 0), primals_3, buf3
class BasicLinearReLULinearNew(nn.Module):
def __init__(self, in_features, out_features=5, bias=False):
super().__init__()
self.fc1 = nn.Linear(in_features, out_features, bias=bias)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(out_features, 1, bias=bias)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_3 = self.fc2.weight
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| sagnik/captum | BasicLinearReLULinear | false | 4,352 | [
"BSD-3-Clause"
] | 0 | d6b663745ee6c01f072a4358233dec381324c283 | https://github.com/sagnik/captum/tree/d6b663745ee6c01f072a4358233dec381324c283 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_features, out_features=5, bias=False):
super().__init__()
self.fc1 = nn.Linear(in_features, out_features, bias=bias)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(out_features, 1, bias=bias)
def forward(self, x):
x = self.fc1(x)
x = self.relu1(x)
x = self.fc2(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
ConcatPositionalEncoding | # 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_7/inductor_cache/vh/cvhtw4c7nwvnmcdnw6gcneuy3h7qtp7uh2lwza5ahdwkbcd5w32l.py
# Topologically Sorted Source Nodes: [out, out_1], Original ATen: [aten.cat, aten.native_layer_norm]
# Source node to ATen node mapping:
# out => cat
# out_1 => add, add_1, mul, mul_1, rsqrt, sub, var_mean
# Graph fragment:
# %cat : [num_users=3] = call_function[target=torch.ops.aten.cat.default](args = ([%primals_2, %expand], -1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%cat, [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 = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%cat, %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_3), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_4), kwargs = {})
triton_per_fused_cat_native_layer_norm_0 = async_compile.triton('triton_per_fused_cat_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.persistent_reduction(
size_hints=[64, 256],
reduction_hint=ReductionHint.INNER,
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*fp32', 7: '*fp32', 8: 'i32', 9: 'i32'}, 'device': 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), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_cat_native_layer_norm_0', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': True, '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_cat_native_layer_norm_0(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel):
xnumel = 64
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
x3 = xindex
x0 = xindex % 4
tmp31 = tl.load(in_ptr2 + (r2), None, eviction_policy='evict_last')
tmp33 = tl.load(in_ptr3 + (r2), None, eviction_policy='evict_last')
tmp0 = r2
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 128, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((128*x3) + r2), tmp4, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 256, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((128*x0) + ((-128) + r2)), tmp6, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = tl.broadcast_to(tmp10, [RBLOCK])
tmp13 = tl.broadcast_to(tmp11, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = tl.full([1], 256, tl.int32)
tmp17 = tmp16.to(tl.float32)
tmp18 = tmp15 / tmp17
tmp19 = tmp11 - tmp18
tmp20 = tmp19 * tmp19
tmp21 = tl.broadcast_to(tmp20, [RBLOCK])
tmp23 = triton_helpers.promote_to_tensor(tl.sum(tmp21, 0))
tmp24 = 256.0
tmp25 = tmp23 / tmp24
tmp26 = 1e-05
tmp27 = tmp25 + tmp26
tmp28 = libdevice.rsqrt(tmp27)
tmp29 = tmp10 - tmp18
tmp30 = tmp29 * tmp28
tmp32 = tmp30 * tmp31
tmp34 = tmp32 + tmp33
tl.store(out_ptr0 + (r2 + (256*x3)), tmp10, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + (x3), tmp28, None)
tl.store(out_ptr2 + (r2 + (256*x3)), tmp34, None)
tl.store(out_ptr1 + (x3), tmp18, None)
''', 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, (512, 128), (128, 1))
assert_size_stride(primals_2, (4, 4, 4, 128), (2048, 512, 128, 1))
assert_size_stride(primals_3, (256, ), (1, ))
assert_size_stride(primals_4, (256, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 256), (4096, 1024, 256, 1), torch.float32)
buf1 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
buf2 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = reinterpret_tensor(buf2, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf2 # reuse
buf5 = empty_strided_cuda((4, 4, 4, 256), (4096, 1024, 256, 1), torch.float32)
# Topologically Sorted Source Nodes: [out, out_1], Original ATen: [aten.cat, aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_per_fused_cat_native_layer_norm_0.run(buf4, primals_2, primals_1, primals_3, primals_4, buf0, buf1, buf5, 64, 256, grid=grid(64), stream=stream0)
del primals_1
del primals_2
del primals_4
return (buf5, primals_3, buf0, buf1, 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((512, 128), (128, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 4, 128), (2048, 512, 128, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((256, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((256, ), (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 ConcatPositionalEncoding(nn.Module):
def __init__(self, d_model=256, max_len=512):
super().__init__()
self.timing_table = nn.Parameter(torch.FloatTensor(max_len, d_model //
2))
nn.init.normal_(self.timing_table)
self.norm = nn.LayerNorm(d_model)
def forward(self, x):
timing = self.timing_table[None, :x.shape[1], :]
x, timing = torch.broadcast_tensors(x, timing)
out = torch.cat([x, timing], dim=-1)
out = self.norm(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 128])]
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
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_cat_native_layer_norm_0(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, in_ptr3, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = 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
x3 = xindex
x0 = xindex % 4
tmp31 = tl.load(in_ptr2 + r2, None, eviction_policy='evict_last')
tmp33 = tl.load(in_ptr3 + r2, None, eviction_policy='evict_last')
tmp0 = r2
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 128, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (128 * x3 + r2), tmp4, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 256, tl.int64)
tmp9 = tl.load(in_ptr1 + (128 * x0 + (-128 + r2)), tmp6,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = tl.broadcast_to(tmp10, [RBLOCK])
tmp13 = tl.broadcast_to(tmp11, [RBLOCK])
tmp15 = triton_helpers.promote_to_tensor(tl.sum(tmp13, 0))
tmp16 = tl.full([1], 256, tl.int32)
tmp17 = tmp16.to(tl.float32)
tmp18 = tmp15 / tmp17
tmp19 = tmp11 - tmp18
tmp20 = tmp19 * tmp19
tmp21 = tl.broadcast_to(tmp20, [RBLOCK])
tmp23 = triton_helpers.promote_to_tensor(tl.sum(tmp21, 0))
tmp24 = 256.0
tmp25 = tmp23 / tmp24
tmp26 = 1e-05
tmp27 = tmp25 + tmp26
tmp28 = libdevice.rsqrt(tmp27)
tmp29 = tmp10 - tmp18
tmp30 = tmp29 * tmp28
tmp32 = tmp30 * tmp31
tmp34 = tmp32 + tmp33
tl.store(out_ptr0 + (r2 + 256 * x3), tmp10, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + x3, tmp28, None)
tl.store(out_ptr2 + (r2 + 256 * x3), tmp34, None)
tl.store(out_ptr1 + x3, tmp18, None)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (512, 128), (128, 1))
assert_size_stride(primals_2, (4, 4, 4, 128), (2048, 512, 128, 1))
assert_size_stride(primals_3, (256,), (1,))
assert_size_stride(primals_4, (256,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 256), (4096, 1024, 256, 1),
torch.float32)
buf1 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
buf2 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf4 = reinterpret_tensor(buf2, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf2
buf5 = empty_strided_cuda((4, 4, 4, 256), (4096, 1024, 256, 1),
torch.float32)
get_raw_stream(0)
triton_per_fused_cat_native_layer_norm_0[grid(64)](buf4, primals_2,
primals_1, primals_3, primals_4, buf0, buf1, buf5, 64, 256,
num_warps=2, num_stages=1)
del primals_1
del primals_2
del primals_4
return buf5, primals_3, buf0, buf1, buf4
class ConcatPositionalEncodingNew(nn.Module):
def __init__(self, d_model=256, max_len=512):
super().__init__()
self.timing_table = nn.Parameter(torch.FloatTensor(max_len, d_model //
2))
nn.init.normal_(self.timing_table)
self.norm = nn.LayerNorm(d_model)
def forward(self, input_0):
primals_1 = self.timing_table
primals_3 = self.norm.weight
primals_4 = self.norm.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| skulick/self-attentive-parser | ConcatPositionalEncoding | false | 4,353 | [
"MIT"
] | 0 | 04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | https://github.com/skulick/self-attentive-parser/tree/04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, d_model=256, max_len=512):
super().__init__()
self.timing_table = nn.Parameter(torch.FloatTensor(max_len, d_model //
2))
nn.init.normal_(self.timing_table)
self.norm = nn.LayerNorm(d_model)
def forward(self, x):
timing = self.timing_table[None, :x.shape[1], :]
x, timing = torch.broadcast_tensors(x, timing)
out = torch.cat([x, timing], dim=-1)
out = self.norm(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 128])]
def get_init_inputs():
return []
|
PartitionedReLU | # 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_7/inductor_cache/zl/czl4nbejjit37i5xq22msnfyunxozv2ffdattlqux3xysiccvkfk.py
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# relu => relu
# Graph fragment:
# %relu : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%getitem,), 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=[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_relu_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_relu_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
x0 = xindex % 2
x1 = (xindex // 2)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (4*x1)), xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/7g/c7gxev54gbtyo3gsvtqsprkutk4roe44pc2njqdk6r76n2rg53ct.py
# Topologically Sorted Source Nodes: [relu_1], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# relu_1 => relu_1
# Graph fragment:
# %relu_1 : [num_users=1] = call_function[target=torch.ops.aten.relu.default](args = (%getitem_1,), kwargs = {})
triton_poi_fused_relu_1 = async_compile.triton('triton_poi_fused_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=[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_relu_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_relu_1(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
x0 = xindex % 2
x1 = (xindex // 2)
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 + x0 + (4*x1)), xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(out_ptr0 + (x2), tmp2, 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, 2), (32, 8, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(arg0_1, buf0, 128, grid=grid(128), stream=stream0)
buf1 = empty_strided_cuda((4, 4, 4, 2), (32, 8, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [relu_1], Original ATen: [aten.relu]
triton_poi_fused_relu_1.run(arg0_1, buf1, 128, grid=grid(128), stream=stream0)
del arg0_1
return (buf0, 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 PartitionedReLU(nn.ReLU):
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
return super().forward(x_c), super().forward(x_p)
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_relu_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
x0 = xindex % 2
x1 = xindex // 2
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 4 * x1), xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused_relu_1(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
x0 = xindex % 2
x1 = xindex // 2
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 + x0 + 4 * x1), xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tl.store(out_ptr0 + x2, tmp2, 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, 2), (32, 8, 2, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_relu_0[grid(128)](arg0_1, buf0, 128, XBLOCK=128,
num_warps=4, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 4, 2), (32, 8, 2, 1), torch.float32)
triton_poi_fused_relu_1[grid(128)](arg0_1, buf1, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
return buf0, buf1
class PartitionedReLUNew(nn.ReLU):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0], output[1]
| skulick/self-attentive-parser | PartitionedReLU | false | 4,354 | [
"MIT"
] | 0 | 04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | https://github.com/skulick/self-attentive-parser/tree/04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | import torch
import torch.nn as nn
class Model(nn.ReLU):
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
return super().forward(x_c), super().forward(x_p)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
LogLoss | # 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_7/inductor_cache/p5/cp5ubgxix66mcwdxvtf7bvbxdvoimtka63njezhp3tfusdmuegld.py
# Topologically Sorted Source Nodes: [inp, tgt, mse_loss, loss], Original ATen: [aten.atan, aten.mse_loss, aten.sqrt]
# Source node to ATen node mapping:
# inp => atan_1
# loss => sqrt
# mse_loss => mean, pow_1, sub
# tgt => atan
# Graph fragment:
# %atan_1 : [num_users=1] = call_function[target=torch.ops.aten.atan.default](args = (%arg1_1,), kwargs = {})
# %atan : [num_users=1] = call_function[target=torch.ops.aten.atan.default](args = (%arg0_1,), kwargs = {})
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%atan_1, %atan), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%pow_1,), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%mean,), kwargs = {})
triton_per_fused_atan_mse_loss_sqrt_0 = async_compile.triton('triton_per_fused_atan_mse_loss_sqrt_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_atan_mse_loss_sqrt_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_atan_mse_loss_sqrt_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)
tmp2 = tl.load(in_ptr1 + (r0), None)
tmp1 = libdevice.atan(tmp0)
tmp3 = libdevice.atan(tmp2)
tmp4 = tmp1 - tmp3
tmp5 = tmp4 * tmp4
tmp6 = tl.broadcast_to(tmp5, [RBLOCK])
tmp8 = triton_helpers.promote_to_tensor(tl.sum(tmp6, 0))
tmp9 = 256.0
tmp10 = tmp8 / tmp9
tmp11 = libdevice.sqrt(tmp10)
tl.debug_barrier()
tl.store(in_out_ptr0 + (tl.full([1], 0, tl.int32)), tmp11, 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: [inp, tgt, mse_loss, loss], Original ATen: [aten.atan, aten.mse_loss, aten.sqrt]
stream0 = get_raw_stream(0)
triton_per_fused_atan_mse_loss_sqrt_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
from torch.nn import MSELoss
class LogLoss(MSELoss):
def __init__(self):
super(LogLoss, self).__init__()
self.loss = torch.nn.MSELoss()
self.loss2 = torch.nn.MSELoss()
def forward(self, input, target):
tgt = torch.atan(target)
inp = torch.atan(input)
loss = torch.sqrt(self.loss(inp, tgt))
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
from torch.nn import MSELoss
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_atan_mse_loss_sqrt_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)
tmp2 = tl.load(in_ptr1 + r0, None)
tmp1 = libdevice.atan(tmp0)
tmp3 = libdevice.atan(tmp2)
tmp4 = tmp1 - tmp3
tmp5 = tmp4 * tmp4
tmp6 = tl.broadcast_to(tmp5, [RBLOCK])
tmp8 = triton_helpers.promote_to_tensor(tl.sum(tmp6, 0))
tmp9 = 256.0
tmp10 = tmp8 / tmp9
tmp11 = libdevice.sqrt(tmp10)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp11, 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_atan_mse_loss_sqrt_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 LogLossNew(MSELoss):
def __init__(self):
super(LogLossNew, self).__init__()
self.loss = torch.nn.MSELoss()
self.loss2 = torch.nn.MSELoss()
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| slaveuser/testRepo20181123 | LogLoss | false | 4,355 | [
"MIT"
] | 0 | 0651de19b3b7d02f8c9094b8b24346ccc2e30480 | https://github.com/slaveuser/testRepo20181123/tree/0651de19b3b7d02f8c9094b8b24346ccc2e30480 | import torch
from torch.nn import MSELoss
class Model(MSELoss):
def __init__(self):
super().__init__()
self.loss = torch.nn.MSELoss()
self.loss2 = torch.nn.MSELoss()
def forward(self, input, target):
tgt = torch.atan(target)
inp = torch.atan(input)
loss = torch.sqrt(self.loss(inp, tgt))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GlobalLayerNorm | # 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_7/inductor_cache/ap/cap3nkzqkfj7rlygvuozblog2seuhvlea6gxfoq5gqtaswv2kdk3.py
# Topologically Sorted Source Nodes: [mean, mean_2, sub, pow_1, mean_3, var], Original ATen: [aten.mean, aten.sub, aten.pow]
# Source node to ATen node mapping:
# mean => mean
# mean_2 => mean_1
# mean_3 => mean_2
# pow_1 => pow_1
# sub => sub
# var => mean_3
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%primals_1, [1], True), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%mean, [2], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %mean_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean_2 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [1], True), kwargs = {})
# %mean_3 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%mean_2, [2], True), kwargs = {})
triton_poi_fused_mean_pow_sub_0 = async_compile.triton('triton_poi_fused_mean_pow_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=[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_mean_pow_sub_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_mean_pow_sub_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 % 4
x1 = (xindex // 4)
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)
tmp9 = tl.load(in_ptr0 + (4 + x0 + (64*x1)), xmask)
tmp10 = tl.load(in_ptr0 + (20 + x0 + (64*x1)), xmask)
tmp12 = tl.load(in_ptr0 + (36 + x0 + (64*x1)), xmask)
tmp14 = tl.load(in_ptr0 + (52 + x0 + (64*x1)), xmask)
tmp18 = tl.load(in_ptr0 + (8 + x0 + (64*x1)), xmask)
tmp19 = tl.load(in_ptr0 + (24 + x0 + (64*x1)), xmask)
tmp21 = tl.load(in_ptr0 + (40 + x0 + (64*x1)), xmask)
tmp23 = tl.load(in_ptr0 + (56 + x0 + (64*x1)), xmask)
tmp27 = tl.load(in_ptr0 + (12 + x0 + (64*x1)), xmask)
tmp28 = tl.load(in_ptr0 + (28 + x0 + (64*x1)), xmask)
tmp30 = tl.load(in_ptr0 + (44 + x0 + (64*x1)), xmask)
tmp32 = tl.load(in_ptr0 + (60 + x0 + (64*x1)), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp11 = tmp9 + tmp10
tmp13 = tmp11 + tmp12
tmp15 = tmp13 + tmp14
tmp16 = tmp15 / tmp7
tmp17 = tmp8 + tmp16
tmp20 = tmp18 + tmp19
tmp22 = tmp20 + tmp21
tmp24 = tmp22 + tmp23
tmp25 = tmp24 / tmp7
tmp26 = tmp17 + tmp25
tmp29 = tmp27 + tmp28
tmp31 = tmp29 + tmp30
tmp33 = tmp31 + tmp32
tmp34 = tmp33 / tmp7
tmp35 = tmp26 + tmp34
tmp36 = tmp35 / tmp7
tmp37 = tmp0 - tmp36
tmp38 = tmp37 * tmp37
tmp39 = tmp1 - tmp36
tmp40 = tmp39 * tmp39
tmp41 = tmp38 + tmp40
tmp42 = tmp3 - tmp36
tmp43 = tmp42 * tmp42
tmp44 = tmp41 + tmp43
tmp45 = tmp5 - tmp36
tmp46 = tmp45 * tmp45
tmp47 = tmp44 + tmp46
tmp48 = tmp47 / tmp7
tmp49 = tmp9 - tmp36
tmp50 = tmp49 * tmp49
tmp51 = tmp10 - tmp36
tmp52 = tmp51 * tmp51
tmp53 = tmp50 + tmp52
tmp54 = tmp12 - tmp36
tmp55 = tmp54 * tmp54
tmp56 = tmp53 + tmp55
tmp57 = tmp14 - tmp36
tmp58 = tmp57 * tmp57
tmp59 = tmp56 + tmp58
tmp60 = tmp59 / tmp7
tmp61 = tmp48 + tmp60
tmp62 = tmp18 - tmp36
tmp63 = tmp62 * tmp62
tmp64 = tmp19 - tmp36
tmp65 = tmp64 * tmp64
tmp66 = tmp63 + tmp65
tmp67 = tmp21 - tmp36
tmp68 = tmp67 * tmp67
tmp69 = tmp66 + tmp68
tmp70 = tmp23 - tmp36
tmp71 = tmp70 * tmp70
tmp72 = tmp69 + tmp71
tmp73 = tmp72 / tmp7
tmp74 = tmp61 + tmp73
tmp75 = tmp27 - tmp36
tmp76 = tmp75 * tmp75
tmp77 = tmp28 - tmp36
tmp78 = tmp77 * tmp77
tmp79 = tmp76 + tmp78
tmp80 = tmp30 - tmp36
tmp81 = tmp80 * tmp80
tmp82 = tmp79 + tmp81
tmp83 = tmp32 - tmp36
tmp84 = tmp83 * tmp83
tmp85 = tmp82 + tmp84
tmp86 = tmp85 / tmp7
tmp87 = tmp74 + tmp86
tmp88 = tmp87 / tmp7
tl.store(out_ptr0 + (x2), tmp36, xmask)
tl.store(out_ptr1 + (x2), tmp88, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/tu/ctunck2watc7guogb7nwwbhl2evcmo3mr4fhhl6ajujqj74x6txp.py
# Topologically Sorted Source Nodes: [sub, mul, add, pow_2, truediv, gLN_y], Original ATen: [aten.sub, aten.mul, aten.add, aten.pow, aten.div]
# Source node to ATen node mapping:
# add => add
# gLN_y => add_1
# mul => mul
# pow_2 => pow_2
# sub => sub
# truediv => div
# Graph fragment:
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%primals_1, %mean_1), 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 = (%mean_3, 1e-08), kwargs = {})
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%add, 0.5), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul, %pow_2), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, %primals_3), kwargs = {})
triton_poi_fused_add_div_mul_pow_sub_1 = async_compile.triton('triton_poi_fused_add_div_mul_pow_sub_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: 'i32'}, 'device': 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_add_div_mul_pow_sub_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_add_div_mul_pow_sub_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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) % 4
x4 = xindex
x0 = xindex % 4
x3 = (xindex // 64)
tmp0 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x4), xmask)
tmp2 = tl.load(in_ptr2 + (x0 + (4*x3)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x0 + (4*x3)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 - tmp2
tmp4 = tmp0 * tmp3
tmp6 = 1e-08
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp4 / tmp8
tmp11 = tmp9 + tmp10
tl.store(out_ptr0 + (x4), tmp11, 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, (1, 4, 1), (4, 1, 1))
assert_size_stride(primals_3, (1, 4, 1), (4, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 1, 4), (4, 16, 16, 1), torch.float32)
buf1 = empty_strided_cuda((4, 1, 1, 4), (4, 16, 16, 1), torch.float32)
# Topologically Sorted Source Nodes: [mean, mean_2, sub, pow_1, mean_3, var], Original ATen: [aten.mean, aten.sub, aten.pow]
stream0 = get_raw_stream(0)
triton_poi_fused_mean_pow_sub_0.run(primals_1, buf0, buf1, 16, grid=grid(16), stream=stream0)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [sub, mul, add, pow_2, truediv, gLN_y], Original ATen: [aten.sub, aten.mul, aten.add, aten.pow, aten.div]
triton_poi_fused_add_div_mul_pow_sub_1.run(primals_2, primals_1, buf0, buf1, primals_3, buf2, 256, grid=grid(256), stream=stream0)
del buf0
del buf1
del primals_2
del primals_3
return (buf2, 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), (4, 1, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((1, 4, 1), (4, 1, 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 itertools import product as product
class GlobalLayerNorm(nn.Module):
def __init__(self, channel_size):
super(GlobalLayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.beta = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.reset_parameters()
def reset_parameters(self):
self.gamma.data.fill_(1)
self.beta.data.zero_()
def forward(self, y):
mean = y.mean(dim=1, keepdim=True).mean(dim=2, keepdim=True)
var = torch.pow(y - mean, 2).mean(dim=1, keepdim=True).mean(dim=2,
keepdim=True)
gLN_y = self.gamma * (y - mean) / torch.pow(var + 1e-08, 0.5
) + self.beta
return gLN_y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'channel_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.triton_helpers import libdevice
import torch.nn as nn
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_mean_pow_sub_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 % 4
x1 = xindex // 4
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)
tmp9 = tl.load(in_ptr0 + (4 + x0 + 64 * x1), xmask)
tmp10 = tl.load(in_ptr0 + (20 + x0 + 64 * x1), xmask)
tmp12 = tl.load(in_ptr0 + (36 + x0 + 64 * x1), xmask)
tmp14 = tl.load(in_ptr0 + (52 + x0 + 64 * x1), xmask)
tmp18 = tl.load(in_ptr0 + (8 + x0 + 64 * x1), xmask)
tmp19 = tl.load(in_ptr0 + (24 + x0 + 64 * x1), xmask)
tmp21 = tl.load(in_ptr0 + (40 + x0 + 64 * x1), xmask)
tmp23 = tl.load(in_ptr0 + (56 + x0 + 64 * x1), xmask)
tmp27 = tl.load(in_ptr0 + (12 + x0 + 64 * x1), xmask)
tmp28 = tl.load(in_ptr0 + (28 + x0 + 64 * x1), xmask)
tmp30 = tl.load(in_ptr0 + (44 + x0 + 64 * x1), xmask)
tmp32 = tl.load(in_ptr0 + (60 + x0 + 64 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp11 = tmp9 + tmp10
tmp13 = tmp11 + tmp12
tmp15 = tmp13 + tmp14
tmp16 = tmp15 / tmp7
tmp17 = tmp8 + tmp16
tmp20 = tmp18 + tmp19
tmp22 = tmp20 + tmp21
tmp24 = tmp22 + tmp23
tmp25 = tmp24 / tmp7
tmp26 = tmp17 + tmp25
tmp29 = tmp27 + tmp28
tmp31 = tmp29 + tmp30
tmp33 = tmp31 + tmp32
tmp34 = tmp33 / tmp7
tmp35 = tmp26 + tmp34
tmp36 = tmp35 / tmp7
tmp37 = tmp0 - tmp36
tmp38 = tmp37 * tmp37
tmp39 = tmp1 - tmp36
tmp40 = tmp39 * tmp39
tmp41 = tmp38 + tmp40
tmp42 = tmp3 - tmp36
tmp43 = tmp42 * tmp42
tmp44 = tmp41 + tmp43
tmp45 = tmp5 - tmp36
tmp46 = tmp45 * tmp45
tmp47 = tmp44 + tmp46
tmp48 = tmp47 / tmp7
tmp49 = tmp9 - tmp36
tmp50 = tmp49 * tmp49
tmp51 = tmp10 - tmp36
tmp52 = tmp51 * tmp51
tmp53 = tmp50 + tmp52
tmp54 = tmp12 - tmp36
tmp55 = tmp54 * tmp54
tmp56 = tmp53 + tmp55
tmp57 = tmp14 - tmp36
tmp58 = tmp57 * tmp57
tmp59 = tmp56 + tmp58
tmp60 = tmp59 / tmp7
tmp61 = tmp48 + tmp60
tmp62 = tmp18 - tmp36
tmp63 = tmp62 * tmp62
tmp64 = tmp19 - tmp36
tmp65 = tmp64 * tmp64
tmp66 = tmp63 + tmp65
tmp67 = tmp21 - tmp36
tmp68 = tmp67 * tmp67
tmp69 = tmp66 + tmp68
tmp70 = tmp23 - tmp36
tmp71 = tmp70 * tmp70
tmp72 = tmp69 + tmp71
tmp73 = tmp72 / tmp7
tmp74 = tmp61 + tmp73
tmp75 = tmp27 - tmp36
tmp76 = tmp75 * tmp75
tmp77 = tmp28 - tmp36
tmp78 = tmp77 * tmp77
tmp79 = tmp76 + tmp78
tmp80 = tmp30 - tmp36
tmp81 = tmp80 * tmp80
tmp82 = tmp79 + tmp81
tmp83 = tmp32 - tmp36
tmp84 = tmp83 * tmp83
tmp85 = tmp82 + tmp84
tmp86 = tmp85 / tmp7
tmp87 = tmp74 + tmp86
tmp88 = tmp87 / tmp7
tl.store(out_ptr0 + x2, tmp36, xmask)
tl.store(out_ptr1 + x2, tmp88, xmask)
@triton.jit
def triton_poi_fused_add_div_mul_pow_sub_1(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, 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 % 4
x4 = xindex
x0 = xindex % 4
x3 = xindex // 64
tmp0 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, xmask)
tmp2 = tl.load(in_ptr2 + (x0 + 4 * x3), xmask, eviction_policy='evict_last'
)
tmp5 = tl.load(in_ptr3 + (x0 + 4 * x3), xmask, eviction_policy='evict_last'
)
tmp10 = tl.load(in_ptr4 + x1, xmask, eviction_policy='evict_last')
tmp3 = tmp1 - tmp2
tmp4 = tmp0 * tmp3
tmp6 = 1e-08
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp4 / tmp8
tmp11 = tmp9 + tmp10
tl.store(out_ptr0 + x4, tmp11, 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, (1, 4, 1), (4, 1, 1))
assert_size_stride(primals_3, (1, 4, 1), (4, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 1, 4), (4, 16, 16, 1), torch.float32)
buf1 = empty_strided_cuda((4, 1, 1, 4), (4, 16, 16, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mean_pow_sub_0[grid(16)](primals_1, buf0, buf1, 16,
XBLOCK=16, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_div_mul_pow_sub_1[grid(256)](primals_2,
primals_1, buf0, buf1, primals_3, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf0
del buf1
del primals_2
del primals_3
return buf2, primals_1
class GlobalLayerNormNew(nn.Module):
def __init__(self, channel_size):
super(GlobalLayerNormNew, self).__init__()
self.gamma = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.beta = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.reset_parameters()
def reset_parameters(self):
self.gamma.data.fill_(1)
self.beta.data.zero_()
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]
| slapshin/TalkNet_ASD | GlobalLayerNorm | false | 4,356 | [
"MIT"
] | 0 | 343fac5c8d2bef2b98244e3acf20ac322711a4c7 | https://github.com/slapshin/TalkNet_ASD/tree/343fac5c8d2bef2b98244e3acf20ac322711a4c7 | import torch
import torch.nn as nn
from itertools import product as product
class Model(nn.Module):
def __init__(self, channel_size):
super().__init__()
self.gamma = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.beta = nn.Parameter(torch.Tensor(1, channel_size, 1))
self.reset_parameters()
def reset_parameters(self):
self.gamma.data.fill_(1)
self.beta.data.zero_()
def forward(self, y):
mean = y.mean(dim=1, keepdim=True).mean(dim=2, keepdim=True)
var = torch.pow(y - mean, 2).mean(dim=1, keepdim=True).mean(dim=2,
keepdim=True)
gLN_y = self.gamma * (y - mean) / torch.pow(var + 1e-08, 0.5
) + self.beta
return gLN_y
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4]
|
PartitionedLinear | # 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_7/inductor_cache/ke/ckelmqewlwwywmknorjlb5bpvskr4igxxhmicz4kwdabkq45jln5.py
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.add]
# Source node to ATen node mapping:
# out_c => add
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_3), 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=[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_add_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_add_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 2
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, 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, (2, 2), (2, 1))
assert_size_stride(primals_3, (2, ), (1, ))
assert_size_stride(primals_4, (2, 2), (2, 1))
assert_size_stride(primals_5, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 2), (4, 1), 0), reinterpret_tensor(primals_2, (2, 2), (1, 2), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 2), (32, 8, 2, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_0.run(buf1, primals_3, 128, grid=grid(128), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((64, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 2), (4, 1), 2), reinterpret_tensor(primals_4, (2, 2), (1, 2), 0), out=buf2)
del primals_4
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 2), (32, 8, 2, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.add]
triton_poi_fused_add_0.run(buf3, primals_5, 128, grid=grid(128), stream=stream0)
del primals_5
return (buf1, buf3, reinterpret_tensor(primals_1, (64, 2), (4, 1), 0), reinterpret_tensor(primals_1, (64, 2), (4, 1), 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((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((2, ), (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 PartitionedLinear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
out_c = self.linear_c(x_c)
out_p = self.linear_p(x_p)
return out_c, out_p
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
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, xnumel, XBLOCK: tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 2
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, 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, (2, 2), (2, 1))
assert_size_stride(primals_3, (2,), (1,))
assert_size_stride(primals_4, (2, 2), (2, 1))
assert_size_stride(primals_5, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 2), (2, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 2), (4, 1), 0),
reinterpret_tensor(primals_2, (2, 2), (1, 2), 0), out=buf0)
del primals_2
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 2), (32, 8, 2, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_add_0[grid(128)](buf1, primals_3, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((64, 2), (2, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 2), (4, 1), 2),
reinterpret_tensor(primals_4, (2, 2), (1, 2), 0), out=buf2)
del primals_4
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 2), (32, 8, 2, 1), 0)
del buf2
triton_poi_fused_add_0[grid(128)](buf3, primals_5, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_5
return buf1, buf3, reinterpret_tensor(primals_1, (64, 2), (4, 1), 0
), reinterpret_tensor(primals_1, (64, 2), (4, 1), 2)
class PartitionedLinearNew(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, input_0):
primals_2 = self.linear_c.weight
primals_3 = self.linear_c.bias
primals_4 = self.linear_p.weight
primals_5 = self.linear_p.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0], output[1]
| skulick/self-attentive-parser | PartitionedLinear | false | 4,357 | [
"MIT"
] | 0 | 04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | https://github.com/skulick/self-attentive-parser/tree/04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
out_c = self.linear_c(x_c)
out_p = self.linear_p(x_p)
return out_c, out_p
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
Normalize | # 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_7/inductor_cache/4a/c4acpsyuqt5yil7whrwdv5s5bi5vxhby4smu36dwgbhemfjncaxh.py
# Topologically Sorted Source Nodes: [qn, top], Original ATen: [aten.add, aten.div]
# Source node to ATen node mapping:
# qn => add
# top => div
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%unsqueeze, 1e-12), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%arg0_1, %add), kwargs = {})
triton_poi_fused_add_div_0 = async_compile.triton('triton_poi_fused_add_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_add_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_add_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 = tmp12 + tmp13
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + (x3), tmp15, 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: [qn, top], Original ATen: [aten.add, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_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 Normalize(nn.Module):
def __init__(self):
super(Normalize, self).__init__()
def forward(self, bottom):
qn = torch.norm(bottom, p=2, dim=1).unsqueeze(dim=1) + 1e-12
top = bottom.div(qn)
return top
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_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 = tmp12 + tmp13
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x3, tmp15, 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_0[grid(256)](arg0_1, buf0, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class NormalizeNew(nn.Module):
def __init__(self):
super(NormalizeNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| slyviacassell/Multi-taks-UNITE | Normalize | false | 4,358 | [
"MIT"
] | 0 | a010a92c94c0ee0f1ffed27df6d89da58d6d34c5 | https://github.com/slyviacassell/Multi-taks-UNITE/tree/a010a92c94c0ee0f1ffed27df6d89da58d6d34c5 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, bottom):
qn = torch.norm(bottom, p=2, dim=1).unsqueeze(dim=1) + 1e-12
top = bottom.div(qn)
return top
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GlobalAveragePool2d | # 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_7/inductor_cache/kc/ckcbqmsuvoeea4h7o3qtxmmkrtt55gl5tbfyiomy7yxcy72334ok.py
# Topologically Sorted Source Nodes: [out_1, view_1], Original ATen: [aten.mean, aten.view]
# Source node to ATen node mapping:
# out_1 => mean
# view_1 => view_1
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%view, [-1]), kwargs = {})
# %view_1 : [num_users=1] = call_function[target=torch.ops.aten.reshape.default](args = (%mean, [4, -1]), kwargs = {})
triton_per_fused_mean_view_0 = async_compile.triton('triton_per_fused_mean_view_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_view_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_view_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):
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, 1), torch.float32)
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [out_1, view_1], Original ATen: [aten.mean, aten.view]
stream0 = get_raw_stream(0)
triton_per_fused_mean_view_0.run(buf1, arg0_1, 16, 16, grid=grid(16), stream=stream0)
del arg0_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)
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 GlobalAveragePool2d(nn.Module):
def __init__(self):
super(GlobalAveragePool2d, self).__init__()
def forward(self, x: 'torch.Tensor'):
assert len(x.size()) >= 2
x_size = x.size()
out = x.view(*x_size[:-2], -1)
out = out.mean(dim=-1)
out = out.view(*x_size[:-3], -1).contiguous()
return out
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
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_mean_view_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):
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, 1), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_mean_view_0[grid(16)](buf1, arg0_1, 16, 16, XBLOCK
=1, num_warps=2, num_stages=1)
del arg0_1
return buf1,
class GlobalAveragePool2dNew(nn.Module):
def __init__(self):
super(GlobalAveragePool2dNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| slyviacassell/Multi-taks-UNITE | GlobalAveragePool2d | false | 4,359 | [
"MIT"
] | 0 | a010a92c94c0ee0f1ffed27df6d89da58d6d34c5 | https://github.com/slyviacassell/Multi-taks-UNITE/tree/a010a92c94c0ee0f1ffed27df6d89da58d6d34c5 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self):
super().__init__()
def forward(self, x: 'torch.Tensor'):
assert len(x.size()) >= 2
x_size = x.size()
out = x.view(*x_size[:-2], -1)
out = out.mean(dim=-1)
out = out.view(*x_size[:-3], -1).contiguous()
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
PointwiseConvolutionLayer | # 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_7/inductor_cache/ut/cutws4mb4ie4j75s4j3hnhmsv2lbeocvsbzwgheafxyhexvlpysn.py
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.softplus]
# Source node to ATen node mapping:
# output => exp, gt, log1p, where
# Graph fragment:
# %exp : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%view_1,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp,), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%view_1, 20), kwargs = {})
# %where : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%gt, %view_1, %log1p), kwargs = {})
triton_poi_fused_softplus_0 = async_compile.triton('triton_poi_fused_softplus_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=[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_softplus_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_softplus_0(in_ptr0, out_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)
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), None)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + (x0), tmp5, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/sq/csqyfw4bobalu6ttnn6mknoqfunyxmc3yjmtqqb64gz5w2zf5xvi.py
# Topologically Sorted Source Nodes: [softplus_1], Original ATen: [aten.softplus]
# Source node to ATen node mapping:
# softplus_1 => exp_1, gt_1, log1p_1, where_1
# Graph fragment:
# %exp_1 : [num_users=1] = call_function[target=torch.ops.aten.exp.default](args = (%view_3,), kwargs = {})
# %log1p_1 : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%exp_1,), kwargs = {})
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%view_3, 20), kwargs = {})
# %where_1 : [num_users=1] = call_function[target=torch.ops.aten.where.self](args = (%gt_1, %view_3, %log1p_1), kwargs = {})
triton_poi_fused_softplus_1 = async_compile.triton('triton_poi_fused_softplus_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_softplus_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_softplus_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 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, 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 = args
args.clear()
assert_size_stride(primals_1, (128, 4), (4, 1))
assert_size_stride(primals_2, (128, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 128), (128, 1))
assert_size_stride(primals_5, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 128), (128, 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, 128), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1), torch.float32)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.softplus]
stream0 = get_raw_stream(0)
triton_poi_fused_softplus_0.run(buf0, buf1, 8192, grid=grid(8192), 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, 128), (128, 1), 0), reinterpret_tensor(primals_4, (128, 4), (1, 128), 0), alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [softplus_1], Original ATen: [aten.softplus]
triton_poi_fused_softplus_1.run(buf2, buf3, 256, grid=grid(256), stream=stream0)
return (buf3, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0, reinterpret_tensor(buf1, (64, 128), (128, 1), 0), buf2, 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((128, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((128, ), (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, 128), (128, 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
class PointwiseConvolutionLayer(torch.nn.Module):
def __init__(self, N, F, F_prime):
super().__init__()
self.f1 = torch.nn.Linear(F, 128)
self.f2 = torch.nn.Linear(128, F_prime)
def forward(self, f_bar_batch):
output = torch.nn.functional.softplus(self.f1(f_bar_batch))
return torch.nn.functional.softplus(self.f2(output))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'N': 4, 'F': 4, 'F_prime': 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, math as tl_math
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_softplus_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)
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, None)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp5, None)
@triton.jit
def triton_poi_fused_softplus_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 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp5, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (128, 4), (4, 1))
assert_size_stride(primals_2, (128,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 128), (128, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 128), (1, 4),
0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_softplus_0[grid(8192)](buf0, buf1, 8192, 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, 128),
(128, 1), 0), reinterpret_tensor(primals_4, (128, 4), (1, 128),
0), alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_softplus_1[grid(256)](buf2, buf3, 256, XBLOCK=128,
num_warps=4, num_stages=1)
return buf3, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf0, reinterpret_tensor(buf1, (64, 128), (128, 1), 0
), buf2, primals_4
class PointwiseConvolutionLayerNew(torch.nn.Module):
def __init__(self, N, F, F_prime):
super().__init__()
self.f1 = torch.nn.Linear(F, 128)
self.f2 = torch.nn.Linear(128, F_prime)
def forward(self, input_0):
primals_1 = self.f1.weight
primals_2 = self.f1.bias
primals_4 = self.f2.weight
primals_5 = self.f2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| slgao/FU-DeepLearningCourse | PointwiseConvolutionLayer | false | 4,360 | [
"MIT"
] | 0 | 2300e8bdaa2afb4c73535d5de80874f6103af6f2 | https://github.com/slgao/FU-DeepLearningCourse/tree/2300e8bdaa2afb4c73535d5de80874f6103af6f2 | import torch
class Model(torch.nn.Module):
def __init__(self, N, F, F_prime):
super().__init__()
self.f1 = torch.nn.Linear(F, 128)
self.f2 = torch.nn.Linear(128, F_prime)
def forward(self, f_bar_batch):
output = torch.nn.functional.softplus(self.f1(f_bar_batch))
return torch.nn.functional.softplus(self.f2(output))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
ArcFaceLinear | # 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_7/inductor_cache/zk/czk5xfokmwnuegxn53eciq25366p2is3a6lxx47tlosf3q225vha.py
# Topologically Sorted Source Nodes: [normalize], Original ATen: [aten.div]
# Source node to ATen node mapping:
# normalize => div
# Graph fragment:
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%primals_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=[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_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 = 16
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')
tmp3 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + (4*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 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + (x2), 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, 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((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [normalize], Original ATen: [aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_div_0.run(primals_1, buf0, 16, grid=grid(16), stream=stream0)
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cosine], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_2, (64, 4), (4, 1), 0), reinterpret_tensor(buf0, (4, 4), (1, 4), 0), out=buf1)
del buf0
return (reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 16, 4, 1), 0), primals_1, reinterpret_tensor(primals_2, (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, 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.nn import Module
import math
import torch
import torch.distributed
import torch.nn.functional as F
class ArcFaceLinear(Module):
def __init__(self, embedding_size, num_classes):
super(ArcFaceLinear, self).__init__()
self.weight = torch.nn.Parameter(data=torch.FloatTensor(num_classes,
embedding_size), requires_grad=True)
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
def forward(self, features):
cosine = F.linear(features, F.normalize(self.weight))
return cosine
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'embedding_size': 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._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
from torch.nn import Module
import math
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_div_0(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
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + 4 * 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 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x2, tmp15, 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((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_0[grid(16)](primals_1, buf0, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_2, (64, 4), (4, 1), 0),
reinterpret_tensor(buf0, (4, 4), (1, 4), 0), out=buf1)
del buf0
return reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 16, 4, 1), 0
), primals_1, reinterpret_tensor(primals_2, (64, 4), (4, 1), 0)
class ArcFaceLinearNew(Module):
def __init__(self, embedding_size, num_classes):
super(ArcFaceLinearNew, self).__init__()
self.weight = torch.nn.Parameter(data=torch.FloatTensor(num_classes,
embedding_size), requires_grad=True)
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
def forward(self, input_0):
primals_1 = self.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
| smivv/kaggle-bengali | ArcFaceLinear | false | 4,361 | [
"Apache-2.0"
] | 0 | ab6a2153b657b4f4210551f7f4a674920d66a272 | https://github.com/smivv/kaggle-bengali/tree/ab6a2153b657b4f4210551f7f4a674920d66a272 | from torch.nn import Module
import math
import torch
import torch.distributed
import torch.nn.functional as F
class Model(Module):
def __init__(self, embedding_size, num_classes):
super().__init__()
self.weight = torch.nn.Parameter(data=torch.FloatTensor(num_classes,
embedding_size), requires_grad=True)
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
def forward(self, features):
cosine = F.linear(features, F.normalize(self.weight))
return cosine
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
Encoder | # 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_7/inductor_cache/x2/cx2hdvwyo7m5jvhhvtugzxqvmy6z4nsfhkkjhvgzbbm3cb6dsum2.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %mul_scalar : [num_users=1] = call_function[target=torch.ops.aten.mul.Scalar](args = (%permute_default, 1.0), kwargs = {})
# %clone_default : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/iz/ciztqj6kop3hxov46yrmzprkzfir3eljcic4mkqznz2j5cfeaudr.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %sum_dim_int_list : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_default, [-1], True), kwargs = {})
# %eq_scalar : [num_users=1] = call_function[target=torch.ops.aten.eq.Scalar](args = (%add_tensor, -inf), kwargs = {})
# %logical_not_default : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%eq_scalar,), kwargs = {})
# %any_dim : [num_users=1] = call_function[target=torch.ops.aten.any.dim](args = (%logical_not_default, -1, True), 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=[64],
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + (4*x2), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x2)), 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*x2)), 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*x2)), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float("-inf")
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = (tmp29 != 0)
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = (tmp33 != 0)
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = (tmp38 != 0)
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = (tmp43 != 0)
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + (x2), tmp14, xmask)
tl.store(out_ptr1 + (x2), tmp25, xmask)
tl.store(out_ptr2 + (x2), tmp45, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/x5/cx5uvbfethxuwwkwxf3xaualzhlcwqsz4jxqpbhintggaypzjwqf.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %add_tensor : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_default_2, %primals_8), kwargs = {})
# %amax_default : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_tensor, [-1], True), kwargs = {})
# %sub_tensor : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_tensor, %amax_default), kwargs = {})
# %exp_default : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_tensor,), kwargs = {})
# %div_tensor : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_default, %sum_dim_int_list), kwargs = {})
# %logical_not_default_1 : [num_users=1] = call_function[target=torch.ops.aten.logical_not.default](args = (%any_dim,), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([4, 4, 4, 4], 0), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %where_self : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%logical_not_default_1, %full_default, %div_tensor), kwargs = {})
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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*i1', 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_2', '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_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, XBLOCK : tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = (xindex // 4)
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last').to(tl.int1)
tmp2 = tl.load(in_out_ptr0 + (x4), xmask)
tmp3 = tl.load(in_ptr1 + (x5), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + (x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + (x3), xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + (x4), tmp11, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/vv/cvvnhithjvmvhfjufxwwzclfobkrgbyyteg66hp24r675f7elw4c.py
# Topologically Sorted Source Nodes: [], Original ATen: []
# Source node to ATen node mapping:
# Graph fragment:
# %clone_default_2 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%expand_default_3,), kwargs = {memory_format: torch.contiguous_format})
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=[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_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_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_7/inductor_cache/6t/c6t5a5ere3lqjiu7zh3uu4oxmpdoujdaqqmeunxqapgzo4m74uav.py
# Topologically Sorted Source Nodes: [attentionScores_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# attentionScores_1 => 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')
# kernel path: runs/run_shard_7/inductor_cache/hk/chkirlrxzb52fxbrq2rynamgt7aligt77yn6j6ihfk46whjvd374.py
# Topologically Sorted Source Nodes: [add_1, mu, sub, pow_1, mean_1], Original ATen: [aten.add, aten.mean, aten.sub, aten.pow]
# Source node to ATen node mapping:
# add_1 => add_1
# mean_1 => mean_1
# mu => mean
# pow_1 => pow_1
# sub => sub_1
# Graph fragment:
# %add_1 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_17, %primals_3), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_1, [-1], True), kwargs = {})
# %sub_1 : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_1, %mean), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub_1, 2), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [-1], True), kwargs = {})
triton_poi_fused_add_mean_pow_sub_5 = async_compile.triton('triton_poi_fused_add_mean_pow_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=[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_mean_pow_sub_5', '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_mean_pow_sub_5(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_7/inductor_cache/3j/c3junievyif7korlhvw3ftme6dpzx2uodc43q34sexam3chrqpdb.py
# Topologically Sorted Source Nodes: [add_1, mu, sub, stdArg, std, input_4, mul, normAttOutput], Original ATen: [aten.add, aten.mean, aten.sub, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# add_1 => add_1
# input_4 => div_2
# mu => mean
# mul => mul
# normAttOutput => add_3
# std => sqrt
# stdArg => add_2
# sub => sub_1
# Graph fragment:
# %add_1 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_17, %primals_3), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_1, [-1], True), kwargs = {})
# %sub_1 : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_1, %mean), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mean_1, 1e-12), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add_2,), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_1, %sqrt), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_11, %div_2), kwargs = {})
# %add_3 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul, %primals_12), kwargs = {})
triton_poi_fused_add_div_mean_mul_sqrt_sub_6 = async_compile.triton('triton_poi_fused_add_div_mean_mul_sqrt_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.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_div_mean_mul_sqrt_sub_6', '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_sqrt_sub_6(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
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 = 1e-12
tmp8 = tmp6 + tmp7
tmp9 = libdevice.sqrt(tmp8)
tmp10 = tmp5 / tmp9
tmp11 = tmp0 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + (x2), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/cy/ccybyvkqbybjlur2lbsis3iyvgty2ylsfkqudq5bk64tzxnqmr3u.py
# Topologically Sorted Source Nodes: [pow_2, mul_1, geluPow, mul_2, geluTanh, geluResult, mul_3, intermediate], Original ATen: [aten.pow, aten.mul, aten.add, aten.tanh]
# Source node to ATen node mapping:
# geluPow => add_4
# geluResult => add_5
# geluTanh => tanh
# intermediate => mul_4
# mul_1 => mul_1
# mul_2 => mul_2
# mul_3 => mul_3
# pow_2 => pow_2
# Graph fragment:
# %pow_2 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_19, 3), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%pow_2, 0.044715), kwargs = {})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_19, %mul_1), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_4, 0.7978845608028654), kwargs = {})
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%mul_2,), kwargs = {})
# %add_5 : [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 = (%view_19, 0.5), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_3, %add_5), kwargs = {})
triton_poi_fused_add_mul_pow_tanh_7 = async_compile.triton('triton_poi_fused_add_mul_pow_tanh_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: 'i32'}, 'device': 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_7', '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_7(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_7/inductor_cache/c6/cc6rtxy2qjdhb5zy6oeynlh6ngmrr4eozy6cfv5wwq2vg7sz7tgy.py
# Topologically Sorted Source Nodes: [add_6, mu_1], Original ATen: [aten.add, aten.mean]
# Source node to ATen node mapping:
# add_6 => add_6
# mu_1 => mean_2
# Graph fragment:
# %add_6 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_21, %add_3), kwargs = {})
# %mean_2 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%add_6, [-1], True), kwargs = {})
triton_poi_fused_add_mean_8 = async_compile.triton('triton_poi_fused_add_mean_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_mean_8', '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_add_mean_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
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (0))
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp4 = tl.load(in_ptr2 + (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))
tmp8 = tl.broadcast_to(tmp7, [XBLOCK])
tmp10 = tl.load(in_ptr2 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (2))
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp17 = tl.load(in_ptr2 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp21 = tl.load(in_ptr1 + (3))
tmp22 = tl.broadcast_to(tmp21, [XBLOCK])
tmp24 = tl.load(in_ptr2 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp3 = tmp0 + tmp2
tmp5 = tmp3 + tmp4
tmp9 = tmp6 + tmp8
tmp11 = tmp9 + tmp10
tmp12 = tmp5 + tmp11
tmp16 = tmp13 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp12 + tmp18
tmp23 = tmp20 + tmp22
tmp25 = tmp23 + tmp24
tmp26 = tmp19 + tmp25
tmp27 = 4.0
tmp28 = tmp26 / tmp27
tl.store(out_ptr0 + (x0), tmp28, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/g2/cg2u374wfwwzi2ujjci66zuzc52o4s5xk7vllzergorbsxbqkozm.py
# Topologically Sorted Source Nodes: [add_6, sub_2], Original ATen: [aten.add, aten.sub]
# Source node to ATen node mapping:
# add_6 => add_6
# sub_2 => sub_3
# Graph fragment:
# %add_6 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_21, %add_3), kwargs = {})
# %sub_3 : [num_users=3] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_6, %mean_2), kwargs = {})
triton_poi_fused_add_sub_9 = async_compile.triton('triton_poi_fused_add_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_sub_9', '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_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
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')
tmp3 = tl.load(in_ptr1 + (x2), xmask)
tmp5 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tl.store(in_out_ptr0 + (x2), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/zq/czqvir3otcgawytuzztkv7stq2kaq23wqm7pfjsppb3e4yrb6jwz.py
# Topologically Sorted Source Nodes: [pow_3, mean_3, stdArg_1, std_1, input_5, mul_5, normFFOutput], Original ATen: [aten.pow, aten.mean, aten.add, aten.sqrt, aten.div, aten.mul]
# Source node to ATen node mapping:
# input_5 => div_3
# mean_3 => mean_3
# mul_5 => mul_5
# normFFOutput => add_8
# pow_3 => pow_3
# stdArg_1 => add_7
# std_1 => sqrt_1
# Graph fragment:
# %pow_3 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub_3, 2), kwargs = {})
# %mean_3 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_3, [-1], True), kwargs = {})
# %add_7 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mean_3, 1e-12), kwargs = {})
# %sqrt_1 : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%add_7,), kwargs = {})
# %div_3 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%sub_3, %sqrt_1), kwargs = {})
# %mul_5 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%primals_17, %div_3), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_5, %primals_18), kwargs = {})
triton_poi_fused_add_div_mean_mul_pow_sqrt_10 = async_compile.triton('triton_poi_fused_add_div_mean_mul_pow_sqrt_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_div_mean_mul_pow_sqrt_10', '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_pow_sqrt_10(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 % 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')
tmp4 = tl.load(in_ptr1 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + (4*x1)), xmask, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr2 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = 4.0
tmp14 = tmp12 / tmp13
tmp15 = 1e-12
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp1 / tmp17
tmp19 = tmp0 * tmp18
tmp21 = tmp19 + 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, 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 = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4, ), (1, ))
assert_size_stride(primals_11, (4, ), (1, ))
assert_size_stride(primals_12, (4, ), (1, ))
assert_size_stride(primals_13, (16, 4), (4, 1))
assert_size_stride(primals_14, (16, ), (1, ))
assert_size_stride(primals_15, (4, 16), (16, 1))
assert_size_stride(primals_16, (4, ), (1, ))
assert_size_stride(primals_17, (4, ), (1, ))
assert_size_stride(primals_18, (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 = 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_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_3, (16, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
stream0 = get_raw_stream(0)
triton_poi_fused_0.run(buf0, primals_2, buf3, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_0.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: [], Original ATen: []
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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0); del buf1 # reuse
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_1.run(buf5, primals_8, buf6, buf7, buf8, 64, grid=grid(64), stream=stream0)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_2.run(buf9, buf8, primals_8, buf6, buf7, 256, grid=grid(256), stream=stream0)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
triton_poi_fused_3.run(buf2, primals_7, buf10, 16, 4, grid=grid(16, 4), stream=stream0)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [attentionScores_1], Original ATen: [aten.clone]
triton_poi_fused_clone_4.run(buf11, buf12, 16, 4, grid=grid(16, 4), stream=stream0)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0); del buf11 # reuse
# Topologically Sorted Source Nodes: [attentionOutput], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4), (4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf13)
del primals_10
buf14 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf15 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [add_1, mu, sub, pow_1, mean_1], Original ATen: [aten.add, aten.mean, aten.sub, aten.pow]
triton_poi_fused_add_mean_pow_sub_5.run(buf13, primals_3, buf14, buf15, 16, grid=grid(16), stream=stream0)
buf16 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [add_1, mu, sub, stdArg, std, input_4, mul, normAttOutput], Original ATen: [aten.add, aten.mean, aten.sub, aten.sqrt, aten.div, aten.mul]
triton_poi_fused_add_div_mean_mul_sqrt_sub_6.run(primals_11, buf13, primals_3, buf14, buf15, primals_12, buf16, 64, grid=grid(64), stream=stream0)
del buf14
del primals_12
buf17 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_4], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_14, reinterpret_tensor(buf16, (16, 4), (4, 1), 0), reinterpret_tensor(primals_13, (4, 16), (1, 4), 0), alpha=1, beta=1, out=buf17)
del primals_14
buf18 = empty_strided_cuda((4, 4, 16), (64, 16, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_2, mul_1, geluPow, mul_2, geluTanh, geluResult, mul_3, intermediate], Original ATen: [aten.pow, aten.mul, aten.add, aten.tanh]
triton_poi_fused_add_mul_pow_tanh_7.run(buf17, buf18, 256, grid=grid(256), stream=stream0)
buf19 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf18, (16, 16), (16, 1), 0), reinterpret_tensor(primals_15, (16, 4), (1, 16), 0), out=buf19)
buf20 = buf15; del buf15 # reuse
# Topologically Sorted Source Nodes: [add_6, mu_1], Original ATen: [aten.add, aten.mean]
triton_poi_fused_add_mean_8.run(buf19, primals_16, buf16, buf20, 16, grid=grid(16), stream=stream0)
buf21 = reinterpret_tensor(buf19, (4, 4, 4), (16, 4, 1), 0); del buf19 # reuse
# Topologically Sorted Source Nodes: [add_6, sub_2], Original ATen: [aten.add, aten.sub]
triton_poi_fused_add_sub_9.run(buf21, primals_16, buf16, buf20, 64, grid=grid(64), stream=stream0)
del buf20
del primals_16
buf22 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_3, mean_3, stdArg_1, std_1, input_5, mul_5, normFFOutput], Original ATen: [aten.pow, aten.mean, aten.add, aten.sqrt, aten.div, aten.mul]
triton_poi_fused_add_div_mean_mul_pow_sqrt_10.run(primals_17, buf21, primals_18, buf22, 64, grid=grid(64), stream=stream0)
del primals_18
return (buf22, primals_3, primals_11, primals_17, buf9, reinterpret_tensor(buf10, (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), reinterpret_tensor(buf12, (16, 4), (4, 1), 0), buf13, reinterpret_tensor(buf16, (16, 4), (4, 1), 0), buf17, reinterpret_tensor(buf18, (16, 16), (16, 1), 0), buf21, primals_15, primals_13, primals_9, )
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)
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), (16, 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, ), 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((16, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((4, 16), (16, 1), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_18 = 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, primals_18])
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 nn
class NormLayer(nn.Module):
"""
Implementation of Layer Normalization (https://arxiv.org/abs/1607.06450)
It consists of Batch Normalization Transform to speed up learning with mean and std computed according to the above paper
normWeights:
weights for this normalization layer which will be learnt during training
normBias:
bias for this normalization layer which will be learnt during training
epsilon:
numerical stability parameter to avoid division by zero
"""
def __init__(self, hiddenSize, epsilon=1e-12):
super(NormLayer, self).__init__()
self.weight = nn.Parameter(torch.ones(hiddenSize))
self.bias = nn.Parameter(torch.zeros(hiddenSize))
self.epsilon = epsilon
def forward(self, input):
mu = input.mean(-1, keepdim=True)
stdArg = (input - mu).pow(2).mean(-1, keepdim=True) + self.epsilon
std = torch.sqrt(stdArg)
input = (input - mu) / std
return self.weight * input + self.bias
class GELU(nn.Module):
def __init__(self):
super(GELU, self).__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class FeedForward(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super(FeedForward, self).__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, tensor):
intermediate = self.activationFuncion(self.w1(tensor))
linearOut = self.w2(intermediate)
return self.dropout(linearOut)
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProduct, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class MultiHeadAttention(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super(MultiHeadAttention, self).__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, hiddenStates, attentionMask):
projQ = self.queriesLinear(hiddenStates)
projK = self.keysLinear(hiddenStates)
projV = self.valuesLinear(hiddenStates)
queries = self.prepareForScores(projQ)
keys = self.prepareForScores(projK)
values = self.prepareForScores(projV)
attentionScores = self.sdp(queries, keys, values, attentionMask)
attentionScores = attentionScores.permute(0, 2, 1, 3).contiguous()
newShape = attentionScores.size()[:-2] + (self.numAttentionHeads *
self.attentionHeadSize,)
attentionScores = attentionScores.view(*newShape)
return self.outputLinear(attentionScores)
class Encoder(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, normEpsilon=1e-12,
dropoutProb=0.1):
super(Encoder, self).__init__()
self.multiHeadAtt = MultiHeadAttention(hiddenSize,
numAttentionHeads, dropoutProb)
self.feedForward = FeedForward(hiddenSize, 4 * hiddenSize, dropoutProb)
self.attNorm = NormLayer(hiddenSize, normEpsilon)
self.outputNorm = NormLayer(hiddenSize, normEpsilon)
self.dropout = nn.Dropout(dropoutProb)
def forward(self, hiddenStates, attentionMask):
attentionOutput = self.multiHeadAtt(hiddenStates, attentionMask)
attentionOutput = self.dropout(attentionOutput)
normAttOutput = self.attNorm(attentionOutput + hiddenStates)
ffOutput = self.feedForward(normAttOutput)
normFFOutput = self.outputNorm(ffOutput + normAttOutput)
return normFFOutput
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'hiddenSize': 4, 'numAttentionHeads': 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 math
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_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_1(in_ptr0, in_ptr1, 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
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_ptr0 + 4 * x2, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x2), 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 * x2), 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 * x2), 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 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tmp26 = float('-inf')
tmp27 = tmp2 == tmp26
tmp28 = tmp27 == 0
tmp29 = tmp28.to(tl.int64)
tmp30 = tmp29 != 0
tmp31 = tmp5 == tmp26
tmp32 = tmp31 == 0
tmp33 = tmp32.to(tl.int64)
tmp34 = tmp33 != 0
tmp35 = tmp30 | tmp34
tmp36 = tmp9 == tmp26
tmp37 = tmp36 == 0
tmp38 = tmp37.to(tl.int64)
tmp39 = tmp38 != 0
tmp40 = tmp35 | tmp39
tmp41 = tmp13 == tmp26
tmp42 = tmp41 == 0
tmp43 = tmp42.to(tl.int64)
tmp44 = tmp43 != 0
tmp45 = tmp40 | tmp44
tl.store(out_ptr0 + x2, tmp14, xmask)
tl.store(out_ptr1 + x2, tmp25, xmask)
tl.store(out_ptr2 + x2, tmp45, xmask)
@triton.jit
def triton_poi_fused_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 4
x4 = xindex
x5 = xindex % 64
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last').to(tl
.int1)
tmp2 = tl.load(in_out_ptr0 + x4, xmask)
tmp3 = tl.load(in_ptr1 + x5, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + x3, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + x3, xmask, eviction_policy='evict_last')
tmp1 = tmp0 == 0
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tmp7 = tl_math.exp(tmp6)
tmp9 = tmp7 / tmp8
tmp10 = 0.0
tmp11 = tl.where(tmp1, tmp10, tmp9)
tl.store(in_out_ptr0 + x4, tmp11, xmask)
@triton.jit
def triton_poi_fused_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_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)
@triton.jit
def triton_poi_fused_add_mean_pow_sub_5(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_div_mean_mul_sqrt_sub_6(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
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 = 1e-12
tmp8 = tmp6 + tmp7
tmp9 = libdevice.sqrt(tmp8)
tmp10 = tmp5 / tmp9
tmp11 = tmp0 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + x2, tmp13, xmask)
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_7(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_mean_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
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp4 = tl.load(in_ptr2 + 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)
tmp8 = tl.broadcast_to(tmp7, [XBLOCK])
tmp10 = tl.load(in_ptr2 + (1 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp14 = tl.load(in_ptr1 + 2)
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp17 = tl.load(in_ptr2 + (2 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp21 = tl.load(in_ptr1 + 3)
tmp22 = tl.broadcast_to(tmp21, [XBLOCK])
tmp24 = tl.load(in_ptr2 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp3 = tmp0 + tmp2
tmp5 = tmp3 + tmp4
tmp9 = tmp6 + tmp8
tmp11 = tmp9 + tmp10
tmp12 = tmp5 + tmp11
tmp16 = tmp13 + tmp15
tmp18 = tmp16 + tmp17
tmp19 = tmp12 + tmp18
tmp23 = tmp20 + tmp22
tmp25 = tmp23 + tmp24
tmp26 = tmp19 + tmp25
tmp27 = 4.0
tmp28 = tmp26 / tmp27
tl.store(out_ptr0 + x0, tmp28, xmask)
@triton.jit
def triton_poi_fused_add_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
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')
tmp3 = tl.load(in_ptr1 + x2, xmask)
tmp5 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 - tmp5
tl.store(in_out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_div_mean_mul_pow_sqrt_10(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 % 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')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = 4.0
tmp14 = tmp12 / tmp13
tmp15 = 1e-12
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp1 / tmp17
tmp19 = tmp0 * tmp18
tmp21 = tmp19 + tmp20
tl.store(out_ptr0 + x2, tmp21, 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, primals_18
) = 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))
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), (16, 4, 1))
assert_size_stride(primals_9, (4, 4), (4, 1))
assert_size_stride(primals_10, (4,), (1,))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (4,), (1,))
assert_size_stride(primals_13, (16, 4), (4, 1))
assert_size_stride(primals_14, (16,), (1,))
assert_size_stride(primals_15, (4, 16), (16, 1))
assert_size_stride(primals_16, (4,), (1,))
assert_size_stride(primals_17, (4,), (1,))
assert_size_stride(primals_18, (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 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (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_3, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(16, 4)](buf0, primals_2, buf3, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_2
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_0[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 = reinterpret_tensor(buf1, (4, 4, 4, 1), (16, 4, 1, 64), 0)
del buf1
buf7 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.bool)
triton_poi_fused_1[grid(64)](buf5, primals_8, buf6, buf7, buf8, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused_2[grid(256)](buf9, buf8, primals_8, buf6, buf7,
256, XBLOCK=128, num_warps=4, num_stages=1)
del buf8
del primals_8
buf10 = reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf7
triton_poi_fused_3[grid(16, 4)](buf2, primals_7, buf10, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del primals_7
buf11 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 0), 0), out=buf11)
buf12 = reinterpret_tensor(buf6, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf6
triton_poi_fused_clone_4[grid(16, 4)](buf11, buf12, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf13 = reinterpret_tensor(buf11, (16, 4), (4, 1), 0)
del buf11
extern_kernels.addmm(primals_10, reinterpret_tensor(buf12, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_9, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf13)
del primals_10
buf14 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf15 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused_add_mean_pow_sub_5[grid(16)](buf13, primals_3,
buf14, buf15, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf16 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_sqrt_sub_6[grid(64)](primals_11,
buf13, primals_3, buf14, buf15, primals_12, buf16, 64, XBLOCK=
64, num_warps=1, num_stages=1)
del buf14
del primals_12
buf17 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.addmm(primals_14, reinterpret_tensor(buf16, (16, 4),
(4, 1), 0), reinterpret_tensor(primals_13, (4, 16), (1, 4), 0),
alpha=1, beta=1, out=buf17)
del primals_14
buf18 = empty_strided_cuda((4, 4, 16), (64, 16, 1), torch.float32)
triton_poi_fused_add_mul_pow_tanh_7[grid(256)](buf17, buf18, 256,
XBLOCK=256, num_warps=4, num_stages=1)
buf19 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf18, (16, 16), (16, 1), 0),
reinterpret_tensor(primals_15, (16, 4), (1, 16), 0), out=buf19)
buf20 = buf15
del buf15
triton_poi_fused_add_mean_8[grid(16)](buf19, primals_16, buf16,
buf20, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf21 = reinterpret_tensor(buf19, (4, 4, 4), (16, 4, 1), 0)
del buf19
triton_poi_fused_add_sub_9[grid(64)](buf21, primals_16, buf16,
buf20, 64, XBLOCK=64, num_warps=1, num_stages=1)
del buf20
del primals_16
buf22 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_pow_sqrt_10[grid(64)](primals_17,
buf21, primals_18, buf22, 64, XBLOCK=64, num_warps=1, num_stages=1)
del primals_18
return buf22, primals_3, primals_11, primals_17, buf9, reinterpret_tensor(
buf10, (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
), reinterpret_tensor(buf12, (16, 4), (4, 1), 0
), buf13, reinterpret_tensor(buf16, (16, 4), (4, 1), 0
), buf17, reinterpret_tensor(buf18, (16, 16), (16, 1), 0
), buf21, primals_15, primals_13, primals_9
class NormLayer(nn.Module):
"""
Implementation of Layer Normalization (https://arxiv.org/abs/1607.06450)
It consists of Batch Normalization Transform to speed up learning with mean and std computed according to the above paper
normWeights:
weights for this normalization layer which will be learnt during training
normBias:
bias for this normalization layer which will be learnt during training
epsilon:
numerical stability parameter to avoid division by zero
"""
def __init__(self, hiddenSize, epsilon=1e-12):
super(NormLayer, self).__init__()
self.weight = nn.Parameter(torch.ones(hiddenSize))
self.bias = nn.Parameter(torch.zeros(hiddenSize))
self.epsilon = epsilon
def forward(self, input):
mu = input.mean(-1, keepdim=True)
stdArg = (input - mu).pow(2).mean(-1, keepdim=True) + self.epsilon
std = torch.sqrt(stdArg)
input = (input - mu) / std
return self.weight * input + self.bias
class GELU(nn.Module):
def __init__(self):
super(GELU, self).__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class FeedForward(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super(FeedForward, self).__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, tensor):
intermediate = self.activationFuncion(self.w1(tensor))
linearOut = self.w2(intermediate)
return self.dropout(linearOut)
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super(ScaledDotProduct, self).__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class MultiHeadAttention(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super(MultiHeadAttention, self).__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, hiddenStates, attentionMask):
projQ = self.queriesLinear(hiddenStates)
projK = self.keysLinear(hiddenStates)
projV = self.valuesLinear(hiddenStates)
queries = self.prepareForScores(projQ)
keys = self.prepareForScores(projK)
values = self.prepareForScores(projV)
attentionScores = self.sdp(queries, keys, values, attentionMask)
attentionScores = attentionScores.permute(0, 2, 1, 3).contiguous()
newShape = attentionScores.size()[:-2] + (self.numAttentionHeads *
self.attentionHeadSize,)
attentionScores = attentionScores.view(*newShape)
return self.outputLinear(attentionScores)
class EncoderNew(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, normEpsilon=1e-12,
dropoutProb=0.1):
super(EncoderNew, self).__init__()
self.multiHeadAtt = MultiHeadAttention(hiddenSize,
numAttentionHeads, dropoutProb)
self.feedForward = FeedForward(hiddenSize, 4 * hiddenSize, dropoutProb)
self.attNorm = NormLayer(hiddenSize, normEpsilon)
self.outputNorm = NormLayer(hiddenSize, normEpsilon)
self.dropout = nn.Dropout(dropoutProb)
def forward(self, input_0, input_1):
primals_1 = self.multiHeadAtt.queriesLinear.weight
primals_2 = self.multiHeadAtt.queriesLinear.bias
primals_4 = self.multiHeadAtt.keysLinear.weight
primals_5 = self.multiHeadAtt.keysLinear.bias
primals_6 = self.multiHeadAtt.valuesLinear.weight
primals_7 = self.multiHeadAtt.valuesLinear.bias
primals_9 = self.multiHeadAtt.outputLinear.weight
primals_10 = self.multiHeadAtt.outputLinear.bias
primals_13 = self.feedForward.w1.weight
primals_14 = self.feedForward.w1.bias
primals_15 = self.feedForward.w2.weight
primals_11 = self.feedForward.w2.bias
primals_12 = self.attNorm.weight
primals_16 = self.attNorm.bias
primals_17 = self.outputNorm.weight
primals_18 = self.outputNorm.bias
primals_3 = input_0
primals_8 = input_1
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])
return output[0]
| simonepreite/QABERT | Encoder | false | 4,362 | [
"MIT"
] | 0 | ed3e49f6619f3ff660068291231909693cb8f5d5 | https://github.com/simonepreite/QABERT/tree/ed3e49f6619f3ff660068291231909693cb8f5d5 | import math
import torch
from torch import nn
class NormLayer(nn.Module):
"""
Implementation of Layer Normalization (https://arxiv.org/abs/1607.06450)
It consists of Batch Normalization Transform to speed up learning with mean and std computed according to the above paper
normWeights:
weights for this normalization layer which will be learnt during training
normBias:
bias for this normalization layer which will be learnt during training
epsilon:
numerical stability parameter to avoid division by zero
"""
def __init__(self, hiddenSize, epsilon=1e-12):
super().__init__()
self.weight = nn.Parameter(torch.ones(hiddenSize))
self.bias = nn.Parameter(torch.zeros(hiddenSize))
self.epsilon = epsilon
def forward(self, input):
mu = input.mean(-1, keepdim=True)
stdArg = (input - mu).pow(2).mean(-1, keepdim=True) + self.epsilon
std = torch.sqrt(stdArg)
input = (input - mu) / std
return self.weight * input + self.bias
class GELU(nn.Module):
def __init__(self):
super().__init__()
def forward(self, tensor):
geluPow = tensor + 0.044715 * torch.pow(tensor, 3)
geluTanh = torch.tanh(math.sqrt(2 / math.pi) * geluPow)
geluResult = 1 + geluTanh
return 0.5 * tensor * geluResult
class FeedForward(nn.Module):
def __init__(self, hiddenSize, innerLayerDimension, dropOutProb=0.1):
super().__init__()
self.activationFuncion = GELU()
self.dropout = nn.Dropout(dropOutProb)
self.w1 = nn.Linear(hiddenSize, innerLayerDimension)
self.w2 = nn.Linear(innerLayerDimension, hiddenSize)
def forward(self, tensor):
intermediate = self.activationFuncion(self.w1(tensor))
linearOut = self.w2(intermediate)
return self.dropout(linearOut)
class ScaledDotProduct(nn.Module):
def __init__(self, attentionHeadSize, dropOutProb=0.1):
super().__init__()
self.attentionHeadSize = attentionHeadSize
self.dropout = nn.Dropout(dropOutProb)
def forward(self, Q, K, V, attentionMask):
aScores = torch.matmul(Q, K.transpose(-1, -2)) / math.sqrt(self.
attentionHeadSize)
aScores = aScores + attentionMask
aProbs = self.dropout(nn.Softmax(dim=-1)(aScores))
return torch.matmul(aProbs, V)
class MultiHeadAttention(nn.Module):
def __init__(self, hiddenSize, numAttentionHeads, dropoutProb):
super().__init__()
if hiddenSize % numAttentionHeads != 0:
raise ValueError(
'The hidden size ({}) is not a multiple of the numeber of attention heads ({})'
.format(hiddenSize, numAttentionHeads))
self.numAttentionHeads = numAttentionHeads
self.attentionHeadSize = hiddenSize // self.numAttentionHeads
self.queriesLinear = nn.Linear(hiddenSize, hiddenSize)
self.keysLinear = nn.Linear(hiddenSize, hiddenSize)
self.valuesLinear = nn.Linear(hiddenSize, hiddenSize)
self.sdp = ScaledDotProduct(self.attentionHeadSize, dropoutProb)
self.outputLinear = nn.Linear(hiddenSize, hiddenSize)
def prepareForScores(self, input):
newShape = input.size()[:-1] + (self.numAttentionHeads, self.
attentionHeadSize)
input = input.view(*newShape)
return input.permute(0, 2, 1, 3)
def forward(self, hiddenStates, attentionMask):
projQ = self.queriesLinear(hiddenStates)
projK = self.keysLinear(hiddenStates)
projV = self.valuesLinear(hiddenStates)
queries = self.prepareForScores(projQ)
keys = self.prepareForScores(projK)
values = self.prepareForScores(projV)
attentionScores = self.sdp(queries, keys, values, attentionMask)
attentionScores = attentionScores.permute(0, 2, 1, 3).contiguous()
newShape = attentionScores.size()[:-2] + (self.numAttentionHeads *
self.attentionHeadSize,)
atten
# ... truncated (>4000 chars) for memory efficiency |
InnerProductDecoder | # 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_7/inductor_cache/mb/cmb72vxh36b4k6lvmt4562lj3nrqtpyzst2qbon2yqx22gdjfa7x.py
# Topologically Sorted Source Nodes: [adj], Original ATen: [aten.sigmoid]
# Source node to ATen node mapping:
# adj => sigmoid
# Graph fragment:
# %sigmoid : [num_users=1] = call_function[target=torch.ops.aten.sigmoid.default](args = (%mm,), kwargs = {})
triton_poi_fused_sigmoid_0 = async_compile.triton('triton_poi_fused_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=[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_sigmoid_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_sigmoid_0(in_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_out_ptr0 + (x0), xmask)
tmp1 = tl.sigmoid(tmp0)
tl.store(in_out_ptr0 + (x0), tmp1, 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, 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: [mm], Original ATen: [aten.mm]
extern_kernels.mm(arg0_1, reinterpret_tensor(arg0_1, (4, 4), (1, 4), 0), out=buf0)
del arg0_1
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [adj], Original ATen: [aten.sigmoid]
stream0 = get_raw_stream(0)
triton_poi_fused_sigmoid_0.run(buf1, 16, grid=grid(16), stream=stream0)
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, 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
import torch.nn.modules.loss
import torch.nn.functional as F
import torch.nn as nn
class InnerProductDecoder(nn.Module):
"""Decoder for using inner product for prediction."""
def __init__(self, dropout, act=torch.sigmoid):
super(InnerProductDecoder, self).__init__()
self.dropout = dropout
self.act = act
def forward(self, z):
z = F.dropout(z, self.dropout, training=self.training)
adj = self.act(torch.mm(z, z.t()))
return adj
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'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
import torch.nn
import torch.nn.modules.loss
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_sigmoid_0(in_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_out_ptr0 + x0, xmask)
tmp1 = tl.sigmoid(tmp0)
tl.store(in_out_ptr0 + x0, tmp1, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (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(arg0_1, reinterpret_tensor(arg0_1, (4, 4), (1, 4),
0), out=buf0)
del arg0_1
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_sigmoid_0[grid(16)](buf1, 16, XBLOCK=16, num_warps
=1, num_stages=1)
return buf1,
class InnerProductDecoderNew(nn.Module):
"""Decoder for using inner product for prediction."""
def __init__(self, dropout, act=torch.sigmoid):
super(InnerProductDecoderNew, self).__init__()
self.dropout = dropout
self.act = act
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| spatial-Transcriptomics/DeepST | InnerProductDecoder | false | 4,363 | [
"MIT"
] | 0 | 47ce64b06b62395cd2983939d4bf2419f558a562 | https://github.com/spatial-Transcriptomics/DeepST/tree/47ce64b06b62395cd2983939d4bf2419f558a562 | import torch
import torch.nn
import torch.nn.modules.loss
import torch.nn.functional as F
import torch.nn as nn
class Model(nn.Module):
"""Decoder for using inner product for prediction."""
def __init__(self, dropout, act=torch.sigmoid):
super().__init__()
self.dropout = dropout
self.act = act
def forward(self, z):
z = F.dropout(z, self.dropout, training=self.training)
adj = self.act(torch.mm(z, z.t()))
return adj
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [0.5]
|
Encoder | # 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_7/inductor_cache/do/cdo22no4lmipk7byduyah2xsadvdcbfr22puoptl5br3l66r6jra.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.leaky_relu]
# Source node to ATen node mapping:
# x => 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, 0.01), 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 = 0.01
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, primals_4, primals_5, primals_6, primals_7, primals_8, primals_9 = 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, ))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (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: [x], 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 primals_2
buf3 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf2, (64, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.leaky_relu]
triton_poi_fused_leaky_relu_0.run(buf3, primals_5, buf4, buf5, 256, grid=grid(256), stream=stream0)
del primals_5
buf6 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [mean], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf5, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf6)
del primals_7
buf7 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [log_variance], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_9, reinterpret_tensor(buf5, (64, 4), (4, 1), 0), reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf7)
del primals_9
return (reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(buf7, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf1, reinterpret_tensor(buf2, (64, 4), (4, 1), 0), buf4, reinterpret_tensor(buf5, (64, 4), (4, 1), 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, 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)
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)
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.functional as F
class Encoder(torch.nn.Module):
"""Documentation for Encoder
"""
def __init__(self, input_dim, hidden_dim, latent_dim):
super(Encoder, self).__init__()
self.e1 = torch.nn.Linear(input_dim, hidden_dim)
self.e2 = torch.nn.Linear(hidden_dim, hidden_dim)
self.e3 = torch.nn.Linear(hidden_dim, latent_dim)
self.e4 = torch.nn.Linear(hidden_dim, latent_dim)
def forward(self, x):
x = F.leaky_relu(self.e1(x))
x = F.leaky_relu(self.e2(x))
mean = self.e3(x)
log_variance = self.e4(x)
return mean, log_variance
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_dim': 4, 'hidden_dim': 4, 'latent_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
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 = 0.01
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, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = 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,))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (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=128, num_warps=4, num_stages=1)
del primals_2
buf3 = buf0
del buf0
extern_kernels.mm(reinterpret_tensor(buf2, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_leaky_relu_0[grid(256)](buf3, primals_5, buf4,
buf5, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf6 = buf3
del buf3
extern_kernels.addmm(primals_7, reinterpret_tensor(buf5, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf6)
del primals_7
buf7 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_9, reinterpret_tensor(buf5, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_8, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf7)
del primals_9
return reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(buf7, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf1, reinterpret_tensor(buf2, (64, 4), (4, 1), 0
), buf4, reinterpret_tensor(buf5, (64, 4), (4, 1), 0
), primals_8, primals_6, primals_4
class EncoderNew(torch.nn.Module):
"""Documentation for Encoder
"""
def __init__(self, input_dim, hidden_dim, latent_dim):
super(EncoderNew, self).__init__()
self.e1 = torch.nn.Linear(input_dim, hidden_dim)
self.e2 = torch.nn.Linear(hidden_dim, hidden_dim)
self.e3 = torch.nn.Linear(hidden_dim, latent_dim)
self.e4 = torch.nn.Linear(hidden_dim, latent_dim)
def forward(self, input_0):
primals_1 = self.e1.weight
primals_2 = self.e1.bias
primals_4 = self.e2.weight
primals_5 = self.e2.bias
primals_6 = self.e3.weight
primals_7 = self.e3.bias
primals_8 = self.e4.weight
primals_9 = self.e4.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], output[1]
| slgao/FU-DeepLearningCourse | Encoder | false | 4,364 | [
"MIT"
] | 0 | 2300e8bdaa2afb4c73535d5de80874f6103af6f2 | https://github.com/slgao/FU-DeepLearningCourse/tree/2300e8bdaa2afb4c73535d5de80874f6103af6f2 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""Documentation for Encoder
"""
def __init__(self, input_dim, hidden_dim, latent_dim):
super().__init__()
self.e1 = torch.nn.Linear(input_dim, hidden_dim)
self.e2 = torch.nn.Linear(hidden_dim, hidden_dim)
self.e3 = torch.nn.Linear(hidden_dim, latent_dim)
self.e4 = torch.nn.Linear(hidden_dim, latent_dim)
def forward(self, x):
x = F.leaky_relu(self.e1(x))
x = F.leaky_relu(self.e2(x))
mean = self.e3(x)
log_variance = self.e4(x)
return mean, log_variance
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
PartitionedMultiHeadAttention | # 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_7/inductor_cache/cc/cccjs22xobelgxtqlsn2fibmbazwulpl6qiesp7clmtct5t76w6i.py
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# qkv_c => 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_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 = 48
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = (xindex // 6) % 4
x2 = (xindex // 24)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (6*x2) + (12*x1)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/va/cva5gwn5lddb4hdqcni22rre3lrnt4hodmwd6zzu7u23zcobkg5i.py
# Topologically Sorted Source Nodes: [cat, q], Original ATen: [aten.cat, aten.mul]
# Source node to ATen node mapping:
# cat => cat
# q => mul
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze, %squeeze_3], -1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%cat, 0.5), kwargs = {})
triton_poi_fused_cat_mul_1 = async_compile.triton('triton_poi_fused_cat_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=[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_cat_mul_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_mul_1(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 % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = 0.5
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/q4/cq4j3evfcbzbqvsc5o6spd3cwbk5ruwxtlfpvyrijy74rzhdizzf.py
# Topologically Sorted Source Nodes: [k], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# k => cat_1
# Graph fragment:
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze_1, %squeeze_4], -1), 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=[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_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 = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2 + (6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + (2 + (6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/s4/cs472yivvl3yzse325afzknsz7ua5dqrqzmwls3lwujk3hte6xkl.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_11, [-1], 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_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 = 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_7/inductor_cache/zh/czh6tw7ngffcygnivwvcjex5edxy3ms4t27ymyn2hemxlpspxzq7.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => 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_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 = 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_7/inductor_cache/wy/cwyd2c5eqe6dmcriegfsxxijhdetgyb5fhlqse4jkwrdzjzz4yra.py
# Topologically Sorted Source Nodes: [v], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# v => cat_2
# Graph fragment:
# %cat_2 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze_2, %squeeze_5], -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=[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_cat_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_cat_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 % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 + (6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + (4 + (6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rb/crbo7mnyz6jhctfcs6irltofws4absfehq57gbubuwmkgqremvir.py
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_c => clone_3
# Graph fragment:
# %clone_3 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_22,), 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=[32, 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 = 32
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
x3 = xindex
y0 = yindex % 2
y1 = (yindex // 2) % 4
y2 = (yindex // 8)
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*y1) + (16*x3) + (64*y2)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + (4*y4)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/uf/cufwzthblw5bpxm6to23u5x5pyxdqityp5l4pgpwkcw24uzkh5ok.py
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_c => clone_4
# Graph fragment:
# %clone_4 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_23,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_7 = async_compile.triton('triton_poi_fused_clone_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: 'i32'}, 'device': 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_7', '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_7(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
x0 = xindex % 2
x1 = (xindex // 2) % 4
x2 = (xindex // 8)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (2*x2) + (4*x1)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/in/cinchtemkwlurwioe377jflj2qj5ijghnp4kjugwuldc3qsv4n6b.py
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_p => clone_5
# Graph fragment:
# %clone_5 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_27,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_8 = async_compile.triton('triton_poi_fused_clone_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=[32, 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_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_clone_8(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = (yindex // 2) % 4
y2 = (yindex // 8)
y4 = yindex
tmp0 = tl.load(in_ptr0 + (2 + y0 + (4*y1) + (16*x3) + (64*y2)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + (4*y4)), 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 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_3, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_4, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_5, (4, 2, 2), (4, 2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((2, 4, 3, 2, 1, 1), (24, 6, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(primals_2, buf0, 48, grid=grid(48), stream=stream0)
del primals_2
buf1 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4, 1), 0), reinterpret_tensor(buf0, (1, 2, 24), (0, 24, 1), 0), out=buf1)
buf2 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [qkv_p], Original ATen: [aten.clone]
triton_poi_fused_clone_0.run(primals_3, buf2, 48, grid=grid(48), stream=stream0)
del primals_3
buf3 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_p], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4, 1), 2), reinterpret_tensor(buf2, (1, 2, 24), (0, 24, 1), 0), out=buf3)
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat, q], Original ATen: [aten.cat, aten.mul]
triton_poi_fused_cat_mul_1.run(buf1, buf3, buf4, 256, grid=grid(256), stream=stream0)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [k], Original ATen: [aten.cat]
triton_poi_fused_cat_2.run(buf1, buf3, buf5, 256, grid=grid(256), stream=stream0)
buf6 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [dots], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf4, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf5, (16, 4, 4), (16, 1, 4), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
triton_poi_fused__softmax_3.run(buf6, buf7, 256, grid=grid(256), stream=stream0)
buf8 = reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
triton_poi_fused__softmax_4.run(buf7, buf8, 256, grid=grid(256), stream=stream0)
buf9 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [v], Original ATen: [aten.cat]
triton_poi_fused_cat_5.run(buf1, buf3, buf9, 256, grid=grid(256), stream=stream0)
del buf1
del buf3
buf10 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [o], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
triton_poi_fused_clone_6.run(buf10, buf11, 32, 4, grid=grid(32, 4), stream=stream0)
buf12 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(primals_4, buf12, 16, grid=grid(16), stream=stream0)
del primals_4
buf13 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf11, (1, 16, 8), (0, 8, 1), 0), reinterpret_tensor(buf12, (1, 8, 2), (0, 2, 1), 0), out=buf13)
buf14 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
triton_poi_fused_clone_8.run(buf10, buf14, 32, 4, grid=grid(32, 4), stream=stream0)
del buf10
buf15 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(primals_5, buf15, 16, grid=grid(16), stream=stream0)
del primals_5
buf16 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf14, (1, 16, 8), (0, 8, 1), 0), reinterpret_tensor(buf15, (1, 8, 2), (0, 2, 1), 0), out=buf16)
return (reinterpret_tensor(buf13, (4, 4, 2), (8, 2, 1), 0), reinterpret_tensor(buf16, (4, 4, 2), (8, 2, 1), 0), buf8, reinterpret_tensor(buf14, (1, 8, 16), (128, 1, 8), 0), reinterpret_tensor(buf15, (1, 2, 8), (16, 1, 2), 0), reinterpret_tensor(buf11, (1, 8, 16), (128, 1, 8), 0), reinterpret_tensor(buf12, (1, 2, 8), (16, 1, 2), 0), reinterpret_tensor(buf9, (16, 4, 4), (16, 1, 4), 0), reinterpret_tensor(buf4, (16, 4, 4), (16, 1, 4), 0), reinterpret_tensor(buf5, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 2), reinterpret_tensor(primals_1, (1, 2, 16), (64, 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, 2, 3, 2), (12, 6, 2, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 2, 3, 2), (12, 6, 2, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 2, 2), (4, 2, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 2, 2), (4, 2, 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
import torch.nn.functional as F
class PartitionedMultiHeadAttention(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, x, mask=None):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
qkv_c = torch.einsum('btf,hfca->bhtca', x_c, self.w_qkv_c)
qkv_p = torch.einsum('btf,hfca->bhtca', x_p, self.w_qkv_p)
q_c, k_c, v_c = [c.squeeze(dim=3) for c in torch.chunk(qkv_c, 3, dim=3)
]
q_p, k_p, v_p = [c.squeeze(dim=3) for c in torch.chunk(qkv_p, 3, dim=3)
]
q = torch.cat([q_c, q_p], dim=-1) * self.scaling_factor
k = torch.cat([k_c, k_p], dim=-1)
v = torch.cat([v_c, v_p], dim=-1)
dots = torch.einsum('bhqa,bhka->bhqk', q, k)
if mask is not None:
dots.data.masked_fill_(~mask[:, None, None, :], -float('inf'))
probs = F.softmax(dots, dim=-1)
probs = self.dropout(probs)
o = torch.einsum('bhqk,bhka->bhqa', probs, v)
o_c, o_p = torch.chunk(o, 2, dim=-1)
out_c = torch.einsum('bhta,haf->btf', o_c, self.w_o_c)
out_p = torch.einsum('bhta,haf->btf', o_p, self.w_o_p)
return out_c, out_p
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4, 'n_head': 4, 'd_qkv': 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_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 48
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = xindex // 6 % 4
x2 = xindex // 24
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 6 * x2 + 12 * x1), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_poi_fused_cat_mul_1(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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (6 * x4 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = 0.5
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp12, xmask)
@triton.jit
def triton_poi_fused_cat_2(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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2 + 6 * x4 + x0), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (2 + 6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp10, 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 = 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_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 = 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_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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 + 6 * x4 + x0), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 + 6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp10, xmask)
@triton.jit
def triton_poi_fused_clone_6(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = yindex // 2 % 4
y2 = yindex // 8
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * y1 + 16 * x3 + 64 * y2), xmask &
ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + 4 * y4), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_7(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
x0 = xindex % 2
x1 = xindex // 2 % 4
x2 = xindex // 8
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 2 * x2 + 4 * x1), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_poi_fused_clone_8(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = yindex // 2 % 4
y2 = yindex // 8
y4 = yindex
tmp0 = tl.load(in_ptr0 + (2 + y0 + 4 * y1 + 16 * x3 + 64 * y2), xmask &
ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + 4 * y4), tmp0, 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, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_3, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_4, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_5, (4, 2, 2), (4, 2, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((2, 4, 3, 2, 1, 1), (24, 6, 2, 1, 1, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(48)](primals_2, buf0, 48, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_2
buf1 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4,
1), 0), reinterpret_tensor(buf0, (1, 2, 24), (0, 24, 1), 0),
out=buf1)
buf2 = buf0
del buf0
triton_poi_fused_clone_0[grid(48)](primals_3, buf2, 48, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_3
buf3 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4,
1), 2), reinterpret_tensor(buf2, (1, 2, 24), (0, 24, 1), 0),
out=buf3)
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_cat_mul_1[grid(256)](buf1, buf3, buf4, 256, XBLOCK
=256, num_warps=4, num_stages=1)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_cat_2[grid(256)](buf1, buf3, buf5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf6 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf4, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf5, (16, 4, 4), (16, 1, 4), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_3[grid(256)](buf6, buf7, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf8 = reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf6
triton_poi_fused__softmax_4[grid(256)](buf7, buf8, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf9 = buf7
del buf7
triton_poi_fused_cat_5[grid(256)](buf1, buf3, buf9, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf1
del buf3
buf10 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch
.float32)
triton_poi_fused_clone_6[grid(32, 4)](buf10, buf11, 32, 4, XBLOCK=4,
YBLOCK=32, num_warps=4, num_stages=1)
buf12 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.
float32)
triton_poi_fused_clone_7[grid(16)](primals_4, buf12, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_4
buf13 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf11, (1, 16, 8), (0, 8, 1),
0), reinterpret_tensor(buf12, (1, 8, 2), (0, 2, 1), 0), out=buf13)
buf14 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch
.float32)
triton_poi_fused_clone_8[grid(32, 4)](buf10, buf14, 32, 4, XBLOCK=4,
YBLOCK=32, num_warps=4, num_stages=1)
del buf10
buf15 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.
float32)
triton_poi_fused_clone_7[grid(16)](primals_5, buf15, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_5
buf16 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf14, (1, 16, 8), (0, 8, 1),
0), reinterpret_tensor(buf15, (1, 8, 2), (0, 2, 1), 0), out=buf16)
return reinterpret_tensor(buf13, (4, 4, 2), (8, 2, 1), 0
), reinterpret_tensor(buf16, (4, 4, 2), (8, 2, 1), 0
), buf8, reinterpret_tensor(buf14, (1, 8, 16), (128, 1, 8), 0
), reinterpret_tensor(buf15, (1, 2, 8), (16, 1, 2), 0
), reinterpret_tensor(buf11, (1, 8, 16), (128, 1, 8), 0
), reinterpret_tensor(buf12, (1, 2, 8), (16, 1, 2), 0
), reinterpret_tensor(buf9, (16, 4, 4), (16, 1, 4), 0
), reinterpret_tensor(buf4, (16, 4, 4), (16, 1, 4), 0
), reinterpret_tensor(buf5, (16, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 2
), reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 0)
class PartitionedMultiHeadAttentionNew(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, input_0):
primals_2 = self.w_qkv_c
primals_3 = self.w_qkv_p
primals_4 = self.w_o_c
primals_5 = self.w_o_p
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0], output[1]
| skulick/self-attentive-parser | PartitionedMultiHeadAttention | false | 4,365 | [
"MIT"
] | 0 | 04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | https://github.com/skulick/self-attentive-parser/tree/04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, x, mask=None):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
qkv_c = torch.einsum('btf,hfca->bhtca', x_c, self.w_qkv_c)
qkv_p = torch.einsum('btf,hfca->bhtca', x_p, self.w_qkv_p)
q_c, k_c, v_c = [c.squeeze(dim=3) for c in torch.chunk(qkv_c, 3, dim=3)
]
q_p, k_p, v_p = [c.squeeze(dim=3) for c in torch.chunk(qkv_p, 3, dim=3)
]
q = torch.cat([q_c, q_p], dim=-1) * self.scaling_factor
k = torch.cat([k_c, k_p], dim=-1)
v = torch.cat([v_c, v_p], dim=-1)
dots = torch.einsum('bhqa,bhka->bhqk', q, k)
if mask is not None:
dots.data.masked_fill_(~mask[:, None, None, :], -float('inf'))
probs = F.softmax(dots, dim=-1)
probs = self.dropout(probs)
o = torch.einsum('bhqk,bhka->bhqa', probs, v)
o_c, o_p = torch.chunk(o, 2, dim=-1)
out_c = torch.einsum('bhta,haf->btf', o_c, self.w_o_c)
out_p = torch.einsum('bhta,haf->btf', o_p, self.w_o_p)
return out_c, out_p
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
CausalConv1d | # 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_7/inductor_cache/pz/cpzxz6lkuih5yejx2jgx2k2tt7datj3a4ghmi4d5empkhgoz5dif.py
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# out => 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], False, [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=[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_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 = 80
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 5) % 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, 2), (8, 2, 1))
assert_size_stride(primals_2, (4, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4), (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(primals_3, primals_1, stride=(1,), padding=(1,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 5), (20, 5, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_0.run(buf1, primals_2, 80, grid=grid(80), stream=stream0)
del primals_2
return (reinterpret_tensor(buf1, (4, 4, 4), (20, 5, 1), 0), 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, 2), (8, 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), (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 CausalConv1d(nn.Conv1d):
def __init__(self, in_channels, out_channels, kernel_size=2, dilation=1,
**kwargs):
super(CausalConv1d, self).__init__(in_channels, out_channels,
kernel_size, padding=dilation * (kernel_size - 1), dilation=
dilation, **kwargs)
def forward(self, input):
out = super(CausalConv1d, self).forward(input)
return out[:, :, :-self.padding[0]]
def get_inputs():
return [torch.rand([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
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 = 80
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 5 % 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, 2), (8, 2, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,),
padding=(1,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 5), (20, 5, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(80)](buf1, primals_2, 80,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
return reinterpret_tensor(buf1, (4, 4, 4), (20, 5, 1), 0
), primals_1, primals_3
class CausalConv1dNew(nn.Conv1d):
def __init__(self, in_channels, out_channels, kernel_size=2, dilation=1,
**kwargs):
super(CausalConv1dNew, self).__init__(in_channels, out_channels,
kernel_size, padding=dilation * (kernel_size - 1), dilation=
dilation, **kwargs)
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]
| soumyac1999/instrumental-music-translation | CausalConv1d | false | 4,366 | [
"MIT"
] | 0 | f0d5edfdf34ef7bc9b329c426089f61d3468caa8 | https://github.com/soumyac1999/instrumental-music-translation/tree/f0d5edfdf34ef7bc9b329c426089f61d3468caa8 | import torch
import torch.nn as nn
class Model(nn.Conv1d):
def __init__(self, in_channels, out_channels, kernel_size=2, dilation=1,
**kwargs):
super().__init__(in_channels, out_channels,
kernel_size, padding=dilation * (kernel_size - 1), dilation=
dilation, **kwargs)
def forward(self, input):
out = super(CausalConv1d, self).forward(input)
return out[:, :, :-self.padding[0]]
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
RefModel3d2 | # 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_7/inductor_cache/lu/cluznd6mcssryycrrxyrun3thcx37iyatgrin4volajsy7rmbgly.py
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.replication_pad3d]
# Source node to ATen node mapping:
# pad => _unsafe_index, _unsafe_index_1, _unsafe_index_2
# Graph fragment:
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%primals_2, [None, None, %clamp_max, None, None]), kwargs = {})
# %_unsafe_index_1 : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index, [None, None, None, %clamp_max, None]), kwargs = {})
# %_unsafe_index_2 : [num_users=2] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%_unsafe_index_1, [None, None, None, None, %clamp_max]), kwargs = {})
triton_poi_fused_replication_pad3d_0 = async_compile.triton('triton_poi_fused_replication_pad3d_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=[4194304],
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_replication_pad3d_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_replication_pad3d_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2299968
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 66
x1 = (xindex // 66) % 66
x2 = (xindex // 4356) % 66
x3 = (xindex // 287496)
x4 = xindex
tmp0 = tl.load(in_ptr0 + ((64*((63) * ((63) <= (((0) * ((0) >= ((-1) + x1)) + ((-1) + x1) * (((-1) + x1) > (0))))) + (((0) * ((0) >= ((-1) + x1)) + ((-1) + x1) * (((-1) + x1) > (0)))) * ((((0) * ((0) >= ((-1) + x1)) + ((-1) + x1) * (((-1) + x1) > (0)))) < (63)))) + (4096*((63) * ((63) <= (((0) * ((0) >= ((-1) + x2)) + ((-1) + x2) * (((-1) + x2) > (0))))) + (((0) * ((0) >= ((-1) + x2)) + ((-1) + x2) * (((-1) + x2) > (0)))) * ((((0) * ((0) >= ((-1) + x2)) + ((-1) + x2) * (((-1) + x2) > (0)))) < (63)))) + (262144*x3) + ((63) * ((63) <= (((0) * ((0) >= ((-1) + x0)) + ((-1) + x0) * (((-1) + x0) > (0))))) + (((0) * ((0) >= ((-1) + x0)) + ((-1) + x0) * (((-1) + x0) > (0)))) * ((((0) * ((0) >= ((-1) + x0)) + ((-1) + x0) * (((-1) + x0) > (0)))) < (63)))), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x4), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/me/cme4fiodyseceoiaqotpbppmkhgcmw5qh545wuq6s3eddheyio2u.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.native_group_norm]
# Source node to ATen node mapping:
# output_1 => var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view, [2, 3]), kwargs = {correction: 0, keepdim: True})
triton_red_fused_native_group_norm_1 = async_compile.triton('triton_red_fused_native_group_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.reduction(
size_hints=[32, 8192],
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, 4, 5), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused_native_group_norm_1', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 1, '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_red_fused_native_group_norm_1(in_ptr0, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 32
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp2_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp2_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp2_weight = tl.zeros([XBLOCK, RBLOCK], tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + (8192*x0)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp2_mean_next, tmp2_m2_next, tmp2_weight_next = triton_helpers.welford_reduce(
tmp1, tmp2_mean, tmp2_m2, tmp2_weight, roffset == 0
)
tmp2_mean = tl.where(rmask & xmask, tmp2_mean_next, tmp2_mean)
tmp2_m2 = tl.where(rmask & xmask, tmp2_m2_next, tmp2_m2)
tmp2_weight = tl.where(rmask & xmask, tmp2_weight_next, tmp2_weight)
tmp2_tmp, tmp3_tmp, tmp4_tmp = triton_helpers.welford(
tmp2_mean, tmp2_m2, tmp2_weight, 1
)
tmp2 = tmp2_tmp[:, None]
tmp3 = tmp3_tmp[:, None]
tmp4 = tmp4_tmp[:, None]
tl.store(out_ptr0 + (x0), tmp2, xmask)
tl.store(out_ptr1 + (x0), tmp3, xmask)
tl.store(out_ptr2 + (x0), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/bv/cbvolr7jvlzbzlmnm5267acxusvkinldk4rkup4ifssehwjmkwb2.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.native_group_norm]
# Source node to ATen node mapping:
# output_1 => add, rsqrt, var_mean
# Graph fragment:
# %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 = {})
triton_per_fused_native_group_norm_2 = async_compile.triton('triton_per_fused_native_group_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=[8, 4],
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), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_native_group_norm_2', 'mutated_arg_names': ['in_out_ptr0'], 'no_x_dim': False, 'num_load': 3, '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_native_group_norm_2(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 8
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_ptr1 + (r1 + (4*x0)), xmask, other=0.0)
tmp2 = tl.load(in_ptr2 + (r1 + (4*x0)), xmask, other=0.0)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp5 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp7 = tl.where(xmask, tmp3, 0)
tmp8 = tl.where(xmask, tmp4, 0)
tmp9 = tl.where(xmask, tmp5, 0)
tmp10, tmp11, tmp12 = triton_helpers.welford(tmp7, tmp8, tmp9, 1)
tmp13 = tmp10[:, None]
tmp14 = tmp11[:, None]
tmp15 = tmp12[:, None]
tmp16 = 32768.0
tmp17 = tmp14 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tl.debug_barrier()
tl.store(in_out_ptr0 + (x0), tmp20, xmask)
tl.store(out_ptr0 + (x0), tmp13, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/lw/clwxbjrozyrufclp6lspsimcypqk4c45zkolg6wwxvnj3474halb.py
# Topologically Sorted Source Nodes: [output_1, output_2], Original ATen: [aten.native_group_norm, aten.leaky_relu]
# Source node to ATen node mapping:
# output_1 => add_1, mul_1
# output_2 => gt, mul_2, where
# Graph fragment:
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, %unsqueeze_7), kwargs = {})
# %add_1 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %unsqueeze_3), kwargs = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%add_1, 0), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_1, 0.02), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %add_1, %mul_2), kwargs = {})
triton_poi_fused_leaky_relu_native_group_norm_3 = async_compile.triton('triton_poi_fused_leaky_relu_native_group_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.pointwise(
size_hints=[262144],
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_leaky_relu_native_group_norm_3', '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_leaky_relu_native_group_norm_3(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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
x4 = (xindex // 32768)
x1 = (xindex // 32768) % 2
tmp0 = tl.load(in_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr1 + (x4), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x4), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (x1), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tmp9 = 0.0
tmp10 = tmp8 > tmp9
tmp11 = 0.02
tmp12 = tmp8 * tmp11
tmp13 = tl.where(tmp10, tmp8, tmp12)
tl.store(in_out_ptr0 + (x3), tmp13, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/w4/cw4htj27jyad6revrg52vfqvuwh2yfl5ehkfap2qfnolhs7penho.py
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.avg_pool3d]
# Source node to ATen node mapping:
# output_3 => avg_pool3d
# Graph fragment:
# %avg_pool3d : [num_users=2] = call_function[target=torch.ops.aten.avg_pool3d.default](args = (%where, [2, 2, 2], [4, 4, 4]), kwargs = {})
triton_poi_fused_avg_pool3d_4 = async_compile.triton('triton_poi_fused_avg_pool3d_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_avg_pool3d_4', '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_avg_pool3d_4(in_ptr0, out_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)
x0 = xindex % 8
x1 = (xindex // 8) % 8
x2 = (xindex // 64)
x3 = xindex
tmp0 = tl.load(in_ptr0 + ((4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (32 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (33 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (1024 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (1025 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (1056 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (1057 + (4*x0) + (128*x1) + (4096*x2)), None, eviction_policy='evict_last')
tmp2 = tmp1 + tmp0
tmp4 = tmp3 + tmp2
tmp6 = tmp5 + tmp4
tmp8 = tmp7 + tmp6
tmp10 = tmp9 + tmp8
tmp12 = tmp11 + tmp10
tmp14 = tmp13 + tmp12
tmp15 = 0.125
tmp16 = tmp14 * tmp15
tl.store(out_ptr0 + (x3), tmp16, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/wr/cwre7zhat33dgo5f4kaggrzhykjbos6x3ww64ahisdtl4iydsusr.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# output_5 => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%_adaptive_avg_pool3d, [-1, -2, -3], True), kwargs = {})
triton_red_fused_mean_5 = async_compile.triton('triton_red_fused_mean_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.reduction(
size_hints=[8, 4096],
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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused_mean_5', '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_red_fused_mean_5(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 8
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
_tmp2 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + (4096*x0)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = _tmp2 + tmp1
_tmp2 = tl.where(rmask & xmask, tmp3, _tmp2)
tmp2 = tl.sum(_tmp2, 1)[:, None]
tmp4 = 4096.0
tmp5 = tmp2 / tmp4
tl.debug_barrier()
tl.store(in_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 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 3, 3, 3), (54, 27, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1))
assert_size_stride(primals_3, (2, ), (1, ))
assert_size_stride(primals_4, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 66, 66, 66), (574992, 287496, 4356, 66, 1), torch.float32)
# Topologically Sorted Source Nodes: [pad], Original ATen: [aten.replication_pad3d]
stream0 = get_raw_stream(0)
triton_poi_fused_replication_pad3d_0.run(primals_2, buf0, 2299968, grid=grid(2299968), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.convolution]
buf1 = extern_kernels.convolution(buf0, primals_1, stride=(2, 2, 2), padding=(0, 0, 0), dilation=(1, 1, 1), transposed=False, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 32, 32, 32), (65536, 32768, 1024, 32, 1))
buf2 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch.float32)
buf3 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch.float32)
buf4 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.native_group_norm]
triton_red_fused_native_group_norm_1.run(buf1, buf2, buf3, buf4, 32, 8192, grid=grid(32), stream=stream0)
buf5 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
buf6 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 8, 8), torch.float32)
buf8 = reinterpret_tensor(buf6, (4, 2, 1, 1), (2, 1, 1, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.native_group_norm]
triton_per_fused_native_group_norm_2.run(buf8, buf2, buf3, buf4, buf5, 8, 4, grid=grid(8), stream=stream0)
del buf2
del buf3
del buf4
buf9 = empty_strided_cuda((4, 2, 32, 32, 32), (65536, 32768, 1024, 32, 1), torch.float32)
buf10 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [output_1, output_2], Original ATen: [aten.native_group_norm, aten.leaky_relu]
triton_poi_fused_leaky_relu_native_group_norm_3.run(buf10, buf1, buf5, buf8, primals_3, primals_4, 262144, grid=grid(262144), stream=stream0)
buf11 = empty_strided_cuda((4, 2, 8, 8, 8), (1024, 512, 64, 8, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_3], Original ATen: [aten.avg_pool3d]
triton_poi_fused_avg_pool3d_4.run(buf10, buf11, 4096, grid=grid(4096), stream=stream0)
# Topologically Sorted Source Nodes: [output_4], Original ATen: [aten._adaptive_avg_pool3d]
buf12 = torch.ops.aten._adaptive_avg_pool3d.default(buf11, [16, 16, 16])
buf13 = buf12
del buf12
buf14 = empty_strided_cuda((4, 2, 1, 1, 1), (2, 1, 8, 8, 8), torch.float32)
buf15 = reinterpret_tensor(buf14, (4, 2, 1, 1, 1), (2, 1, 1, 1, 1), 0); del buf14 # reuse
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.mean]
triton_red_fused_mean_5.run(buf15, buf13, 8, 4096, grid=grid(8), stream=stream0)
del buf13
return (buf15, primals_1, primals_3, primals_4, buf0, buf1, buf5, buf8, buf10, 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((2, 2, 3, 3, 3), (54, 27, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, ), (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
class RefModel3d2(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 3, padding=1, stride=2,
padding_mode='replicate', bias=False)
self.l2 = torch.nn.GroupNorm(2, 2)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool3d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='area', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 64, 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 libdevice
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_replication_pad3d_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 2299968
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 66
x1 = xindex // 66 % 66
x2 = xindex // 4356 % 66
x3 = xindex // 287496
x4 = xindex
tmp0 = tl.load(in_ptr0 + (64 * (63 * (63 <= 0 * (0 >= -1 + x1) + (-1 +
x1) * (-1 + x1 > 0)) + (0 * (0 >= -1 + x1) + (-1 + x1) * (-1 + x1 >
0)) * (0 * (0 >= -1 + x1) + (-1 + x1) * (-1 + x1 > 0) < 63)) + 4096 *
(63 * (63 <= 0 * (0 >= -1 + x2) + (-1 + x2) * (-1 + x2 > 0)) + (0 *
(0 >= -1 + x2) + (-1 + x2) * (-1 + x2 > 0)) * (0 * (0 >= -1 + x2) +
(-1 + x2) * (-1 + x2 > 0) < 63)) + 262144 * x3 + (63 * (63 <= 0 * (
0 >= -1 + x0) + (-1 + x0) * (-1 + x0 > 0)) + (0 * (0 >= -1 + x0) +
(-1 + x0) * (-1 + x0 > 0)) * (0 * (0 >= -1 + x0) + (-1 + x0) * (-1 +
x0 > 0) < 63))), xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + x4, tmp0, xmask)
@triton.jit
def triton_red_fused_native_group_norm_1(in_ptr0, out_ptr0, out_ptr1,
out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 32
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
tmp2_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp2_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp2_weight = tl.zeros([XBLOCK, RBLOCK], tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + 8192 * x0), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp2_mean_next, tmp2_m2_next, tmp2_weight_next = (triton_helpers.
welford_reduce(tmp1, tmp2_mean, tmp2_m2, tmp2_weight, roffset == 0)
)
tmp2_mean = tl.where(rmask & xmask, tmp2_mean_next, tmp2_mean)
tmp2_m2 = tl.where(rmask & xmask, tmp2_m2_next, tmp2_m2)
tmp2_weight = tl.where(rmask & xmask, tmp2_weight_next, tmp2_weight)
tmp2_tmp, tmp3_tmp, tmp4_tmp = triton_helpers.welford(tmp2_mean,
tmp2_m2, tmp2_weight, 1)
tmp2 = tmp2_tmp[:, None]
tmp3 = tmp3_tmp[:, None]
tmp4 = tmp4_tmp[:, None]
tl.store(out_ptr0 + x0, tmp2, xmask)
tl.store(out_ptr1 + x0, tmp3, xmask)
tl.store(out_ptr2 + x0, tmp4, xmask)
@triton.jit
def triton_per_fused_native_group_norm_2(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, out_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 8
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_ptr1 + (r1 + 4 * x0), xmask, other=0.0)
tmp2 = tl.load(in_ptr2 + (r1 + 4 * x0), xmask, other=0.0)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp5 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp7 = tl.where(xmask, tmp3, 0)
tmp8 = tl.where(xmask, tmp4, 0)
tmp9 = tl.where(xmask, tmp5, 0)
tmp10, tmp11, tmp12 = triton_helpers.welford(tmp7, tmp8, tmp9, 1)
tmp13 = tmp10[:, None]
tmp14 = tmp11[:, None]
tmp12[:, None]
tmp16 = 32768.0
tmp17 = tmp14 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp20, xmask)
tl.store(out_ptr0 + x0, tmp13, xmask)
@triton.jit
def triton_poi_fused_leaky_relu_native_group_norm_3(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x4 = xindex // 32768
x1 = xindex // 32768 % 2
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_ptr1 + x4, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x1, None, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tmp9 = 0.0
tmp10 = tmp8 > tmp9
tmp11 = 0.02
tmp12 = tmp8 * tmp11
tmp13 = tl.where(tmp10, tmp8, tmp12)
tl.store(in_out_ptr0 + x3, tmp13, None)
@triton.jit
def triton_poi_fused_avg_pool3d_4(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 % 8
x1 = xindex // 8 % 8
x2 = xindex // 64
x3 = xindex
tmp0 = tl.load(in_ptr0 + (4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (32 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (33 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (1024 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (1025 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (1056 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (1057 + 4 * x0 + 128 * x1 + 4096 * x2), None,
eviction_policy='evict_last')
tmp2 = tmp1 + tmp0
tmp4 = tmp3 + tmp2
tmp6 = tmp5 + tmp4
tmp8 = tmp7 + tmp6
tmp10 = tmp9 + tmp8
tmp12 = tmp11 + tmp10
tmp14 = tmp13 + tmp12
tmp15 = 0.125
tmp16 = tmp14 * tmp15
tl.store(out_ptr0 + x3, tmp16, None)
@triton.jit
def triton_red_fused_mean_5(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK:
tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 8
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex
_tmp2 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r1 = rindex
tmp0 = tl.load(in_ptr0 + (r1 + 4096 * x0), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = _tmp2 + tmp1
_tmp2 = tl.where(rmask & xmask, tmp3, _tmp2)
tmp2 = tl.sum(_tmp2, 1)[:, None]
tmp4 = 4096.0
tmp5 = tmp2 / tmp4
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp5, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 3, 3, 3), (54, 27, 9, 3, 1))
assert_size_stride(primals_2, (4, 2, 64, 64, 64), (524288, 262144, 4096,
64, 1))
assert_size_stride(primals_3, (2,), (1,))
assert_size_stride(primals_4, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 2, 66, 66, 66), (574992, 287496, 4356,
66, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_replication_pad3d_0[grid(2299968)](primals_2, buf0,
2299968, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_2
buf1 = extern_kernels.convolution(buf0, primals_1, stride=(2, 2, 2),
padding=(0, 0, 0), dilation=(1, 1, 1), transposed=False,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 2, 32, 32, 32), (65536, 32768, 1024,
32, 1))
buf2 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch
.float32)
buf3 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch
.float32)
buf4 = empty_strided_cuda((4, 2, 1, 1, 4), (8, 4, 32, 32, 1), torch
.float32)
triton_red_fused_native_group_norm_1[grid(32)](buf1, buf2, buf3,
buf4, 32, 8192, XBLOCK=1, RBLOCK=2048, num_warps=16, num_stages=1)
buf5 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 1, 1), torch.float32)
buf6 = empty_strided_cuda((4, 2, 1, 1), (2, 1, 8, 8), torch.float32)
buf8 = reinterpret_tensor(buf6, (4, 2, 1, 1), (2, 1, 1, 1), 0)
del buf6
triton_per_fused_native_group_norm_2[grid(8)](buf8, buf2, buf3,
buf4, buf5, 8, 4, XBLOCK=8, num_warps=2, num_stages=1)
del buf2
del buf3
del buf4
buf9 = empty_strided_cuda((4, 2, 32, 32, 32), (65536, 32768, 1024,
32, 1), torch.float32)
buf10 = buf9
del buf9
triton_poi_fused_leaky_relu_native_group_norm_3[grid(262144)](buf10,
buf1, buf5, buf8, primals_3, primals_4, 262144, XBLOCK=512,
num_warps=8, num_stages=1)
buf11 = empty_strided_cuda((4, 2, 8, 8, 8), (1024, 512, 64, 8, 1),
torch.float32)
triton_poi_fused_avg_pool3d_4[grid(4096)](buf10, buf11, 4096,
XBLOCK=128, num_warps=4, num_stages=1)
buf12 = torch.ops.aten._adaptive_avg_pool3d.default(buf11, [16, 16, 16]
)
buf13 = buf12
del buf12
buf14 = empty_strided_cuda((4, 2, 1, 1, 1), (2, 1, 8, 8, 8), torch.
float32)
buf15 = reinterpret_tensor(buf14, (4, 2, 1, 1, 1), (2, 1, 1, 1, 1), 0)
del buf14
triton_red_fused_mean_5[grid(8)](buf15, buf13, 8, 4096, XBLOCK=1,
RBLOCK=2048, num_warps=16, num_stages=1)
del buf13
return (buf15, primals_1, primals_3, primals_4, buf0, buf1, buf5, buf8,
buf10, buf11)
class RefModel3d2New(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 3, padding=1, stride=2,
padding_mode='replicate', bias=False)
self.l2 = torch.nn.GroupNorm(2, 2)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool3d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
def forward(self, input_0):
primals_1 = self.l1.weight
primals_3 = self.l2.weight
primals_4 = self.l2.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
| shuohan/pytorch-layers | RefModel3d2 | false | 4,367 | [
"MIT"
] | 0 | 020846fd02d501cf477552179c19ba4b5e9a0695 | https://github.com/shuohan/pytorch-layers/tree/020846fd02d501cf477552179c19ba4b5e9a0695 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 3, padding=1, stride=2,
padding_mode='replicate', bias=False)
self.l2 = torch.nn.GroupNorm(2, 2)
self.l3 = torch.nn.LeakyReLU(0.02)
self.l4 = torch.nn.Identity()
self.l5 = torch.nn.AvgPool3d(2, stride=4)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='area', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 64, 64, 64])]
def get_init_inputs():
return []
|
RefModel3d | # 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_7/inductor_cache/l6/cl6777b5oo4mtwbmmrb56fzsbooogvalisfatsqsvwdxxvj3qtm3.py
# Topologically Sorted Source Nodes: [output, output_1], Original ATen: [aten.convolution, aten._native_batch_norm_legit]
# Source node to ATen node mapping:
# output => convolution
# output_1 => var_mean
# Graph fragment:
# %convolution : [num_users=2] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_3, %primals_1, %primals_2, [1, 1, 1], [0, 0, 0], [1, 1, 1], False, [0, 0, 0], 1), kwargs = {})
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view, [0, 2, 3, 4]), kwargs = {correction: 0, keepdim: True})
triton_red_fused__native_batch_norm_legit_convolution_0 = async_compile.triton('triton_red_fused__native_batch_norm_legit_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.reduction(
size_hints=[256, 8192],
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, 5, 6), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused__native_batch_norm_legit_convolution_0', 'mutated_arg_names': ['in_out_ptr0'], '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_red_fused__native_batch_norm_legit_convolution_0(in_out_ptr0, in_ptr0, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 256
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x4 = xindex
x1 = (xindex // 32) % 2
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp4_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp4_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp4_weight = tl.zeros([XBLOCK, RBLOCK], tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r3 = rindex
tmp0 = tl.load(in_out_ptr0 + (r3 + (8192*x4)), rmask & xmask, eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4_mean_next, tmp4_m2_next, tmp4_weight_next = triton_helpers.welford_reduce(
tmp3, tmp4_mean, tmp4_m2, tmp4_weight, roffset == 0
)
tmp4_mean = tl.where(rmask & xmask, tmp4_mean_next, tmp4_mean)
tmp4_m2 = tl.where(rmask & xmask, tmp4_m2_next, tmp4_m2)
tmp4_weight = tl.where(rmask & xmask, tmp4_weight_next, tmp4_weight)
tl.store(in_out_ptr0 + (r3 + (8192*x4)), tmp2, rmask & xmask)
tmp4_tmp, tmp5_tmp, tmp6_tmp = triton_helpers.welford(
tmp4_mean, tmp4_m2, tmp4_weight, 1
)
tmp4 = tmp4_tmp[:, None]
tmp5 = tmp5_tmp[:, None]
tmp6 = tmp6_tmp[:, None]
tl.store(out_ptr0 + (x4), tmp4, xmask)
tl.store(out_ptr1 + (x4), tmp5, xmask)
tl.store(out_ptr2 + (x4), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/n4/cn4kzmqi736xyk336mpk7353ywgjx3ahbcpi5uxs5a4wumlctiqf.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.repeat]
# Source node to ATen node mapping:
# output_1 => repeat
# Graph fragment:
# %repeat : [num_users=2] = call_function[target=torch.ops.aten.repeat.default](args = (%primals_4, [4]), kwargs = {})
triton_poi_fused_repeat_1 = async_compile.triton('triton_poi_fused_repeat_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=[8],
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_repeat_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_repeat_1(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 8
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0 % 2), xmask)
tl.store(out_ptr0 + (x0), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/3i/c3i5gjzqzfthfrw3wdjs55c2r64csamkhn3dhjmenthah5zeqbf4.py
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten._native_batch_norm_legit]
# Source node to ATen node mapping:
# output_1 => add, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view, [0, 2, 3, 4]), 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 = {})
triton_per_fused__native_batch_norm_legit_2 = async_compile.triton('triton_per_fused__native_batch_norm_legit_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=[8, 32],
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': {}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3, 4, 5, 7), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused__native_batch_norm_legit_2', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 3, '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__native_batch_norm_legit_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 8
rnumel = 32
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 = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (32*x0)), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1 + (32*x0)), xmask, other=0.0)
tmp2 = tl.load(in_ptr2 + (r1 + (32*x0)), xmask, other=0.0)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp5 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp7 = tl.where(xmask, tmp3, 0)
tmp8 = tl.where(xmask, tmp4, 0)
tmp9 = tl.where(xmask, tmp5, 0)
tmp10, tmp11, tmp12 = triton_helpers.welford(tmp7, tmp8, tmp9, 1)
tmp13 = tmp10[:, None]
tmp14 = tmp11[:, None]
tmp15 = tmp12[:, None]
tmp16 = 262144.0
tmp17 = tmp14 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tl.store(out_ptr2 + (x0), tmp20, xmask)
tl.store(out_ptr0 + (x0), tmp13, xmask)
tl.store(out_ptr1 + (x0), tmp14, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/4w/c4wucv4kdxkjlqe3aggek2cygacn5abqdwr2pb3snbbdjfio2wkt.py
# Topologically Sorted Source Nodes: [output_2], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# output_2 => relu
# Graph fragment:
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_1,), kwargs = {})
triton_poi_fused_relu_3 = async_compile.triton('triton_poi_fused_relu_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=[2097152],
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_relu_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_relu_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 2097152
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x4 = (xindex // 262144)
x1 = (xindex // 262144) % 2
tmp0 = tl.load(in_ptr0 + (x3), None)
tmp1 = tl.load(in_ptr1 + (x4), None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + (x4), None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + (x4), None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + (x1), None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 262144.0
tmp5 = tmp3 / tmp4
tmp6 = 1e-05
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp2 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + (x3), tmp15, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/r2/cr2gusnlwfmg2roftivwxqzr6sryb527df57v66vp3fasortqoiv.py
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten.add, aten.mul, aten._to_copy]
# Source node to ATen node mapping:
# output_5 => add_2, add_3, convert_element_type, convert_element_type_1, iota, mul_2, mul_3
# Graph fragment:
# %iota : [num_users=1] = call_function[target=torch.ops.prims.iota.default](args = (64,), kwargs = {start: 0, step: 1, dtype: torch.int64, device: cuda:0, requires_grad: False})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%iota, 1), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, 0), kwargs = {})
# %convert_element_type : [num_users=1] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%add_2, torch.float32), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%convert_element_type, 0.0), kwargs = {})
# %mul_3 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%add_3, 0.5), kwargs = {})
# %convert_element_type_1 : [num_users=3] = call_function[target=torch.ops.prims.convert_element_type.default](args = (%mul_3, torch.int64), kwargs = {})
triton_poi_fused__to_copy_add_arange_mul_4 = async_compile.triton('triton_poi_fused__to_copy_add_arange_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: '*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, 1), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused__to_copy_add_arange_mul_4', '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__to_copy_add_arange_mul_4(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tl.store(out_ptr0 + (x0), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/or/cornkii7keqevhdsfvavlhdtdcj3mu5qnsobbrjri3s37szjjarh.py
# Topologically Sorted Source Nodes: [output_4, output_5, output_6], Original ATen: [aten.avg_pool3d, aten._unsafe_index, aten.mean]
# Source node to ATen node mapping:
# output_4 => avg_pool3d
# output_5 => _unsafe_index
# output_6 => mean
# Graph fragment:
# %avg_pool3d : [num_users=1] = call_function[target=torch.ops.aten.avg_pool3d.default](args = (%relu, [2, 2, 2], [2, 2, 2]), kwargs = {})
# %_unsafe_index : [num_users=1] = call_function[target=torch.ops.aten._unsafe_index.Tensor](args = (%avg_pool3d, [None, None, %unsqueeze_7, %unsqueeze_6, %convert_element_type_1]), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%_unsafe_index, [-1, -2, -3], True), kwargs = {})
triton_red_fused__unsafe_index_avg_pool3d_mean_5 = async_compile.triton('triton_red_fused__unsafe_index_avg_pool3d_mean_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.reduction(
size_hints=[256, 8192],
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, 4), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_red_fused__unsafe_index_avg_pool3d_mean_5', '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_red_fused__unsafe_index_avg_pool3d_mean_5(in_ptr0, in_ptr1, out_ptr1, xnumel, rnumel, XBLOCK : tl.constexpr, RBLOCK : tl.constexpr):
xnumel = 256
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 32
x1 = (xindex // 32)
x5 = xindex
_tmp31 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r4 = (rindex // 4096)
r3 = (rindex // 64) % 64
r2 = rindex % 64
r6 = rindex
tmp0 = tl.load(in_ptr0 + (r4 + (2*x0)), rmask & xmask, eviction_policy='evict_last', other=0.0)
tmp5 = tl.load(in_ptr0 + (r3), rmask, eviction_policy='evict_last', other=0.0)
tmp9 = tl.load(in_ptr0 + (r2), rmask, eviction_policy='evict_last', other=0.0)
tmp1 = tl.full([XBLOCK, RBLOCK], 32, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp10 = tmp9 + tmp1
tmp11 = tmp9 < 0
tmp12 = tl.where(tmp11, tmp10, tmp9)
tmp13 = tl.load(in_ptr1 + ((2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (1 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = tl.load(in_ptr1 + (64 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp17 = tmp16 + tmp15
tmp18 = tl.load(in_ptr1 + (65 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp19 = tmp18 + tmp17
tmp20 = tl.load(in_ptr1 + (4096 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp21 = tmp20 + tmp19
tmp22 = tl.load(in_ptr1 + (4097 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp23 = tmp22 + tmp21
tmp24 = tl.load(in_ptr1 + (4160 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp25 = tmp24 + tmp23
tmp26 = tl.load(in_ptr1 + (4161 + (2*tmp12) + (128*tmp8) + (8192*tmp4) + (262144*x1)), rmask & xmask, eviction_policy='evict_last')
tmp27 = tmp26 + tmp25
tmp28 = 0.125
tmp29 = tmp27 * tmp28
tmp30 = tl.broadcast_to(tmp29, [XBLOCK, RBLOCK])
tmp32 = _tmp31 + tmp30
_tmp31 = tl.where(rmask & xmask, tmp32, _tmp31)
tmp31 = tl.sum(_tmp31, 1)[:, None]
tl.store(out_ptr1 + (x5), tmp31, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/vy/cvyohxn2kuwdzd4cn3xjjpoopixuttlh7p53eodyip5jjlpbejt2.py
# Topologically Sorted Source Nodes: [output_6], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# output_6 => mean
# Graph fragment:
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%_unsafe_index, [-1, -2, -3], True), kwargs = {})
triton_per_fused_mean_6 = async_compile.triton('triton_per_fused_mean_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=[8, 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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_per_fused_mean_6', '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_6(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK : tl.constexpr):
xnumel = 8
rnumel = 32
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 = tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + (32*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 = 262144.0
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, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 1, 1, 1), (2, 1, 1, 1, 1))
assert_size_stride(primals_2, (2, ), (1, ))
assert_size_stride(primals_3, (4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1))
assert_size_stride(primals_4, (2, ), (1, ))
assert_size_stride(primals_5, (2, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1, 1, 1), padding=(0, 0, 0), dilation=(1, 1, 1), transposed=False, output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1))
buf1 = buf0; del buf0 # reuse
buf3 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256, 256, 1), torch.float32)
buf4 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256, 256, 1), torch.float32)
buf5 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256, 256, 1), torch.float32)
# Topologically Sorted Source Nodes: [output, output_1], Original ATen: [aten.convolution, aten._native_batch_norm_legit]
stream0 = get_raw_stream(0)
triton_red_fused__native_batch_norm_legit_convolution_0.run(buf1, primals_2, buf3, buf4, buf5, 256, 8192, grid=grid(256), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((8, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten.repeat]
triton_poi_fused_repeat_1.run(primals_4, buf2, 8, grid=grid(8), stream=stream0)
del primals_4
buf6 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.float32)
buf7 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.float32)
buf9 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.float32)
# Topologically Sorted Source Nodes: [output_1], Original ATen: [aten._native_batch_norm_legit]
triton_per_fused__native_batch_norm_legit_2.run(buf3, buf4, buf5, buf6, buf7, buf9, 8, 32, grid=grid(8), stream=stream0)
del buf3
del buf4
buf10 = empty_strided_cuda((4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1), torch.float32)
# Topologically Sorted Source Nodes: [output_2], Original ATen: [aten.relu]
triton_poi_fused_relu_3.run(buf1, buf6, buf7, buf2, primals_5, buf10, 2097152, grid=grid(2097152), stream=stream0)
del primals_5
buf11 = empty_strided_cuda((64, ), (1, ), torch.int64)
# Topologically Sorted Source Nodes: [output_5], Original ATen: [aten.arange, aten.add, aten.mul, aten._to_copy]
triton_poi_fused__to_copy_add_arange_mul_4.run(buf11, 64, grid=grid(64), stream=stream0)
buf13 = reinterpret_tensor(buf5, (4, 2, 1, 1, 1, 32), (64, 32, 256, 256, 256, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [output_4, output_5, output_6], Original ATen: [aten.avg_pool3d, aten._unsafe_index, aten.mean]
triton_red_fused__unsafe_index_avg_pool3d_mean_5.run(buf11, buf10, buf13, 256, 8192, grid=grid(256), stream=stream0)
buf14 = reinterpret_tensor(buf7, (4, 2, 1, 1, 1), (2, 1, 8, 8, 8), 0); del buf7 # reuse
buf15 = reinterpret_tensor(buf14, (4, 2, 1, 1, 1), (2, 1, 1, 1, 1), 0); del buf14 # reuse
# Topologically Sorted Source Nodes: [output_6], Original ATen: [aten.mean]
triton_per_fused_mean_6.run(buf15, buf13, 8, 32, grid=grid(8), stream=stream0)
del buf13
return (buf15, primals_1, primals_3, buf1, buf2, reinterpret_tensor(buf9, (8, ), (1, ), 0), buf10, buf11, reinterpret_tensor(buf6, (1, 8, 1, 1, 1), (8, 1, 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((2, 2, 1, 1, 1), (2, 1, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 2, 64, 64, 64), (524288, 262144, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((2, ), (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
class RefModel3d(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm3d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout3d(0.2)
self.l5 = torch.nn.AvgPool3d(2)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='nearest', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 64, 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 libdevice
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_red_fused__native_batch_norm_legit_convolution_0(in_out_ptr0,
in_ptr0, out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK: tl.
constexpr, RBLOCK: tl.constexpr):
xnumel = 256
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x4 = xindex
x1 = xindex // 32 % 2
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp4_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp4_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp4_weight = tl.zeros([XBLOCK, RBLOCK], tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r3 = rindex
tmp0 = tl.load(in_out_ptr0 + (r3 + 8192 * x4), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp4_mean_next, tmp4_m2_next, tmp4_weight_next = (triton_helpers.
welford_reduce(tmp3, tmp4_mean, tmp4_m2, tmp4_weight, roffset == 0)
)
tmp4_mean = tl.where(rmask & xmask, tmp4_mean_next, tmp4_mean)
tmp4_m2 = tl.where(rmask & xmask, tmp4_m2_next, tmp4_m2)
tmp4_weight = tl.where(rmask & xmask, tmp4_weight_next, tmp4_weight)
tl.store(in_out_ptr0 + (r3 + 8192 * x4), tmp2, rmask & xmask)
tmp4_tmp, tmp5_tmp, tmp6_tmp = triton_helpers.welford(tmp4_mean,
tmp4_m2, tmp4_weight, 1)
tmp4 = tmp4_tmp[:, None]
tmp5 = tmp5_tmp[:, None]
tmp6 = tmp6_tmp[:, None]
tl.store(out_ptr0 + x4, tmp4, xmask)
tl.store(out_ptr1 + x4, tmp5, xmask)
tl.store(out_ptr2 + x4, tmp6, xmask)
@triton.jit
def triton_poi_fused_repeat_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 8
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0 % 2, xmask)
tl.store(out_ptr0 + x0, tmp0, xmask)
@triton.jit
def triton_per_fused__native_batch_norm_legit_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, out_ptr1, out_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 8
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, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 32 * x0), xmask, other=0.0)
tmp1 = tl.load(in_ptr1 + (r1 + 32 * x0), xmask, other=0.0)
tmp2 = tl.load(in_ptr2 + (r1 + 32 * x0), xmask, other=0.0)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp5 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp7 = tl.where(xmask, tmp3, 0)
tmp8 = tl.where(xmask, tmp4, 0)
tmp9 = tl.where(xmask, tmp5, 0)
tmp10, tmp11, tmp12 = triton_helpers.welford(tmp7, tmp8, tmp9, 1)
tmp13 = tmp10[:, None]
tmp14 = tmp11[:, None]
tmp12[:, None]
tmp16 = 262144.0
tmp17 = tmp14 / tmp16
tmp18 = 1e-05
tmp19 = tmp17 + tmp18
tmp20 = libdevice.rsqrt(tmp19)
tl.store(out_ptr2 + x0, tmp20, xmask)
tl.store(out_ptr0 + x0, tmp13, xmask)
tl.store(out_ptr1 + x0, tmp14, xmask)
@triton.jit
def triton_poi_fused_relu_3(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
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
x4 = xindex // 262144
x1 = xindex // 262144 % 2
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_ptr1 + x4, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x4, None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 262144.0
tmp5 = tmp3 / tmp4
tmp6 = 1e-05
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp2 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + x3, tmp15, None)
@triton.jit
def triton_poi_fused__to_copy_add_arange_mul_4(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 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_red_fused__unsafe_index_avg_pool3d_mean_5(in_ptr0, in_ptr1,
out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
xnumel = 256
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x0 = xindex % 32
x1 = xindex // 32
x5 = xindex
_tmp31 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r4 = rindex // 4096
r3 = rindex // 64 % 64
r2 = rindex % 64
tmp0 = tl.load(in_ptr0 + (r4 + 2 * x0), rmask & xmask,
eviction_policy='evict_last', other=0.0)
tmp5 = tl.load(in_ptr0 + r3, rmask, eviction_policy='evict_last',
other=0.0)
tmp9 = tl.load(in_ptr0 + r2, rmask, eviction_policy='evict_last',
other=0.0)
tmp1 = tl.full([XBLOCK, RBLOCK], 32, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp10 = tmp9 + tmp1
tmp11 = tmp9 < 0
tmp12 = tl.where(tmp11, tmp10, tmp9)
tmp13 = tl.load(in_ptr1 + (2 * tmp12 + 128 * tmp8 + 8192 * tmp4 +
262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr1 + (1 + 2 * tmp12 + 128 * tmp8 + 8192 * tmp4 +
262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp15 = tmp14 + tmp13
tmp16 = tl.load(in_ptr1 + (64 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp17 = tmp16 + tmp15
tmp18 = tl.load(in_ptr1 + (65 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp19 = tmp18 + tmp17
tmp20 = tl.load(in_ptr1 + (4096 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp21 = tmp20 + tmp19
tmp22 = tl.load(in_ptr1 + (4097 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp23 = tmp22 + tmp21
tmp24 = tl.load(in_ptr1 + (4160 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp25 = tmp24 + tmp23
tmp26 = tl.load(in_ptr1 + (4161 + 2 * tmp12 + 128 * tmp8 + 8192 *
tmp4 + 262144 * x1), rmask & xmask, eviction_policy='evict_last')
tmp27 = tmp26 + tmp25
tmp28 = 0.125
tmp29 = tmp27 * tmp28
tmp30 = tl.broadcast_to(tmp29, [XBLOCK, RBLOCK])
tmp32 = _tmp31 + tmp30
_tmp31 = tl.where(rmask & xmask, tmp32, _tmp31)
tmp31 = tl.sum(_tmp31, 1)[:, None]
tl.store(out_ptr1 + x5, tmp31, xmask)
@triton.jit
def triton_per_fused_mean_6(in_out_ptr0, in_ptr0, xnumel, rnumel, XBLOCK:
tl.constexpr):
xnumel = 8
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, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (r1 + 32 * 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 = 262144.0
tmp6 = tmp4 / tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (2, 2, 1, 1, 1), (2, 1, 1, 1, 1))
assert_size_stride(primals_2, (2,), (1,))
assert_size_stride(primals_3, (4, 2, 64, 64, 64), (524288, 262144, 4096,
64, 1))
assert_size_stride(primals_4, (2,), (1,))
assert_size_stride(primals_5, (2,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1, 1), padding=(0, 0, 0), dilation=(1, 1, 1), transposed=False,
output_padding=(0, 0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 2, 64, 64, 64), (524288, 262144, 4096,
64, 1))
buf1 = buf0
del buf0
buf3 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256,
256, 1), torch.float32)
buf4 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256,
256, 1), torch.float32)
buf5 = empty_strided_cuda((1, 8, 1, 1, 1, 32), (256, 32, 256, 256,
256, 1), torch.float32)
get_raw_stream(0)
triton_red_fused__native_batch_norm_legit_convolution_0[grid(256)](buf1
, primals_2, buf3, buf4, buf5, 256, 8192, XBLOCK=1, RBLOCK=2048,
num_warps=16, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((8,), (1,), torch.float32)
triton_poi_fused_repeat_1[grid(8)](primals_4, buf2, 8, XBLOCK=8,
num_warps=1, num_stages=1)
del primals_4
buf6 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.
float32)
buf7 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.
float32)
buf9 = empty_strided_cuda((1, 8, 1, 1, 1), (8, 1, 8, 8, 8), torch.
float32)
triton_per_fused__native_batch_norm_legit_2[grid(8)](buf3, buf4,
buf5, buf6, buf7, buf9, 8, 32, XBLOCK=8, num_warps=2, num_stages=1)
del buf3
del buf4
buf10 = empty_strided_cuda((4, 2, 64, 64, 64), (524288, 262144,
4096, 64, 1), torch.float32)
triton_poi_fused_relu_3[grid(2097152)](buf1, buf6, buf7, buf2,
primals_5, buf10, 2097152, XBLOCK=512, num_warps=8, num_stages=1)
del primals_5
buf11 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_add_arange_mul_4[grid(64)](buf11, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf13 = reinterpret_tensor(buf5, (4, 2, 1, 1, 1, 32), (64, 32, 256,
256, 256, 1), 0)
del buf5
triton_red_fused__unsafe_index_avg_pool3d_mean_5[grid(256)](buf11,
buf10, buf13, 256, 8192, XBLOCK=1, RBLOCK=1024, num_warps=16,
num_stages=1)
buf14 = reinterpret_tensor(buf7, (4, 2, 1, 1, 1), (2, 1, 8, 8, 8), 0)
del buf7
buf15 = reinterpret_tensor(buf14, (4, 2, 1, 1, 1), (2, 1, 1, 1, 1), 0)
del buf14
triton_per_fused_mean_6[grid(8)](buf15, buf13, 8, 32, XBLOCK=8,
num_warps=2, num_stages=1)
del buf13
return buf15, primals_1, primals_3, buf1, buf2, reinterpret_tensor(buf9,
(8,), (1,), 0), buf10, buf11, reinterpret_tensor(buf6, (1, 8, 1, 1,
1), (8, 1, 1, 1, 1), 0)
class RefModel3dNew(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm3d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout3d(0.2)
self.l5 = torch.nn.AvgPool3d(2)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
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]
| shuohan/pytorch-layers | RefModel3d | false | 4,368 | [
"MIT"
] | 0 | 020846fd02d501cf477552179c19ba4b5e9a0695 | https://github.com/shuohan/pytorch-layers/tree/020846fd02d501cf477552179c19ba4b5e9a0695 | import torch
import torch.nn.functional as F
class Model(torch.nn.Module):
"""The 3D reference model."""
def __init__(self):
super().__init__()
self.l1 = torch.nn.Conv3d(2, 2, 1, bias=True)
self.l2 = torch.nn.InstanceNorm3d(2, affine=True)
self.l3 = torch.nn.ReLU()
self.l4 = torch.nn.Dropout3d(0.2)
self.l5 = torch.nn.AvgPool3d(2)
self.l7 = torch.nn.AdaptiveAvgPool3d(1)
def forward(self, x):
output = self.l1(x)
output = self.l2(output)
output = self.l3(output)
output = self.l4(output)
output = self.l5(output)
output = F.interpolate(output, mode='nearest', scale_factor=2)
output = self.l7(output)
return output
def get_inputs():
return [torch.rand([4, 2, 64, 64, 64])]
def get_init_inputs():
return []
|
HardSwish | # 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_7/inductor_cache/jj/cjjcpa4jfom3kmx4ufnxtda3bmq466cpemkegyhzep2ymmlsg35l.py
# Topologically Sorted Source Nodes: [add, hardtanh, mul, truediv], Original ATen: [aten.add, aten.hardtanh, aten.mul, aten.div]
# Source node to ATen node mapping:
# add => add
# hardtanh => clamp_max, clamp_min
# mul => mul
# truediv => div
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%arg0_1, 3.0), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%add, 0.0), kwargs = {})
# %clamp_max : [num_users=1] = call_function[target=torch.ops.aten.clamp_max.default](args = (%clamp_min, 6.0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%arg0_1, %clamp_max), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%mul, 6.0), kwargs = {})
triton_poi_fused_add_div_hardtanh_mul_0 = async_compile.triton('triton_poi_fused_add_div_hardtanh_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: 'i32'}, 'device': 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_hardtanh_mul_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_div_hardtanh_mul_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 = 3.0
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 6.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = tmp0 * tmp6
tmp8 = 0.16666666666666666
tmp9 = tmp7 * tmp8
tl.store(out_ptr0 + (x0), tmp9, 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: [add, hardtanh, mul, truediv], Original ATen: [aten.add, aten.hardtanh, aten.mul, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_hardtanh_mul_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 HardSwish(nn.Module):
"""hardswish activation func (see MobileNetV3)"""
def __init__(self):
super(HardSwish, self).__init__()
def forward(self, x):
return x * nn.ReLU6(inplace=True)(x + 3.0) / 6.0
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
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_hardtanh_mul_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 = 3.0
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 6.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = tmp0 * tmp6
tmp8 = 0.16666666666666666
tmp9 = tmp7 * tmp8
tl.store(out_ptr0 + x0, tmp9, 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_hardtanh_mul_0[grid(256)](arg0_1, buf0,
256, XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class HardSwishNew(nn.Module):
"""hardswish activation func (see MobileNetV3)"""
def __init__(self):
super(HardSwishNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| stepbuystep/LightNAS | HardSwish | false | 4,369 | [
"Apache-2.0"
] | 0 | 030d0e13e0c85354ed711e36fc4b91b1541f95e5 | https://github.com/stepbuystep/LightNAS/tree/030d0e13e0c85354ed711e36fc4b91b1541f95e5 | import torch
import torch.nn as nn
class Model(nn.Module):
"""hardswish activation func (see MobileNetV3)"""
def __init__(self):
super().__init__()
def forward(self, x):
return x * nn.ReLU6(inplace=True)(x + 3.0) / 6.0
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
DDPGActor | # 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_7/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, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view_1, [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=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=[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_7/inductor_cache/6e/c6etrnrwnqlbwohg25wyqa2rkojloi6lu4ex24dati4k5ok6ywe4.py
# Topologically Sorted Source Nodes: [layer_norm, x], Original ATen: [aten.native_layer_norm, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# layer_norm => add, add_1, mul, mul_1, rsqrt, sub, var_mean
# x => relu
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%view_1, [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=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 = (%view_1, %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_4), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_5), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_1,), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_native_layer_norm_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_native_layer_norm_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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*fp32', 6: '*i1', 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_native_layer_norm_relu_threshold_backward_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_relu_threshold_backward_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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
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
tmp9 = tl.full([1], 0, tl.int32)
tmp10 = triton_helpers.maximum(tmp9, tmp8)
tmp11 = 0.0
tmp12 = tmp10 <= tmp11
tl.store(out_ptr0 + (x2), tmp10, xmask)
tl.store(out_ptr1 + (x2), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/4t/c4tinndbfifahpyer77rm3uc2bmrxd7ki263fv7g4fyely5r2ouh.py
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.tanh, aten.mul]
# Source node to ATen node mapping:
# x_2 => tanh
# x_3 => mul_4
# Graph fragment:
# %tanh : [num_users=1] = call_function[target=torch.ops.aten.tanh.default](args = (%view_5,), kwargs = {})
# %mul_4 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%tanh, 4), kwargs = {})
triton_poi_fused_mul_tanh_2 = async_compile.triton('triton_poi_fused_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=[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_mul_tanh_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_mul_tanh_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
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0), xmask)
tmp1 = libdevice.tanh(tmp0)
tmp2 = 4.0
tmp3 = tmp1 * 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, primals_6, primals_7, primals_8, primals_9, primals_10, primals_11 = 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, ), (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, ), (1, ))
assert_size_stride(primals_9, (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((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, 1), (16, 4, 1, 64), torch.float32)
buf2 = 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(buf0, buf1, buf2, 64, grid=grid(64), stream=stream0)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [layer_norm, x], Original ATen: [aten.native_layer_norm, aten.relu, aten.threshold_backward]
triton_poi_fused_native_layer_norm_relu_threshold_backward_1.run(buf0, buf1, buf2, primals_4, primals_5, buf3, buf11, 256, grid=grid(256), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_1], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf4)
del primals_7
buf5 = buf2; del buf2 # reuse
buf6 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [layer_norm_1], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_0.run(buf4, buf5, buf6, 64, grid=grid(64), stream=stream0)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf10 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [layer_norm_1, x_1], Original ATen: [aten.native_layer_norm, aten.relu, aten.threshold_backward]
triton_poi_fused_native_layer_norm_relu_threshold_backward_1.run(buf4, buf5, buf6, primals_8, primals_9, buf7, buf10, 256, grid=grid(256), stream=stream0)
del buf5
del buf6
del primals_9
buf8 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [linear_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_11, reinterpret_tensor(buf7, (64, 4), (4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf8)
del primals_11
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.tanh, aten.mul]
triton_poi_fused_mul_tanh_2.run(buf8, buf9, 256, grid=grid(256), stream=stream0)
return (buf9, primals_4, primals_8, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0, reinterpret_tensor(buf3, (64, 4), (4, 1), 0), buf4, reinterpret_tensor(buf7, (64, 4), (4, 1), 0), buf8, primals_10, buf10, primals_6, 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, 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)
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, ), (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)
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 numpy as np
import torch.nn as nn
import torch.nn.functional as F
def fanin_init(size, fanin=None):
"""
Initilise network weights
"""
fanin = fanin or size[0]
v = 1.0 / np.sqrt(fanin)
return torch.Tensor(size).uniform_(-v, v)
class DDPGActor(nn.Module):
"""
Pytorch neural network for Actor model
"""
def __init__(self, state_dim, action_dim, action_bound, hidden_size,
init_w=0.003):
super(DDPGActor, self).__init__()
self.action_bound = action_bound
self.l1 = nn.Linear(state_dim, hidden_size)
self.bn1 = nn.LayerNorm(hidden_size)
self.l2 = nn.Linear(hidden_size, hidden_size)
self.bn2 = nn.LayerNorm(hidden_size)
self.l3 = nn.Linear(hidden_size, action_dim)
self.init_weights(init_w)
self.device = torch.device('cuda' if torch.cuda.is_available() else
'cpu')
self
def init_weights(self, init_w):
self.l1.weight.data = fanin_init(self.l1.weight.data.size())
self.l2.weight.data = fanin_init(self.l2.weight.data.size())
self.l3.weight.data.uniform_(-init_w, init_w)
def forward(self, x):
x = F.relu(self.bn1(self.l1(x)))
x = F.relu(self.bn2(self.l2(x)))
x = torch.tanh(self.l3(x))
x = x * self.action_bound
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'state_dim': 4, 'action_dim': 4, 'action_bound': 4,
'hidden_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 libdevice
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_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_native_layer_norm_relu_threshold_backward_1(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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
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
tmp9 = tl.full([1], 0, tl.int32)
tmp10 = triton_helpers.maximum(tmp9, tmp8)
tmp11 = 0.0
tmp12 = tmp10 <= tmp11
tl.store(out_ptr0 + x2, tmp10, xmask)
tl.store(out_ptr1 + x2, tmp12, xmask)
@triton.jit
def triton_poi_fused_mul_tanh_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
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = libdevice.tanh(tmp0)
tmp2 = 4.0
tmp3 = tmp1 * tmp2
tl.store(out_ptr0 + x0, tmp3, 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, 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,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (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,))
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, 1), (16, 4, 1, 64), torch.float32)
buf2 = 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)](buf0, buf1, buf2, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_native_layer_norm_relu_threshold_backward_1[grid(256)
](buf0, buf1, buf2, primals_4, primals_5, buf3, buf11, 256,
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, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf4)
del primals_7
buf5 = buf2
del buf2
buf6 = buf1
del buf1
triton_poi_fused_native_layer_norm_0[grid(64)](buf4, buf5, buf6, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf10 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_native_layer_norm_relu_threshold_backward_1[grid(256)
](buf4, buf5, buf6, primals_8, primals_9, buf7, buf10, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del buf5
del buf6
del primals_9
buf8 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(buf7, (64, 4),
(4, 1), 0), reinterpret_tensor(primals_10, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf8)
del primals_11
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_mul_tanh_2[grid(256)](buf8, buf9, 256, XBLOCK=128,
num_warps=4, num_stages=1)
return buf9, primals_4, primals_8, reinterpret_tensor(primals_3, (64, 4
), (4, 1), 0), buf0, reinterpret_tensor(buf3, (64, 4), (4, 1), 0
), buf4, reinterpret_tensor(buf7, (64, 4), (4, 1), 0
), buf8, primals_10, buf10, primals_6, buf11
def fanin_init(size, fanin=None):
"""
Initilise network weights
"""
fanin = fanin or size[0]
v = 1.0 / np.sqrt(fanin)
return torch.Tensor(size).uniform_(-v, v)
class DDPGActorNew(nn.Module):
"""
Pytorch neural network for Actor model
"""
def __init__(self, state_dim, action_dim, action_bound, hidden_size,
init_w=0.003):
super(DDPGActorNew, self).__init__()
self.action_bound = action_bound
self.l1 = nn.Linear(state_dim, hidden_size)
self.bn1 = nn.LayerNorm(hidden_size)
self.l2 = nn.Linear(hidden_size, hidden_size)
self.bn2 = nn.LayerNorm(hidden_size)
self.l3 = nn.Linear(hidden_size, action_dim)
self.init_weights(init_w)
self.device = torch.device('cuda' if torch.cuda.is_available() else
'cpu')
self
def init_weights(self, init_w):
self.l1.weight.data = fanin_init(self.l1.weight.data.size())
self.l2.weight.data = fanin_init(self.l2.weight.data.size())
self.l3.weight.data.uniform_(-init_w, init_w)
def forward(self, input_0):
primals_1 = self.l1.weight
primals_2 = self.l1.bias
primals_4 = self.bn1.weight
primals_5 = self.bn1.bias
primals_6 = self.l2.weight
primals_7 = self.l2.bias
primals_8 = self.bn2.weight
primals_9 = self.bn2.bias
primals_10 = self.l3.weight
primals_11 = self.l3.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])
return output[0]
| Nikhil-Paleti/sawyer_analysis_reinforcement_learning | DDPGActor | false | 4,370 | [
"MIT"
] | 0 | dc774c9a162fabb98493b69d7656cb14cb37f094 | https://github.com/Nikhil-Paleti/sawyer_analysis_reinforcement_learning/tree/dc774c9a162fabb98493b69d7656cb14cb37f094 | import torch
import numpy as np
import torch.nn as nn
import torch.nn.functional as F
def fanin_init(size, fanin=None):
"""
Initilise network weights
"""
fanin = fanin or size[0]
v = 1.0 / np.sqrt(fanin)
return torch.Tensor(size).uniform_(-v, v)
class Model(nn.Module):
"""
Pytorch neural network for Actor model
"""
def __init__(self, state_dim, action_dim, action_bound, hidden_size,
init_w=0.003):
super().__init__()
self.action_bound = action_bound
self.l1 = nn.Linear(state_dim, hidden_size)
self.bn1 = nn.LayerNorm(hidden_size)
self.l2 = nn.Linear(hidden_size, hidden_size)
self.bn2 = nn.LayerNorm(hidden_size)
self.l3 = nn.Linear(hidden_size, action_dim)
self.init_weights(init_w)
self.device = torch.device('cuda' if torch.cuda.is_available() else
'cpu')
self
def init_weights(self, init_w):
self.l1.weight.data = fanin_init(self.l1.weight.data.size())
self.l2.weight.data = fanin_init(self.l2.weight.data.size())
self.l3.weight.data.uniform_(-init_w, init_w)
def forward(self, x):
x = F.relu(self.bn1(self.l1(x)))
x = F.relu(self.bn2(self.l2(x)))
x = torch.tanh(self.l3(x))
x = x * self.action_bound
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'state_dim': 4, 'action_dim': 4, 'action_bound': 4,
'hidden_size': 4}]
|
attentionLayer | # 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_7/inductor_cache/3l/c3lu4ccbjruychszpewk67ythz75gaj4rslgmbux6fatrywe7g7t.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_2
# Graph fragment:
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_5, %getitem_5), 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=[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_add_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_add_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 + (8 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ol/colbiyeeegfdyyzeckjnylgg3xt3rkh3aadcz7fjtfx5472nedsg.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_5, 1.0), kwargs = {})
triton_poi_fused_mul_1 = async_compile.triton('triton_poi_fused_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=[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_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_mul_1(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_7/inductor_cache/c5/cc5cm2utkmzhcjdhw5qgs7t254ixwfil74kthoebdprvhlljdmul.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_1
# Graph fragment:
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %getitem_4), 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=[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_add_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_add_2(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 + (4 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/7s/c7spagnqvsgjrukyw5jujzjmswxuigeuvpyhxgdob766q2gfvgzr.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_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=[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_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 = 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_7/inductor_cache/dw/cdwqsjnh2osfmjr2utzzaqdg2vrfivzkuhareq3urgidllj2bsvr.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_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=[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_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 = 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_7/inductor_cache/y5/cy5gjrtl7netbzcjhig66pdorub2vbq2qvwmv3tamld2ehimmlz7.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_9,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_5 = async_compile.triton('triton_poi_fused_clone_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=[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_5', '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_5(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_7/inductor_cache/vv/cvvbofouoyxjzf3vusx62ssynyztxp4c2lk76gekn6hxry2vdjht.py
# Topologically Sorted Source Nodes: [src_1, src_2], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# src_1 => add_3
# src_2 => clone_2
# Graph fragment:
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%permute, %squeeze), kwargs = {})
# %clone_2 : [num_users=3] = call_function[target=torch.ops.aten.clone.default](args = (%add_3,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_add_native_layer_norm_6 = async_compile.triton('triton_poi_fused_add_native_layer_norm_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, 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), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_native_layer_norm_6', '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_native_layer_norm_6(in_out_ptr0, in_ptr0, in_ptr1, 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)
tmp1 = tl.load(in_out_ptr0 + (x1 + (4*y0)), xmask & ymask)
tmp2 = tl.load(in_ptr1 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + (x1 + (4*y0)), tmp4, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/a4/ca4uo2wxiyozqosox66kdczthghpe6qlxcx3xgdhcs2i6efpvbaa.py
# Topologically Sorted Source Nodes: [src_2], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# src_2 => add_4, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%clone_2, [1]), kwargs = {correction: 0, keepdim: True})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_6, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_4,), kwargs = {})
triton_poi_fused_native_layer_norm_7 = async_compile.triton('triton_poi_fused_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=[4],
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_native_layer_norm_7', '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_7(in_ptr0, out_ptr0, out_ptr1, 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 + (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_7/inductor_cache/z2/cz2x6uoruxdkowfiawvku3nxpx4vgbvkygpozaunzozrsmzk6q75.py
# Topologically Sorted Source Nodes: [src_2], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# src_2 => add_4, add_5, mul_1, mul_2, rsqrt, sub_1, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%clone_2, [1]), kwargs = {correction: 0, keepdim: True})
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_6, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_4,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%clone_2, %getitem_7), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %rsqrt), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_1, %primals_7), kwargs = {})
# %add_5 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %primals_8), kwargs = {})
triton_poi_fused_native_layer_norm_8 = async_compile.triton('triton_poi_fused_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=[16],
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_8', '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_8(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, 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
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_7/inductor_cache/q6/cq6bj5cbehpfrboy2hqptxwsffz2qanka4axd2qviiot545u347l.py
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# relu => relu
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_10), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_9 = async_compile.triton('triton_poi_fused_relu_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_relu_9', '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_9(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 % 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)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/4e/c4escp7v3hikmzkxuyjmzdj3juzy4ty3yheqrkztwu2gvsteirte.py
# Topologically Sorted Source Nodes: [src_3], Original ATen: [aten.add]
# Source node to ATen node mapping:
# src_3 => add_6
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_12), kwargs = {})
# %add_6 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_5, %add_tensor), 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=[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_10', '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_10(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, 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, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (12, 4), (4, 1))
assert_size_stride(primals_4, (12, ), (1, ))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, ), (1, ))
assert_size_stride(primals_8, (4, ), (1, ))
assert_size_stride(primals_9, (16, 4), (4, 1))
assert_size_stride(primals_10, (16, ), (1, ))
assert_size_stride(primals_11, (4, 16), (16, 1))
assert_size_stride(primals_12, (4, ), (1, ))
assert_size_stride(primals_13, (4, ), (1, ))
assert_size_stride(primals_14, (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: [multi_head_attention_forward], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_3, (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.mm]
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), reinterpret_tensor(primals_3, (4, 4), (1, 4), 16), out=buf1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), reinterpret_tensor(primals_3, (4, 4), (1, 4), 32), out=buf2)
del primals_3
buf3 = reinterpret_tensor(buf2, (4, 1, 4), (4, 4, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_0.run(buf3, primals_4, 16, grid=grid(16), stream=stream0)
buf4 = 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]
triton_poi_fused_mul_1.run(buf4, primals_4, 16, grid=grid(16), stream=stream0)
buf5 = reinterpret_tensor(buf1, (4, 1, 4), (4, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
triton_poi_fused_add_2.run(buf5, primals_4, 16, grid=grid(16), stream=stream0)
del primals_4
buf6 = 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(buf4, reinterpret_tensor(buf5, (4, 1, 4), (1, 0, 4), 0), out=buf6)
buf7 = 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_3.run(buf6, buf7, 64, grid=grid(64), stream=stream0)
buf8 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
triton_poi_fused__softmax_4.run(buf7, buf8, 64, grid=grid(64), stream=stream0)
buf9 = 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(buf8, reinterpret_tensor(buf3, (4, 4, 1), (1, 4, 0), 0), out=buf9)
buf10 = 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_5.run(buf9, buf10, 4, 4, grid=grid(4, 4), stream=stream0)
buf11 = reinterpret_tensor(buf9, (4, 4), (4, 1), 0); del buf9 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf10, (4, 4), (4, 1), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf11)
buf12 = buf11; del buf11 # reuse
# Topologically Sorted Source Nodes: [src_1, src_2], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_6.run(buf12, primals_1, primals_6, 4, 4, grid=grid(4, 4), stream=stream0)
del primals_6
buf13 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
buf14 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
# Topologically Sorted Source Nodes: [src_2], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_7.run(buf12, buf13, buf14, 4, grid=grid(4), stream=stream0)
buf15 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [src_2], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_8.run(buf12, buf13, buf14, primals_7, primals_8, buf15, 16, grid=grid(16), stream=stream0)
del primals_8
buf16 = reinterpret_tensor(buf7, (4, 16), (16, 1), 0); del buf7 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf15, reinterpret_tensor(primals_9, (4, 16), (1, 4), 0), out=buf16)
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
triton_poi_fused_relu_9.run(buf17, primals_10, 64, grid=grid(64), stream=stream0)
del primals_10
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf17, reinterpret_tensor(primals_11, (16, 4), (1, 16), 0), out=buf18)
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [src_3], Original ATen: [aten.add]
triton_poi_fused_add_10.run(buf19, buf15, primals_12, 16, grid=grid(16), stream=stream0)
del primals_12
buf20 = buf14; del buf14 # reuse
buf21 = buf13; del buf13 # reuse
# Topologically Sorted Source Nodes: [src_4], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_7.run(buf19, buf20, buf21, 4, grid=grid(4), stream=stream0)
buf22 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [src_4], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_8.run(buf19, buf20, buf21, primals_13, primals_14, buf22, 16, grid=grid(16), stream=stream0)
del buf20
del buf21
del primals_14
return (reinterpret_tensor(buf22, (4, 4), (1, 4), 0), primals_7, primals_13, reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), buf8, reinterpret_tensor(buf10, (4, 4), (4, 1), 0), buf12, buf15, buf17, buf19, primals_11, primals_9, primals_5, reinterpret_tensor(buf3, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf4, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf5, (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((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((12, 4), (4, 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), (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)
primals_8 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((16, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((16, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, 16), (16, 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, ), (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 torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import MultiheadAttention
from itertools import product as product
class attentionLayer(nn.Module):
def __init__(self, d_model, nhead, dropout=0.1):
super(attentionLayer, self).__init__()
self.self_attn = MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, d_model * 4)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(d_model * 4, d_model)
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.activation = F.relu
def forward(self, src, tar):
src = src.transpose(0, 1)
tar = tar.transpose(0, 1)
src2 = self.self_attn(tar, src, src, attn_mask=None,
key_padding_mask=None)[0]
src = src + self.dropout1(src2)
src = self.norm1(src)
src2 = self.linear2(self.dropout(self.activation(self.linear1(src))))
src = src + self.dropout2(src2)
src = self.norm2(src)
src = src.transpose(0, 1)
return src
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'d_model': 4, 'nhead': 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.nn.functional as F
from torch.nn import MultiheadAttention
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
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_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 + (8 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused_mul_1(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_add_2(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 + (4 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused__softmax_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
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_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
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_5(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_native_layer_norm_6(in_out_ptr0, in_ptr0, in_ptr1,
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)
tmp1 = tl.load(in_out_ptr0 + (x1 + 4 * y0), xmask & ymask)
tmp2 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tmp0 + tmp3
tl.store(in_out_ptr0 + (x1 + 4 * y0), tmp4, xmask & ymask)
@triton.jit
def triton_poi_fused_native_layer_norm_7(in_ptr0, out_ptr0, out_ptr1,
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 + 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_8(in_ptr0, in_ptr1, in_ptr2, in_ptr3,
in_ptr4, 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
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_relu_9(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 % 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)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
@triton.jit
def triton_poi_fused_add_10(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, 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, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (12, 4), (4, 1))
assert_size_stride(primals_4, (12,), (1,))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4,), (1,))
assert_size_stride(primals_9, (16, 4), (4, 1))
assert_size_stride(primals_10, (16,), (1,))
assert_size_stride(primals_11, (4, 16), (16, 1))
assert_size_stride(primals_12, (4,), (1,))
assert_size_stride(primals_13, (4,), (1,))
assert_size_stride(primals_14, (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_2, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), out=buf0)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_3, (4, 4), (1, 4), 16), out=buf1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_3, (4, 4), (1, 4), 32), out=buf2)
del primals_3
buf3 = reinterpret_tensor(buf2, (4, 1, 4), (4, 4, 1), 0)
del buf2
get_raw_stream(0)
triton_poi_fused_add_0[grid(16)](buf3, primals_4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf4 = reinterpret_tensor(buf0, (4, 4, 1), (1, 4, 16), 0)
del buf0
triton_poi_fused_mul_1[grid(16)](buf4, primals_4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf5 = reinterpret_tensor(buf1, (4, 1, 4), (4, 4, 1), 0)
del buf1
triton_poi_fused_add_2[grid(16)](buf5, primals_4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_4
buf6 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf4, reinterpret_tensor(buf5, (4, 1, 4), (1, 0,
4), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_3[grid(64)](buf6, buf7, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf8 = buf6
del buf6
triton_poi_fused__softmax_4[grid(64)](buf7, buf8, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf9 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
extern_kernels.bmm(buf8, reinterpret_tensor(buf3, (4, 4, 1), (1, 4,
0), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
triton_poi_fused_clone_5[grid(4, 4)](buf9, buf10, 4, 4, XBLOCK=4,
YBLOCK=4, num_warps=1, num_stages=1)
buf11 = reinterpret_tensor(buf9, (4, 4), (4, 1), 0)
del buf9
extern_kernels.mm(reinterpret_tensor(buf10, (4, 4), (4, 1), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf11)
buf12 = buf11
del buf11
triton_poi_fused_add_native_layer_norm_6[grid(4, 4)](buf12,
primals_1, primals_6, 4, 4, XBLOCK=4, YBLOCK=4, num_warps=1,
num_stages=1)
del primals_6
buf13 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
buf14 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
triton_poi_fused_native_layer_norm_7[grid(4)](buf12, buf13, buf14,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf15 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_native_layer_norm_8[grid(16)](buf12, buf13, buf14,
primals_7, primals_8, buf15, 16, XBLOCK=16, num_warps=1,
num_stages=1)
del primals_8
buf16 = reinterpret_tensor(buf7, (4, 16), (16, 1), 0)
del buf7
extern_kernels.mm(buf15, reinterpret_tensor(primals_9, (4, 16), (1,
4), 0), out=buf16)
buf17 = buf16
del buf16
triton_poi_fused_relu_9[grid(64)](buf17, primals_10, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_10
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf17, reinterpret_tensor(primals_11, (16, 4), (1,
16), 0), out=buf18)
buf19 = buf18
del buf18
triton_poi_fused_add_10[grid(16)](buf19, buf15, primals_12, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_12
buf20 = buf14
del buf14
buf21 = buf13
del buf13
triton_poi_fused_native_layer_norm_7[grid(4)](buf19, buf20, buf21,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf22 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_native_layer_norm_8[grid(16)](buf19, buf20, buf21,
primals_13, primals_14, buf22, 16, XBLOCK=16, num_warps=1,
num_stages=1)
del buf20
del buf21
del primals_14
return (reinterpret_tensor(buf22, (4, 4), (1, 4), 0), primals_7,
primals_13, reinterpret_tensor(primals_2, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), buf8,
reinterpret_tensor(buf10, (4, 4), (4, 1), 0), buf12, buf15, buf17,
buf19, primals_11, primals_9, primals_5, reinterpret_tensor(buf3, (
4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf4, (4, 1, 4), (1, 1,
4), 0), reinterpret_tensor(buf5, (4, 4, 1), (1, 4, 1), 0))
class attentionLayerNew(nn.Module):
def __init__(self, d_model, nhead, dropout=0.1):
super(attentionLayerNew, self).__init__()
self.self_attn = MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, d_model * 4)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(d_model * 4, d_model)
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.activation = F.relu
def forward(self, input_0, input_1):
primals_3 = self.self_attn.in_proj_weight
primals_4 = self.self_attn.in_proj_bias
primals_1 = self.self_attn.out_proj.weight
primals_6 = self.self_attn.out_proj.bias
primals_9 = self.linear1.weight
primals_10 = self.linear1.bias
primals_11 = self.linear2.weight
primals_7 = self.linear2.bias
primals_8 = self.norm1.weight
primals_12 = self.norm1.bias
primals_13 = self.norm2.weight
primals_14 = self.norm2.bias
primals_2 = input_0
primals_5 = input_1
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]
| slapshin/TalkNet_ASD | attentionLayer | false | 4,371 | [
"MIT"
] | 0 | 343fac5c8d2bef2b98244e3acf20ac322711a4c7 | https://github.com/slapshin/TalkNet_ASD/tree/343fac5c8d2bef2b98244e3acf20ac322711a4c7 | import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.nn import MultiheadAttention
from itertools import product as product
class Model(nn.Module):
def __init__(self, d_model, nhead, dropout=0.1):
super().__init__()
self.self_attn = MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, d_model * 4)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(d_model * 4, d_model)
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.activation = F.relu
def forward(self, src, tar):
src = src.transpose(0, 1)
tar = tar.transpose(0, 1)
src2 = self.self_attn(tar, src, src, attn_mask=None,
key_padding_mask=None)[0]
src = src + self.dropout1(src2)
src = self.norm1(src)
src2 = self.linear2(self.dropout(self.activation(self.linear1(src))))
src = src + self.dropout2(src2)
src = self.norm2(src)
src = src.transpose(0, 1)
return src
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [4, 4]
|
BananaResNet | # 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_7/inductor_cache/sz/cszoyuuj4zvzkzs2nd4mplxzdwf2wbqaaytgyv47lszb3x4lea74.py
# Topologically Sorted Source Nodes: [skip], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# skip => relu
# Graph fragment:
# %relu : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%view_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=[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_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 = 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_7/inductor_cache/cm/ccmmvfrzj6ezqdqbwxjnljmvjeabooeq6tnlretadgfpzikha52e.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.relu, aten.add, aten.threshold_backward]
# Source node to ATen node mapping:
# x => relu_1
# x_1 => add
# Graph fragment:
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_3,), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%relu_1, %relu), kwargs = {})
# %le_4 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_1, 0), kwargs = {})
# %le_5 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_add_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=[8192],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*i1', 5: '*i1', 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_add_relu_threshold_backward_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_relu_threshold_backward_1(in_ptr0, in_ptr1, in_ptr2, out_ptr0, out_ptr1, out_ptr2, 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_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr1 + (x0), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + (x2), None)
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = tmp4 + tmp5
tmp7 = 0.0
tmp8 = tmp4 <= tmp7
tmp9 = tmp5 <= tmp7
tl.store(out_ptr0 + (x2), tmp6, None)
tl.store(out_ptr1 + (x2), tmp8, None)
tl.store(out_ptr2 + (x2), tmp9, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/cl/cclfbafil76viay3vxdhvp7r7wsplzuz252u2rpv64iobshvbg3m.py
# Topologically Sorted Source Nodes: [skip_1], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# skip_1 => relu_2
# Graph fragment:
# %relu_2 : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%view_5,), kwargs = {})
triton_poi_fused_relu_2 = async_compile.triton('triton_poi_fused_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=[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_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_relu_2(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_7/inductor_cache/5j/c5jc62gahidjmezpxmrjc6lrnd4jn2aro26ecgbhmwvknukgqnl7.py
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_2 => relu_3
# Graph fragment:
# %relu_3 : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%view_7,), kwargs = {})
# %le_2 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_3, 0), kwargs = {})
triton_poi_fused_relu_threshold_backward_3 = async_compile.triton('triton_poi_fused_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=[4096],
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_relu_threshold_backward_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_relu_threshold_backward_3(in_ptr0, in_ptr1, 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)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr1 + (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(out_ptr0 + (x2), tmp4, None)
tl.store(out_ptr1 + (x2), tmp6, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/iy/ciytefnlsncbtwx7g3hdl73hgupqxyzoel5pn3jbocp3bu6vxbyi.py
# Topologically Sorted Source Nodes: [x_2, skip_2, x_3, skip_3], Original ATen: [aten.relu, aten.add, aten.threshold_backward]
# Source node to ATen node mapping:
# skip_2 => add_1
# skip_3 => add_2
# x_2 => relu_3
# x_3 => relu_4
# Graph fragment:
# %relu_3 : [num_users=3] = call_function[target=torch.ops.aten.relu.default](args = (%view_7,), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%relu_3, %relu_2), kwargs = {})
# %relu_4 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_9,), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%relu_4, %add_1), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_4, 0), kwargs = {})
# %le_3 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_2, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_4 = async_compile.triton('triton_poi_fused_add_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=[4096],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*i1', 6: '*i1', 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_relu_threshold_backward_4', '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_relu_threshold_backward_4(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, in_ptr3, 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)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + (x2), None)
tmp1 = tl.load(in_ptr1 + (x0), None, eviction_policy='evict_last')
tmp5 = tl.load(in_out_ptr0 + (x2), None)
tmp6 = tl.load(in_ptr2 + (x0), None, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr3 + (x2), None)
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp7 = tmp5 + tmp6
tmp8 = triton_helpers.maximum(tmp3, tmp7)
tmp10 = tmp8 + tmp9
tmp11 = tmp4 + tmp10
tmp12 = 0.0
tmp13 = tmp4 <= tmp12
tmp14 = tmp9 <= tmp12
tl.store(in_out_ptr0 + (x2), tmp11, None)
tl.store(out_ptr0 + (x2), tmp13, None)
tl.store(out_ptr1 + (x2), tmp14, None)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/av/cavoncpszthddtfpaezjs7pph3sxcwx6vh2ostnzsxg3qw75y7gp.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# x_4 => relu_5
# Graph fragment:
# %relu_5 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%view_11,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_5, 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=[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_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 = 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')
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, (128, 4), (4, 1))
assert_size_stride(primals_2, (128, ), (1, ))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 128), (128, 1))
assert_size_stride(primals_5, (128, ), (1, ))
assert_size_stride(primals_6, (64, 128), (128, 1))
assert_size_stride(primals_7, (64, ), (1, ))
assert_size_stride(primals_8, (64, 64), (64, 1))
assert_size_stride(primals_9, (64, ), (1, ))
assert_size_stride(primals_10, (64, 64), (64, 1))
assert_size_stride(primals_11, (64, ), (1, ))
assert_size_stride(primals_12, (32, 64), (64, 1))
assert_size_stride(primals_13, (32, ), (1, ))
assert_size_stride(primals_14, (4, 32), (32, 1))
assert_size_stride(primals_15, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 128), (128, 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, 128), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 128), (2048, 512, 128, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [skip], Original ATen: [aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_relu_0.run(buf1, primals_2, 8192, grid=grid(8192), stream=stream0)
del primals_2
buf2 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf1, (64, 128), (128, 1), 0), reinterpret_tensor(primals_4, (128, 128), (1, 128), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1), torch.float32)
buf17 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1), torch.bool)
buf18 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1), torch.bool)
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.relu, aten.add, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_1.run(buf2, primals_5, buf1, buf3, buf17, buf18, 8192, grid=grid(8192), stream=stream0)
del buf2
del primals_5
buf4 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf3, (64, 128), (128, 1), 0), reinterpret_tensor(primals_6, (128, 64), (1, 128), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 64), (1024, 256, 64, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [skip_1], Original ATen: [aten.relu]
triton_poi_fused_relu_2.run(buf5, primals_7, 4096, grid=grid(4096), stream=stream0)
del primals_7
buf6 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf5, (64, 64), (64, 1), 0), reinterpret_tensor(primals_8, (64, 64), (1, 64), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch.float32)
buf15 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_3.run(buf6, primals_9, buf7, buf15, 4096, grid=grid(4096), stream=stream0)
buf8 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf7, (64, 64), (64, 1), 0), reinterpret_tensor(primals_10, (64, 64), (1, 64), 0), out=buf8)
buf9 = reinterpret_tensor(buf6, (4, 4, 4, 64), (1024, 256, 64, 1), 0); del buf6 # reuse
buf14 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch.bool)
buf16 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_2, skip_2, x_3, skip_3], Original ATen: [aten.relu, aten.add, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_4.run(buf9, buf8, primals_11, primals_9, buf5, buf14, buf16, 4096, grid=grid(4096), stream=stream0)
del buf8
del primals_11
del primals_9
buf10 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf9, (64, 64), (64, 1), 0), reinterpret_tensor(primals_12, (64, 32), (1, 64), 0), out=buf10)
buf11 = reinterpret_tensor(buf10, (4, 4, 4, 32), (512, 128, 32, 1), 0); del buf10 # reuse
buf13 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.relu, aten.threshold_backward]
triton_poi_fused_relu_threshold_backward_5.run(buf11, primals_13, buf13, 2048, grid=grid(2048), stream=stream0)
del primals_13
buf12 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_5], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_15, reinterpret_tensor(buf11, (64, 32), (32, 1), 0), reinterpret_tensor(primals_14, (32, 4), (1, 32), 0), alpha=1, beta=1, out=buf12)
del primals_15
return (reinterpret_tensor(buf12, (4, 4, 4, 4), (64, 16, 4, 1), 0), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), reinterpret_tensor(buf1, (64, 128), (128, 1), 0), reinterpret_tensor(buf3, (64, 128), (128, 1), 0), reinterpret_tensor(buf5, (64, 64), (64, 1), 0), reinterpret_tensor(buf7, (64, 64), (64, 1), 0), reinterpret_tensor(buf9, (64, 64), (64, 1), 0), reinterpret_tensor(buf11, (64, 32), (32, 1), 0), primals_14, buf13, primals_12, buf14, primals_10, buf15, primals_8, buf16, primals_6, buf17, primals_4, buf18, )
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((128, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((128, ), (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((128, 128), (128, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 128), (128, 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), (64, 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), (64, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((32, 64), (64, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((32, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((4, 32), (32, 1), device='cuda:0', dtype=torch.float32)
primals_15 = 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])
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 BananaResNet(nn.Module):
def __init__(self, state_size, action_size):
super(BananaResNet, self).__init__()
self.blk1fc1 = nn.Linear(state_size, 128)
self.blk1fc2 = nn.Linear(128, 128)
self.blk1fc3 = nn.Linear(128, 64)
self.blk2fc1 = nn.Linear(64, 64)
self.blk2fc2 = nn.Linear(64, 64)
self.blk2fc3 = nn.Linear(64, 32)
self.outfc = nn.Linear(32, action_size)
def forward(self, state):
skip = F.relu(self.blk1fc1(state))
x = F.relu(self.blk1fc2(skip))
x = x + skip
skip = F.relu(self.blk1fc3(x))
x = F.relu(self.blk2fc1(skip))
skip = x + skip
x = F.relu(self.blk2fc2(x))
skip = x + skip
x = F.relu(self.blk2fc3(skip))
x = self.outfc(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'state_size': 4, 'action_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
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
):
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_add_relu_threshold_backward_1(in_ptr0, in_ptr1,
in_ptr2, out_ptr0, out_ptr1, out_ptr2, 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_ptr0 + x2, None)
tmp1 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr2 + x2, None)
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = tmp4 + tmp5
tmp7 = 0.0
tmp8 = tmp4 <= tmp7
tmp9 = tmp5 <= tmp7
tl.store(out_ptr0 + x2, tmp6, None)
tl.store(out_ptr1 + x2, tmp8, None)
tl.store(out_ptr2 + x2, tmp9, None)
@triton.jit
def triton_poi_fused_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)
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_relu_threshold_backward_3(in_ptr0, in_ptr1, 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)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + x2, None)
tmp1 = tl.load(in_ptr1 + 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(out_ptr0 + x2, tmp4, None)
tl.store(out_ptr1 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_add_relu_threshold_backward_4(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, in_ptr3, 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)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_ptr0 + x2, None)
tmp1 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp5 = tl.load(in_out_ptr0 + x2, None)
tmp6 = tl.load(in_ptr2 + x0, None, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr3 + x2, None)
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp7 = tmp5 + tmp6
tmp8 = triton_helpers.maximum(tmp3, tmp7)
tmp10 = tmp8 + tmp9
tmp11 = tmp4 + tmp10
tmp12 = 0.0
tmp13 = tmp4 <= tmp12
tmp14 = tmp9 <= tmp12
tl.store(in_out_ptr0 + x2, tmp11, None)
tl.store(out_ptr0 + x2, tmp13, None)
tl.store(out_ptr1 + x2, tmp14, None)
@triton.jit
def triton_poi_fused_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)
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)
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, (128, 4), (4, 1))
assert_size_stride(primals_2, (128,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 128), (128, 1))
assert_size_stride(primals_5, (128,), (1,))
assert_size_stride(primals_6, (64, 128), (128, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (64, 64), (64, 1))
assert_size_stride(primals_9, (64,), (1,))
assert_size_stride(primals_10, (64, 64), (64, 1))
assert_size_stride(primals_11, (64,), (1,))
assert_size_stride(primals_12, (32, 64), (64, 1))
assert_size_stride(primals_13, (32,), (1,))
assert_size_stride(primals_14, (4, 32), (32, 1))
assert_size_stride(primals_15, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 128), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 128), (2048, 512, 128, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_relu_0[grid(8192)](buf1, primals_2, 8192, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 128), (128, 1), 0),
reinterpret_tensor(primals_4, (128, 128), (1, 128), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.float32)
buf17 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.bool)
buf18 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.bool)
triton_poi_fused_add_relu_threshold_backward_1[grid(8192)](buf2,
primals_5, buf1, buf3, buf17, buf18, 8192, XBLOCK=256,
num_warps=4, num_stages=1)
del buf2
del primals_5
buf4 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 128), (128, 1), 0),
reinterpret_tensor(primals_6, (128, 64), (1, 128), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 64), (1024, 256, 64, 1), 0)
del buf4
triton_poi_fused_relu_2[grid(4096)](buf5, primals_7, 4096, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf5, (64, 64), (64, 1), 0),
reinterpret_tensor(primals_8, (64, 64), (1, 64), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch.
float32)
buf15 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch
.bool)
triton_poi_fused_relu_threshold_backward_3[grid(4096)](buf6,
primals_9, buf7, buf15, 4096, XBLOCK=256, num_warps=4, num_stages=1
)
buf8 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (64, 64), (64, 1), 0),
reinterpret_tensor(primals_10, (64, 64), (1, 64), 0), out=buf8)
buf9 = reinterpret_tensor(buf6, (4, 4, 4, 64), (1024, 256, 64, 1), 0)
del buf6
buf14 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch
.bool)
buf16 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch
.bool)
triton_poi_fused_add_relu_threshold_backward_4[grid(4096)](buf9,
buf8, primals_11, primals_9, buf5, buf14, buf16, 4096, XBLOCK=
256, num_warps=4, num_stages=1)
del buf8
del primals_11
del primals_9
buf10 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf9, (64, 64), (64, 1), 0),
reinterpret_tensor(primals_12, (64, 32), (1, 64), 0), out=buf10)
buf11 = reinterpret_tensor(buf10, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf10
buf13 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool
)
triton_poi_fused_relu_threshold_backward_5[grid(2048)](buf11,
primals_13, buf13, 2048, XBLOCK=128, num_warps=4, num_stages=1)
del primals_13
buf12 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_15, reinterpret_tensor(buf11, (64, 32),
(32, 1), 0), reinterpret_tensor(primals_14, (32, 4), (1, 32), 0
), alpha=1, beta=1, out=buf12)
del primals_15
return (reinterpret_tensor(buf12, (4, 4, 4, 4), (64, 16, 4, 1), 0),
reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(buf1, (64, 128), (128, 1), 0),
reinterpret_tensor(buf3, (64, 128), (128, 1), 0),
reinterpret_tensor(buf5, (64, 64), (64, 1), 0), reinterpret_tensor(
buf7, (64, 64), (64, 1), 0), reinterpret_tensor(buf9, (64, 64), (64,
1), 0), reinterpret_tensor(buf11, (64, 32), (32, 1), 0), primals_14,
buf13, primals_12, buf14, primals_10, buf15, primals_8, buf16,
primals_6, buf17, primals_4, buf18)
class BananaResNetNew(nn.Module):
def __init__(self, state_size, action_size):
super(BananaResNetNew, self).__init__()
self.blk1fc1 = nn.Linear(state_size, 128)
self.blk1fc2 = nn.Linear(128, 128)
self.blk1fc3 = nn.Linear(128, 64)
self.blk2fc1 = nn.Linear(64, 64)
self.blk2fc2 = nn.Linear(64, 64)
self.blk2fc3 = nn.Linear(64, 32)
self.outfc = nn.Linear(32, action_size)
def forward(self, input_0):
primals_1 = self.blk1fc1.weight
primals_2 = self.blk1fc1.bias
primals_4 = self.blk1fc2.weight
primals_5 = self.blk1fc2.bias
primals_6 = self.blk1fc3.weight
primals_7 = self.blk1fc3.bias
primals_8 = self.blk2fc1.weight
primals_9 = self.blk2fc1.bias
primals_10 = self.blk2fc2.weight
primals_11 = self.blk2fc2.bias
primals_12 = self.blk2fc3.weight
primals_13 = self.blk2fc3.bias
primals_14 = self.outfc.weight
primals_15 = self.outfc.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]
| slash-fury/DRL-Navigation | BananaResNet | false | 4,372 | [
"MIT"
] | 0 | 5989dca62590b611ab39ac8722a22d897c65cc88 | https://github.com/slash-fury/DRL-Navigation/tree/5989dca62590b611ab39ac8722a22d897c65cc88 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, state_size, action_size):
super().__init__()
self.blk1fc1 = nn.Linear(state_size, 128)
self.blk1fc2 = nn.Linear(128, 128)
self.blk1fc3 = nn.Linear(128, 64)
self.blk2fc1 = nn.Linear(64, 64)
self.blk2fc2 = nn.Linear(64, 64)
self.blk2fc3 = nn.Linear(64, 32)
self.outfc = nn.Linear(32, action_size)
def forward(self, state):
skip = F.relu(self.blk1fc1(state))
x = F.relu(self.blk1fc2(skip))
x = x + skip
skip = F.relu(self.blk1fc3(x))
x = F.relu(self.blk2fc1(skip))
skip = x + skip
x = F.relu(self.blk2fc2(x))
skip = x + skip
x = F.relu(self.blk2fc3(skip))
x = self.outfc(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
HardSigmoid | # 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_7/inductor_cache/gl/cgljna3wfarubemgd6d2p3bgazvfhdxtrcu7luu5yza3rrfkty2s.py
# Topologically Sorted Source Nodes: [add, hardtanh, truediv], Original ATen: [aten.add, aten.hardtanh, aten.div]
# Source node to ATen node mapping:
# add => add
# hardtanh => clamp_max, clamp_min
# truediv => div
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%arg0_1, 3.0), kwargs = {})
# %clamp_min : [num_users=1] = call_function[target=torch.ops.aten.clamp_min.default](args = (%add, 0.0), kwargs = {})
# %clamp_max : [num_users=1] = call_function[target=torch.ops.aten.clamp_max.default](args = (%clamp_min, 6.0), kwargs = {})
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%clamp_max, 6.0), kwargs = {})
triton_poi_fused_add_div_hardtanh_0 = async_compile.triton('triton_poi_fused_add_div_hardtanh_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_hardtanh_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_div_hardtanh_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 = 3.0
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 6.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = 0.16666666666666666
tmp8 = tmp6 * tmp7
tl.store(out_ptr0 + (x0), 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: [add, hardtanh, truediv], Original ATen: [aten.add, aten.hardtanh, aten.div]
stream0 = get_raw_stream(0)
triton_poi_fused_add_div_hardtanh_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 HardSigmoid(nn.Module):
"""hardsigmoid activation func used in squeeze-and-excitation module (see MobileNetV3)"""
def __init__(self):
super(HardSigmoid, self).__init__()
def forward(self, x):
return nn.ReLU6(inplace=True)(x + 3.0) / 6.0
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
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_hardtanh_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 = 3.0
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 6.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = 0.16666666666666666
tmp8 = tmp6 * tmp7
tl.store(out_ptr0 + x0, 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_add_div_hardtanh_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class HardSigmoidNew(nn.Module):
"""hardsigmoid activation func used in squeeze-and-excitation module (see MobileNetV3)"""
def __init__(self):
super(HardSigmoidNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| stepbuystep/LightNAS | HardSigmoid | false | 4,373 | [
"Apache-2.0"
] | 0 | 030d0e13e0c85354ed711e36fc4b91b1541f95e5 | https://github.com/stepbuystep/LightNAS/tree/030d0e13e0c85354ed711e36fc4b91b1541f95e5 | import torch
import torch.nn as nn
class Model(nn.Module):
"""hardsigmoid activation func used in squeeze-and-excitation module (see MobileNetV3)"""
def __init__(self):
super().__init__()
def forward(self, x):
return nn.ReLU6(inplace=True)(x + 3.0) / 6.0
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
HeatmapLoss | # 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_7/inductor_cache/xf/cxf2mveni2qzuqw5oi7mwzm5vuj4a63unfx2dikmtlxpnubv3xi5.py
# Topologically Sorted Source Nodes: [sub, l, mean, mean_1], Original ATen: [aten.sub, aten.pow, aten.mean]
# Source node to ATen node mapping:
# l => pow_1
# mean => mean
# mean_1 => mean_1
# sub => sub
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %arg1_1), kwargs = {})
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%sub, 2), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%pow_1, [3]), kwargs = {})
# %mean_1 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%mean, [2]), kwargs = {})
triton_poi_fused_mean_pow_sub_0 = async_compile.triton('triton_poi_fused_mean_pow_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=[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_mean_pow_sub_0', 'mutated_arg_names': [], 'no_x_dim': False, 'num_load': 32, '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_mean_pow_sub_0(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 + (16*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (16*x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (1 + (16*x0)), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + (1 + (16*x0)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (2 + (16*x0)), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (2 + (16*x0)), xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr0 + (3 + (16*x0)), xmask, eviction_policy='evict_last')
tmp15 = tl.load(in_ptr1 + (3 + (16*x0)), xmask, eviction_policy='evict_last')
tmp21 = tl.load(in_ptr0 + (4 + (16*x0)), xmask, eviction_policy='evict_last')
tmp22 = tl.load(in_ptr1 + (4 + (16*x0)), xmask, eviction_policy='evict_last')
tmp25 = tl.load(in_ptr0 + (5 + (16*x0)), xmask, eviction_policy='evict_last')
tmp26 = tl.load(in_ptr1 + (5 + (16*x0)), xmask, eviction_policy='evict_last')
tmp30 = tl.load(in_ptr0 + (6 + (16*x0)), xmask, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr1 + (6 + (16*x0)), xmask, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr0 + (7 + (16*x0)), xmask, eviction_policy='evict_last')
tmp36 = tl.load(in_ptr1 + (7 + (16*x0)), xmask, eviction_policy='evict_last')
tmp42 = tl.load(in_ptr0 + (8 + (16*x0)), xmask, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr1 + (8 + (16*x0)), xmask, eviction_policy='evict_last')
tmp46 = tl.load(in_ptr0 + (9 + (16*x0)), xmask, eviction_policy='evict_last')
tmp47 = tl.load(in_ptr1 + (9 + (16*x0)), xmask, eviction_policy='evict_last')
tmp51 = tl.load(in_ptr0 + (10 + (16*x0)), xmask, eviction_policy='evict_last')
tmp52 = tl.load(in_ptr1 + (10 + (16*x0)), xmask, eviction_policy='evict_last')
tmp56 = tl.load(in_ptr0 + (11 + (16*x0)), xmask, eviction_policy='evict_last')
tmp57 = tl.load(in_ptr1 + (11 + (16*x0)), xmask, eviction_policy='evict_last')
tmp63 = tl.load(in_ptr0 + (12 + (16*x0)), xmask, eviction_policy='evict_last')
tmp64 = tl.load(in_ptr1 + (12 + (16*x0)), xmask, eviction_policy='evict_last')
tmp67 = tl.load(in_ptr0 + (13 + (16*x0)), xmask, eviction_policy='evict_last')
tmp68 = tl.load(in_ptr1 + (13 + (16*x0)), xmask, eviction_policy='evict_last')
tmp72 = tl.load(in_ptr0 + (14 + (16*x0)), xmask, eviction_policy='evict_last')
tmp73 = tl.load(in_ptr1 + (14 + (16*x0)), xmask, eviction_policy='evict_last')
tmp77 = tl.load(in_ptr0 + (15 + (16*x0)), xmask, eviction_policy='evict_last')
tmp78 = tl.load(in_ptr1 + (15 + (16*x0)), xmask, eviction_policy='evict_last')
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
tmp19 = 4.0
tmp20 = tmp18 / tmp19
tmp23 = tmp21 - tmp22
tmp24 = tmp23 * tmp23
tmp27 = tmp25 - tmp26
tmp28 = tmp27 * tmp27
tmp29 = tmp24 + tmp28
tmp32 = tmp30 - tmp31
tmp33 = tmp32 * tmp32
tmp34 = tmp29 + tmp33
tmp37 = tmp35 - tmp36
tmp38 = tmp37 * tmp37
tmp39 = tmp34 + tmp38
tmp40 = tmp39 / tmp19
tmp41 = tmp20 + tmp40
tmp44 = tmp42 - tmp43
tmp45 = tmp44 * tmp44
tmp48 = tmp46 - tmp47
tmp49 = tmp48 * tmp48
tmp50 = tmp45 + tmp49
tmp53 = tmp51 - tmp52
tmp54 = tmp53 * tmp53
tmp55 = tmp50 + tmp54
tmp58 = tmp56 - tmp57
tmp59 = tmp58 * tmp58
tmp60 = tmp55 + tmp59
tmp61 = tmp60 / tmp19
tmp62 = tmp41 + tmp61
tmp65 = tmp63 - tmp64
tmp66 = tmp65 * tmp65
tmp69 = tmp67 - tmp68
tmp70 = tmp69 * tmp69
tmp71 = tmp66 + tmp70
tmp74 = tmp72 - tmp73
tmp75 = tmp74 * tmp74
tmp76 = tmp71 + tmp75
tmp79 = tmp77 - tmp78
tmp80 = tmp79 * tmp79
tmp81 = tmp76 + tmp80
tmp82 = tmp81 / tmp19
tmp83 = tmp62 + tmp82
tmp84 = tmp83 / tmp19
tl.store(out_ptr0 + (x0), tmp84, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/35/c35v4ojil7vwot3cvbaaxwst2htsi5dh2y4o4d7fr4ma3zmlulqt.py
# Topologically Sorted Source Nodes: [l_1], Original ATen: [aten.mean]
# Source node to ATen node mapping:
# l_1 => mean_2
# Graph fragment:
# %mean_2 : [num_users=1] = call_function[target=torch.ops.aten.mean.dim](args = (%mean_1, [1]), kwargs = {})
triton_poi_fused_mean_1 = async_compile.triton('triton_poi_fused_mean_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: 'i32'}, 'device': 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_mean_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_mean_1(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 + (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
tl.store(out_ptr0 + (x0), 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, 1), torch.float32)
# Topologically Sorted Source Nodes: [sub, l, mean, mean_1], Original ATen: [aten.sub, aten.pow, aten.mean]
stream0 = get_raw_stream(0)
triton_poi_fused_mean_pow_sub_0.run(arg0_1, arg1_1, buf0, 16, grid=grid(16), stream=stream0)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4, ), (1, ), torch.float32)
# Topologically Sorted Source Nodes: [l_1], Original ATen: [aten.mean]
triton_poi_fused_mean_1.run(buf0, buf1, 4, grid=grid(4), 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.utils.data
class HeatmapLoss(torch.nn.Module):
"""
loss for detection heatmap
"""
def __init__(self):
super(HeatmapLoss, self).__init__()
def forward(self, pred, gt):
l = (pred - gt) ** 2
l = l.mean(dim=3).mean(dim=2).mean(dim=1)
return l
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.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_mean_pow_sub_0(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 + 16 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 16 * x0, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (1 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp5 = tl.load(in_ptr1 + (1 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp9 = tl.load(in_ptr0 + (2 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp10 = tl.load(in_ptr1 + (2 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp14 = tl.load(in_ptr0 + (3 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp15 = tl.load(in_ptr1 + (3 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp21 = tl.load(in_ptr0 + (4 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr1 + (4 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr0 + (5 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp26 = tl.load(in_ptr1 + (5 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp30 = tl.load(in_ptr0 + (6 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp31 = tl.load(in_ptr1 + (6 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp35 = tl.load(in_ptr0 + (7 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp36 = tl.load(in_ptr1 + (7 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp42 = tl.load(in_ptr0 + (8 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp43 = tl.load(in_ptr1 + (8 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp46 = tl.load(in_ptr0 + (9 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp47 = tl.load(in_ptr1 + (9 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp51 = tl.load(in_ptr0 + (10 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp52 = tl.load(in_ptr1 + (10 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp56 = tl.load(in_ptr0 + (11 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp57 = tl.load(in_ptr1 + (11 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp63 = tl.load(in_ptr0 + (12 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp64 = tl.load(in_ptr1 + (12 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp67 = tl.load(in_ptr0 + (13 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp68 = tl.load(in_ptr1 + (13 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp72 = tl.load(in_ptr0 + (14 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp73 = tl.load(in_ptr1 + (14 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp77 = tl.load(in_ptr0 + (15 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp78 = tl.load(in_ptr1 + (15 + 16 * x0), xmask, eviction_policy=
'evict_last')
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
tmp19 = 4.0
tmp20 = tmp18 / tmp19
tmp23 = tmp21 - tmp22
tmp24 = tmp23 * tmp23
tmp27 = tmp25 - tmp26
tmp28 = tmp27 * tmp27
tmp29 = tmp24 + tmp28
tmp32 = tmp30 - tmp31
tmp33 = tmp32 * tmp32
tmp34 = tmp29 + tmp33
tmp37 = tmp35 - tmp36
tmp38 = tmp37 * tmp37
tmp39 = tmp34 + tmp38
tmp40 = tmp39 / tmp19
tmp41 = tmp20 + tmp40
tmp44 = tmp42 - tmp43
tmp45 = tmp44 * tmp44
tmp48 = tmp46 - tmp47
tmp49 = tmp48 * tmp48
tmp50 = tmp45 + tmp49
tmp53 = tmp51 - tmp52
tmp54 = tmp53 * tmp53
tmp55 = tmp50 + tmp54
tmp58 = tmp56 - tmp57
tmp59 = tmp58 * tmp58
tmp60 = tmp55 + tmp59
tmp61 = tmp60 / tmp19
tmp62 = tmp41 + tmp61
tmp65 = tmp63 - tmp64
tmp66 = tmp65 * tmp65
tmp69 = tmp67 - tmp68
tmp70 = tmp69 * tmp69
tmp71 = tmp66 + tmp70
tmp74 = tmp72 - tmp73
tmp75 = tmp74 * tmp74
tmp76 = tmp71 + tmp75
tmp79 = tmp77 - tmp78
tmp80 = tmp79 * tmp79
tmp81 = tmp76 + tmp80
tmp82 = tmp81 / tmp19
tmp83 = tmp62 + tmp82
tmp84 = tmp83 / tmp19
tl.store(out_ptr0 + x0, tmp84, xmask)
@triton.jit
def triton_poi_fused_mean_1(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 + 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
tl.store(out_ptr0 + x0, 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, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mean_pow_sub_0[grid(16)](arg0_1, arg1_1, buf0, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del arg0_1
del arg1_1
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
triton_poi_fused_mean_1[grid(4)](buf0, buf1, 4, XBLOCK=4, num_warps
=1, num_stages=1)
del buf0
return buf1,
class HeatmapLossNew(torch.nn.Module):
"""
loss for detection heatmap
"""
def __init__(self):
super(HeatmapLossNew, 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]
| seeinggreen/pyslr | HeatmapLoss | false | 4,374 | [
"BSD-3-Clause"
] | 0 | 17009582f70aed09a9174ce47f9414f715173018 | https://github.com/seeinggreen/pyslr/tree/17009582f70aed09a9174ce47f9414f715173018 | import torch
import torch.utils.data
class Model(torch.nn.Module):
"""
loss for detection heatmap
"""
def __init__(self):
super().__init__()
def forward(self, pred, gt):
l = (pred - gt) ** 2
l = l.mean(dim=3).mean(dim=2).mean(dim=1)
return l
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
GCN | # 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_7/inductor_cache/rp/crpg2xwsfquznkqzvnyazfnl5y4eqbe55cxzbx4lgyyat3i355e3.py
# Topologically Sorted Source Nodes: [add, x], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# add => add
# x => relu
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_1, %primals_4), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_0 = async_compile.triton('triton_poi_fused_add_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_add_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_add_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')
# kernel path: runs/run_shard_7/inductor_cache/ul/culvxc5xcnacfjypzxghwcyc2445sqsz25ci4rib6axjxs3fv3so.py
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# log_softmax => amax, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%addmm_default, [1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%addmm_default, %amax), kwargs = {})
triton_poi_fused__log_softmax_1 = async_compile.triton('triton_poi_fused__log_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: 'i32'}, 'device': 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_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__log_softmax_1(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
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
tl.store(out_ptr0 + (x2), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/yr/cyr6fatjcqc5np3quy6arljtkkff4qjmueyb5b4pk5xvkxgrzuvd.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=[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__log_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__log_softmax_2(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
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (4*x1), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + (4*x1)), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + (4*x1)), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + (4*x1)), xmask, 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
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 = 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, 4), (4, 1))
assert_size_stride(primals_6, (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: [support], Original ATen: [aten.mm]
extern_kernels.mm(primals_2, primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [output], Original ATen: [aten.mm]
extern_kernels.mm(primals_3, buf0, out=buf1)
buf2 = buf1; del buf1 # reuse
buf10 = empty_strided_cuda((4, 4), (4, 1), torch.bool)
# Topologically Sorted Source Nodes: [add, x], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_add_relu_threshold_backward_0.run(buf2, primals_4, buf10, 16, grid=grid(16), stream=stream0)
del primals_4
# Topologically Sorted Source Nodes: [add, x, x_1], Original ATen: [aten.add, aten.relu, aten.native_dropout]
buf3 = torch.ops.aten.native_dropout.default(buf2, 0.5, True)
buf4 = buf3[0]
buf5 = buf3[1]
del buf3
buf6 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [support_1], Original ATen: [aten.mm]
extern_kernels.mm(buf4, primals_5, out=buf6)
buf7 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.addmm(primals_6, primals_3, buf6, alpha=1, beta=1, out=buf7)
del primals_6
buf8 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
triton_poi_fused__log_softmax_1.run(buf7, buf8, 16, grid=grid(16), stream=stream0)
buf9 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
triton_poi_fused__log_softmax_2.run(buf8, buf9, 16, grid=grid(16), stream=stream0)
del buf8
return (buf9, buf5, buf9, reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), reinterpret_tensor(buf4, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), buf10, reinterpret_tensor(primals_2, (4, 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, 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, 4), (4, 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)
| from torch.nn import Module
import math
import torch
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
import torch.nn as nn
import torch.nn.functional as F
class GCLayer(Module):
def __init__(self, dim_in, dim_out):
super(GCLayer, self).__init__()
self.dim_in = dim_in
self.dim_out = dim_out
self.weight = Parameter(torch.FloatTensor(self.dim_in, self.dim_out))
self.bias = Parameter(torch.FloatTensor(self.dim_out))
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
return output + self.bias
class GCN(nn.Module):
def __init__(self, features, hidden, classes, dropout, layers=2):
super(GCN, self).__init__()
self.gc1 = GCLayer(features, hidden)
self.gc2 = GCLayer(hidden, classes)
self.gc3 = None
self.dropout = dropout
if layers == 3:
self.gc2 = GCLayer(hidden, hidden)
self.gc3 = GCLayer(hidden, classes)
def forward(self, x, adj):
x = F.relu(self.gc1(x, adj))
x = F.dropout(x, self.dropout)
x = self.gc2(x, adj)
if self.gc3 is not None:
x = self.gc3(x, adj)
return F.log_softmax(x, dim=1)
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'features': 4, 'hidden': 4, 'classes': 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 import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch.nn import Module
import math
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
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_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)
@triton.jit
def triton_poi_fused__log_softmax_1(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
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
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_softmax_2(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
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, 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
tl.store(out_ptr0 + x2, tmp13, 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, 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, 4), (4, 1))
assert_size_stride(primals_6, (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_2, primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_3, buf0, out=buf1)
buf2 = buf1
del buf1
buf10 = empty_strided_cuda((4, 4), (4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_add_relu_threshold_backward_0[grid(16)](buf2,
primals_4, buf10, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_4
buf3 = torch.ops.aten.native_dropout.default(buf2, 0.5, True)
buf4 = buf3[0]
buf5 = buf3[1]
del buf3
buf6 = buf2
del buf2
extern_kernels.mm(buf4, primals_5, out=buf6)
buf7 = buf0
del buf0
extern_kernels.addmm(primals_6, primals_3, buf6, alpha=1, beta=1,
out=buf7)
del primals_6
buf8 = buf6
del buf6
triton_poi_fused__log_softmax_1[grid(16)](buf7, buf8, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf9 = buf7
del buf7
triton_poi_fused__log_softmax_2[grid(16)](buf8, buf9, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf8
return buf9, buf5, buf9, reinterpret_tensor(primals_3, (4, 4), (1, 4), 0
), reinterpret_tensor(buf4, (4, 4), (1, 4), 0), reinterpret_tensor(
primals_5, (4, 4), (1, 4), 0), buf10, reinterpret_tensor(primals_2,
(4, 4), (1, 4), 0)
class GCLayer(Module):
def __init__(self, dim_in, dim_out):
super(GCLayer, self).__init__()
self.dim_in = dim_in
self.dim_out = dim_out
self.weight = Parameter(torch.FloatTensor(self.dim_in, self.dim_out))
self.bias = Parameter(torch.FloatTensor(self.dim_out))
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
return output + self.bias
class GCNNew(nn.Module):
def __init__(self, features, hidden, classes, dropout, layers=2):
super(GCNNew, self).__init__()
self.gc1 = GCLayer(features, hidden)
self.gc2 = GCLayer(hidden, classes)
self.gc3 = None
self.dropout = dropout
if layers == 3:
self.gc2 = GCLayer(hidden, hidden)
self.gc3 = GCLayer(hidden, classes)
def forward(self, input_0, input_1):
primals_1 = self.gc1.weight
primals_4 = self.gc1.bias
primals_2 = self.gc2.weight
primals_6 = self.gc2.bias
primals_3 = input_0
primals_5 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| spacemanidol/CS512DM | GCN | false | 4,375 | [
"MIT"
] | 0 | fa664ceb7526e27b9cccd372b65b15c587095c49 | https://github.com/spacemanidol/CS512DM/tree/fa664ceb7526e27b9cccd372b65b15c587095c49 | from torch.nn import Module
import math
import torch
from torch.nn.parameter import Parameter
from torch.nn.modules.module import Module
import torch.nn as nn
import torch.nn.functional as F
class GCLayer(Module):
def __init__(self, dim_in, dim_out):
super().__init__()
self.dim_in = dim_in
self.dim_out = dim_out
self.weight = Parameter(torch.FloatTensor(self.dim_in, self.dim_out))
self.bias = Parameter(torch.FloatTensor(self.dim_out))
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
return output + self.bias
class Model(nn.Module):
def __init__(self, features, hidden, classes, dropout, layers=2):
super().__init__()
self.gc1 = GCLayer(features, hidden)
self.gc2 = GCLayer(hidden, classes)
self.gc3 = None
self.dropout = dropout
if layers == 3:
self.gc2 = GCLayer(hidden, hidden)
self.gc3 = GCLayer(hidden, classes)
def forward(self, x, adj):
x = F.relu(self.gc1(x, adj))
x = F.dropout(x, self.dropout)
x = self.gc2(x, adj)
if self.gc3 is not None:
x = self.gc3(x, adj)
return F.log_softmax(x, dim=1)
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [4, 4, 4, 0.5]
|
DilatedResConv | # 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_7/inductor_cache/fk/cfkcunh3plyysuvib63zgkougyqv2ia22pa4qcifvxy3tij7w7nx.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_3, 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=[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_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 = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = (xindex // 4)
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')
# kernel path: runs/run_shard_7/inductor_cache/np/cnpq67ju2jjvmhdlii5rqv3ajv3tl7ugd3lald4s6jzn2wy4gvbv.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_4), kwargs = {})
triton_poi_fused_add_1 = async_compile.triton('triton_poi_fused_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=[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_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_1(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
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = tl.load(in_out_ptr0 + (x2), xmask)
tmp2 = tl.load(in_ptr1 + (x1), 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, (4, 4, 3), (12, 3, 1))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (4, 4, 1), (4, 1, 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(reinterpret_tensor(primals_1, (1, 4, 4), (16, 4, 1), 0), primals_2, stride=(1,), padding=(1,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf0, (1, 4, 4), (16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4), (4, 1), 0); del buf0 # reuse
buf4 = empty_strided_cuda((4, 4), (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_3, buf4, 16, grid=grid(16), stream=stream0)
del primals_3
# Topologically Sorted Source Nodes: [x_2], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (1, 4, 4), (0, 4, 1), 0), primals_4, stride=(1,), padding=(0,), dilation=(1,), transposed=False, output_padding=(0,), groups=1, bias=None)
assert_size_stride(buf2, (1, 4, 4), (16, 4, 1))
buf3 = reinterpret_tensor(buf2, (4, 4), (4, 1), 0); del buf2 # reuse
# Topologically Sorted Source Nodes: [add], Original ATen: [aten.add]
triton_poi_fused_add_1.run(buf3, primals_1, primals_5, 16, grid=grid(16), stream=stream0)
del primals_5
return (buf3, primals_2, primals_4, reinterpret_tensor(primals_1, (1, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf1, (1, 4, 4), (16, 4, 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), (4, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((4, 4, 3), (12, 3, 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), (4, 1, 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.functional as F
class DilatedResConv(nn.Module):
def __init__(self, channels, dilation=1, activation='relu', padding=1,
kernel_size=3, left_pad=0):
super().__init__()
in_channels = channels
if activation == 'relu':
self.activation = lambda *args, **kwargs: F.relu(*args, **
kwargs, inplace=True)
elif activation == 'tanh':
self.activation = F.tanh
elif activation == 'glu':
self.activation = F.glu
in_channels = channels // 2
self.left_pad = left_pad
self.dilated_conv = nn.Conv1d(in_channels, channels, kernel_size=
kernel_size, stride=1, padding=dilation * padding, dilation=
dilation, bias=True)
self.conv_1x1 = nn.Conv1d(in_channels, channels, kernel_size=1,
bias=True)
def forward(self, input):
x = input
if self.left_pad > 0:
x = F.pad(x, (self.left_pad, 0))
x = self.dilated_conv(x)
x = self.activation(x)
x = self.conv_1x1(x)
return input + x
def get_inputs():
return [torch.rand([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 import triton_helpers
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
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 = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
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)
@triton.jit
def triton_poi_fused_add_1(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
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_out_ptr0 + x2, xmask)
tmp2 = tl.load(in_ptr1 + x1, 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, (4, 4, 3), (12, 3, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_1, (1,
4, 4), (16, 4, 1), 0), primals_2, stride=(1,), padding=(1,),
dilation=(1,), transposed=False, output_padding=(0,), groups=1,
bias=None)
assert_size_stride(buf0, (1, 4, 4), (16, 4, 1))
buf1 = reinterpret_tensor(buf0, (4, 4), (4, 1), 0)
del buf0
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(16)](buf1,
primals_3, buf4, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_3
buf2 = extern_kernels.convolution(reinterpret_tensor(buf1, (1, 4, 4
), (0, 4, 1), 0), primals_4, stride=(1,), padding=(0,),
dilation=(1,), transposed=False, output_padding=(0,), groups=1,
bias=None)
assert_size_stride(buf2, (1, 4, 4), (16, 4, 1))
buf3 = reinterpret_tensor(buf2, (4, 4), (4, 1), 0)
del buf2
triton_poi_fused_add_1[grid(16)](buf3, primals_1, primals_5, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_5
return buf3, primals_2, primals_4, reinterpret_tensor(primals_1, (1, 4,
4), (16, 4, 1), 0), reinterpret_tensor(buf1, (1, 4, 4), (16, 4, 1), 0
), buf4
class DilatedResConvNew(nn.Module):
def __init__(self, channels, dilation=1, activation='relu', padding=1,
kernel_size=3, left_pad=0):
super().__init__()
in_channels = channels
if activation == 'relu':
self.activation = lambda *args, **kwargs: F.relu(*args, **
kwargs, inplace=True)
elif activation == 'tanh':
self.activation = F.tanh
elif activation == 'glu':
self.activation = F.glu
in_channels = channels // 2
self.left_pad = left_pad
self.dilated_conv = nn.Conv1d(in_channels, channels, kernel_size=
kernel_size, stride=1, padding=dilation * padding, dilation=
dilation, bias=True)
self.conv_1x1 = nn.Conv1d(in_channels, channels, kernel_size=1,
bias=True)
def forward(self, input_0):
primals_2 = self.dilated_conv.weight
primals_3 = self.dilated_conv.bias
primals_4 = self.conv_1x1.weight
primals_5 = self.conv_1x1.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| soumyac1999/instrumental-music-translation | DilatedResConv | false | 4,376 | [
"MIT"
] | 0 | f0d5edfdf34ef7bc9b329c426089f61d3468caa8 | https://github.com/soumyac1999/instrumental-music-translation/tree/f0d5edfdf34ef7bc9b329c426089f61d3468caa8 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, channels, dilation=1, activation='relu', padding=1,
kernel_size=3, left_pad=0):
super().__init__()
in_channels = channels
if activation == 'relu':
self.activation = lambda *args, **kwargs: F.relu(*args, **
kwargs, inplace=True)
elif activation == 'tanh':
self.activation = F.tanh
elif activation == 'glu':
self.activation = F.glu
in_channels = channels // 2
self.left_pad = left_pad
self.dilated_conv = nn.Conv1d(in_channels, channels, kernel_size=
kernel_size, stride=1, padding=dilation * padding, dilation=
dilation, bias=True)
self.conv_1x1 = nn.Conv1d(in_channels, channels, kernel_size=1,
bias=True)
def forward(self, input):
x = input
if self.left_pad > 0:
x = F.pad(x, (self.left_pad, 0))
x = self.dilated_conv(x)
x = self.activation(x)
x = self.conv_1x1(x)
return input + x
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [4]
|
VitMlpHead | # 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_7/inductor_cache/yy/cyya3js6wt64vdji3sfisvrqyfvqxwkwqq5mzg5bqjl2crzjs4t3.py
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# x_1 => 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_7/inductor_cache/2g/c2gw7362i2a6wsfdx2sxyywx4o6ronjg6goebvdn44w6gpjsxpbc.py
# Topologically Sorted Source Nodes: [x_1, x_2], Original ATen: [aten.add, aten.tanh]
# Source node to ATen node mapping:
# x_1 => add
# x_2 => tanh
# Graph fragment:
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_1, %primals_3), kwargs = {})
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%add,), kwargs = {})
triton_poi_fused_add_tanh_1 = async_compile.triton('triton_poi_fused_add_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: 'i32'}, 'device': 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_tanh_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_add_tanh_1(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')
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, 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, ))
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: [x_1], 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, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf0, (16, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf1)
del primals_2
buf2 = reinterpret_tensor(buf1, (4, 4, 4), (16, 4, 1), 0); del buf1 # reuse
# Topologically Sorted Source Nodes: [x_1, x_2], Original ATen: [aten.add, aten.tanh]
triton_poi_fused_add_tanh_1.run(buf2, primals_3, 64, grid=grid(64), stream=stream0)
del primals_3
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_3], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_5, reinterpret_tensor(buf2, (16, 4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf3)
del primals_5
return (reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf0, (16, 4), (4, 1), 0), buf2, 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, 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)
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
def get_args():
parser = argparse.ArgumentParser()
group = parser.add_argument_group(title='input data')
group.add_argument('--input', type=str, required=True, help=
'Path to input JSON')
group.add_argument('--json-keys', nargs='+', default=['text'], help=
'space separate listed of keys to extract from json')
group.add_argument('--split-sentences', action='store_true', help=
'Split documents into sentences.')
group.add_argument('--keep-newlines', action='store_true', help=
'Keep newlines between sentences when splitting.')
group = parser.add_argument_group(title='tokenizer')
group.add_argument('--tokenizer-type', type=str, required=True, choices
=['BertWordPieceLowerCase', 'BertWordPieceCase', 'GPT2BPETokenizer'
], help='What type of tokenizer to use.')
group.add_argument('--vocab-file', type=str, default=None, help=
'Path to the vocab file')
group.add_argument('--merge-file', type=str, default=None, help=
'Path to the BPE merge file (if necessary).')
group.add_argument('--append-eod', action='store_true', help=
'Append an <eod> token to the end of a document.')
group = parser.add_argument_group(title='output data')
group.add_argument('--output-prefix', type=str, required=True, help=
'Path to binary output file without suffix')
group.add_argument('--dataset-impl', type=str, default='mmap', choices=
['lazy', 'cached', 'mmap'])
group = parser.add_argument_group(title='runtime')
group.add_argument('--workers', type=int, default=1, help=
'Number of worker processes to launch')
group.add_argument('--log-interval', type=int, default=100, help=
'Interval between progress updates')
args = parser.parse_args()
args.keep_empty = False
if args.tokenizer_type.lower().startswith('bert'
) and not args.split_sentences:
None
args.rank = 0
args.make_vocab_size_divisible_by = 128
args.tensor_model_parallel_size = 1
return args
def init_method_normal(sigma):
"""Init method based on N(0, sigma)."""
def init_(tensor):
return torch.nn.init.normal_(tensor, mean=0.0, std=sigma)
return init_
class MegatronModule(torch.nn.Module):
"""Megatron specific extensions of torch Module with support
for pipelining."""
def __init__(self, share_word_embeddings=True):
super(MegatronModule, self).__init__()
self.share_word_embeddings = share_word_embeddings
def state_dict_for_save_checkpoint(self, destination=None, prefix='',
keep_vars=False):
"""Use this function to override the state dict for
saving checkpoints."""
return self.state_dict(destination, prefix, keep_vars)
def word_embeddings_weight(self):
if mpu.is_pipeline_first_stage(ignore_virtual=True):
return self.language_model.embedding.word_embeddings.weight
if mpu.is_pipeline_last_stage(ignore_virtual=True):
if not self.share_word_embeddings:
raise Exception(
'word_embeddings_weight() called for last stage, but share_word_embeddings is false'
)
return self.word_embeddings.weight
raise Exception(
'word_embeddings_weight() should be called for first and last stage only'
)
def initialize_word_embeddings(self, init_method_normal):
args = get_args()
if not self.share_word_embeddings:
raise Exception(
'initialize_word_embeddings() was called but share_word_embeddings is false'
)
if args.pipeline_model_parallel_size == 1:
return
if mpu.is_pipeline_last_stage():
assert not mpu.is_pipeline_first_stage()
self._word_embeddings_for_head_key = 'word_embeddings_for_head'
self.word_embeddings = mpu.VocabParallelEmbedding(args.
padded_vocab_size, args.hidden_size, init_method=
init_method_normal(args.init_method_std))
self.word_embeddings.weight.data.fill_(0)
self.word_embeddings.weight.shared = True
if torch.distributed.is_initialized():
if mpu.is_pipeline_first_stage() or mpu.is_pipeline_last_stage():
torch.distributed.all_reduce(self.word_embeddings_weight().
data, group=mpu.get_embedding_group())
else:
None
class VitMlpHead(MegatronModule):
"""Pooler layer.
Pool hidden states of a specific token (for example start of the
sequence) and add a linear transformation followed by a tanh.
Arguments:
hidden_size: hidden size
init_method: weight initialization method for the linear layer.
bias is set to zero.
"""
def __init__(self, hidden_size, num_classes):
super(VitMlpHead, self).__init__()
self.dense_in = torch.nn.Linear(hidden_size, hidden_size)
self.dense_out = torch.nn.Linear(hidden_size, num_classes)
torch.nn.init.constant_(self.dense_out.bias, -10)
def forward(self, hidden_states, sequence_index=0):
x = hidden_states[:, sequence_index, :]
x = self.dense_in(x)
x = torch.tanh(x)
x = self.dense_out(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'hidden_size': 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._inductor.runtime.triton_helpers import libdevice
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_add_tanh_1(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)
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, 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,))
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, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf1)
del primals_2
buf2 = reinterpret_tensor(buf1, (4, 4, 4), (16, 4, 1), 0)
del buf1
triton_poi_fused_add_tanh_1[grid(64)](buf2, primals_3, 64, XBLOCK=
64, num_warps=1, num_stages=1)
del primals_3
buf3 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf2, (16, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf3)
del primals_5
return reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(buf0, (16, 4), (4, 1), 0), buf2, primals_4
def get_args():
parser = argparse.ArgumentParser()
group = parser.add_argument_group(title='input data')
group.add_argument('--input', type=str, required=True, help=
'Path to input JSON')
group.add_argument('--json-keys', nargs='+', default=['text'], help=
'space separate listed of keys to extract from json')
group.add_argument('--split-sentences', action='store_true', help=
'Split documents into sentences.')
group.add_argument('--keep-newlines', action='store_true', help=
'Keep newlines between sentences when splitting.')
group = parser.add_argument_group(title='tokenizer')
group.add_argument('--tokenizer-type', type=str, required=True, choices
=['BertWordPieceLowerCase', 'BertWordPieceCase', 'GPT2BPETokenizer'
], help='What type of tokenizer to use.')
group.add_argument('--vocab-file', type=str, default=None, help=
'Path to the vocab file')
group.add_argument('--merge-file', type=str, default=None, help=
'Path to the BPE merge file (if necessary).')
group.add_argument('--append-eod', action='store_true', help=
'Append an <eod> token to the end of a document.')
group = parser.add_argument_group(title='output data')
group.add_argument('--output-prefix', type=str, required=True, help=
'Path to binary output file without suffix')
group.add_argument('--dataset-impl', type=str, default='mmap', choices=
['lazy', 'cached', 'mmap'])
group = parser.add_argument_group(title='runtime')
group.add_argument('--workers', type=int, default=1, help=
'Number of worker processes to launch')
group.add_argument('--log-interval', type=int, default=100, help=
'Interval between progress updates')
args = parser.parse_args()
args.keep_empty = False
if args.tokenizer_type.lower().startswith('bert'
) and not args.split_sentences:
None
args.rank = 0
args.make_vocab_size_divisible_by = 128
args.tensor_model_parallel_size = 1
return args
def init_method_normal(sigma):
"""Init method based on N(0, sigma)."""
def init_(tensor):
return torch.nn.init.normal_(tensor, mean=0.0, std=sigma)
return init_
class MegatronModule(torch.nn.Module):
"""Megatron specific extensions of torch Module with support
for pipelining."""
def __init__(self, share_word_embeddings=True):
super(MegatronModule, self).__init__()
self.share_word_embeddings = share_word_embeddings
def state_dict_for_save_checkpoint(self, destination=None, prefix='',
keep_vars=False):
"""Use this function to override the state dict for
saving checkpoints."""
return self.state_dict(destination, prefix, keep_vars)
def word_embeddings_weight(self):
if mpu.is_pipeline_first_stage(ignore_virtual=True):
return self.language_model.embedding.word_embeddings.weight
if mpu.is_pipeline_last_stage(ignore_virtual=True):
if not self.share_word_embeddings:
raise Exception(
'word_embeddings_weight() called for last stage, but share_word_embeddings is false'
)
return self.word_embeddings.weight
raise Exception(
'word_embeddings_weight() should be called for first and last stage only'
)
def initialize_word_embeddings(self, init_method_normal):
args = get_args()
if not self.share_word_embeddings:
raise Exception(
'initialize_word_embeddings() was called but share_word_embeddings is false'
)
if args.pipeline_model_parallel_size == 1:
return
if mpu.is_pipeline_last_stage():
assert not mpu.is_pipeline_first_stage()
self._word_embeddings_for_head_key = 'word_embeddings_for_head'
self.word_embeddings = mpu.VocabParallelEmbedding(args.
padded_vocab_size, args.hidden_size, init_method=
init_method_normal(args.init_method_std))
self.word_embeddings.weight.data.fill_(0)
self.word_embeddings.weight.shared = True
if torch.distributed.is_initialized():
if mpu.is_pipeline_first_stage() or mpu.is_pipeline_last_stage():
torch.distributed.all_reduce(self.word_embeddings_weight().
data, group=mpu.get_embedding_group())
else:
None
class VitMlpHeadNew(MegatronModule):
"""Pooler layer.
Pool hidden states of a specific token (for example start of the
sequence) and add a linear transformation followed by a tanh.
Arguments:
hidden_size: hidden size
init_method: weight initialization method for the linear layer.
bias is set to zero.
"""
def __init__(self, hidden_size, num_classes):
super(VitMlpHeadNew, self).__init__()
self.dense_in = torch.nn.Linear(hidden_size, hidden_size)
self.dense_out = torch.nn.Linear(hidden_size, num_classes)
torch.nn.init.constant_(self.dense_out.bias, -10)
def forward(self, input_0):
primals_2 = self.dense_in.weight
primals_3 = self.dense_in.bias
primals_4 = self.dense_out.weight
primals_5 = self.dense_out.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| sourcery-ai-bot/Megatron-LM | VitMlpHead | false | 4,377 | [
"MIT"
] | 0 | f27f44e2c49d1cb39b2288bef6f7d837e11094cb | https://github.com/sourcery-ai-bot/Megatron-LM/tree/f27f44e2c49d1cb39b2288bef6f7d837e11094cb | import torch
def get_args():
parser = argparse.ArgumentParser()
group = parser.add_argument_group(title='input data')
group.add_argument('--input', type=str, required=True, help=
'Path to input JSON')
group.add_argument('--json-keys', nargs='+', default=['text'], help=
'space separate listed of keys to extract from json')
group.add_argument('--split-sentences', action='store_true', help=
'Split documents into sentences.')
group.add_argument('--keep-newlines', action='store_true', help=
'Keep newlines between sentences when splitting.')
group = parser.add_argument_group(title='tokenizer')
group.add_argument('--tokenizer-type', type=str, required=True, choices
=['BertWordPieceLowerCase', 'BertWordPieceCase', 'GPT2BPETokenizer'
], help='What type of tokenizer to use.')
group.add_argument('--vocab-file', type=str, default=None, help=
'Path to the vocab file')
group.add_argument('--merge-file', type=str, default=None, help=
'Path to the BPE merge file (if necessary).')
group.add_argument('--append-eod', action='store_true', help=
'Append an <eod> token to the end of a document.')
group = parser.add_argument_group(title='output data')
group.add_argument('--output-prefix', type=str, required=True, help=
'Path to binary output file without suffix')
group.add_argument('--dataset-impl', type=str, default='mmap', choices=
['lazy', 'cached', 'mmap'])
group = parser.add_argument_group(title='runtime')
group.add_argument('--workers', type=int, default=1, help=
'Number of worker processes to launch')
group.add_argument('--log-interval', type=int, default=100, help=
'Interval between progress updates')
args = parser.parse_args()
args.keep_empty = False
if args.tokenizer_type.lower().startswith('bert'
) and not args.split_sentences:
None
args.rank = 0
args.make_vocab_size_divisible_by = 128
args.tensor_model_parallel_size = 1
return args
def init_method_normal(sigma):
"""Init method based on N(0, sigma)."""
def init_(tensor):
return torch.nn.init.normal_(tensor, mean=0.0, std=sigma)
return init_
class MegatronModule(torch.nn.Module):
"""Megatron specific extensions of torch Module with support
for pipelining."""
def __init__(self, share_word_embeddings=True):
super().__init__()
self.share_word_embeddings = share_word_embeddings
def state_dict_for_save_checkpoint(self, destination=None, prefix='',
keep_vars=False):
"""Use this function to override the state dict for
saving checkpoints."""
return self.state_dict(destination, prefix, keep_vars)
def word_embeddings_weight(self):
if mpu.is_pipeline_first_stage(ignore_virtual=True):
return self.language_model.embedding.word_embeddings.weight
if mpu.is_pipeline_last_stage(ignore_virtual=True):
if not self.share_word_embeddings:
raise Exception(
'word_embeddings_weight() called for last stage, but share_word_embeddings is false'
)
return self.word_embeddings.weight
raise Exception(
'word_embeddings_weight() should be called for first and last stage only'
)
def initialize_word_embeddings(self, init_method_normal):
args = get_args()
if not self.share_word_embeddings:
raise Exception(
'initialize_word_embeddings() was called but share_word_embeddings is false'
)
if args.pipeline_model_parallel_size == 1:
return
if mpu.is_pipeline_last_stage():
assert not mpu.is_pipeline_first_stage()
self._word_embeddings_for_head_key = 'word_embeddings_for_head'
self.word_embeddings = mpu.VocabParallelEmbedding(args.
padded_vocab_si
# ... truncated (>4000 chars) for memory efficiency |
Attention | # 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_7/inductor_cache/jw/cjwwa4uchk246nlrjndt3n65ojf447r3xpsci6idv5lw24heuhdo.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, %repeat], 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
x3 = (xindex // 8)
x2 = (xindex // 32)
x4 = 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*x3) + 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*x2) + ((-4) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x4), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/lz/clzc7c4rqtr7ky6jrepxpu2dlmeo4y66gzcis5bqhwixpt7ktopj.py
# Topologically Sorted Source Nodes: [tanh], Original ATen: [aten.tanh]
# Source node to ATen node mapping:
# tanh => tanh
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_4), kwargs = {})
# %tanh : [num_users=2] = call_function[target=torch.ops.aten.tanh.default](args = (%add_tensor,), kwargs = {})
triton_poi_fused_tanh_1 = async_compile.triton('triton_poi_fused_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: 'i32'}, 'device': 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_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_tanh_1(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_7/inductor_cache/gd/cgdtd7uw2iemby2kfb22fx3vkhdbrpyx2y2l6nq45fmox3ad7stv.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_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],
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 = 16
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_7/inductor_cache/qs/cqsyda2m63ct5ijcfgcipyyfn273chi5d3kmpjuf5asa7h4wdpdv.py
# Topologically Sorted Source Nodes: [alpha], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# alpha => 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=[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__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 = 16
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):
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, 4), (4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (1, 4), (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: [cat], 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: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf0, (16, 8), (8, 1), 0), reinterpret_tensor(primals_3, (8, 4), (1, 8), 0), out=buf1)
del primals_3
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [tanh], Original ATen: [aten.tanh]
triton_poi_fused_tanh_1.run(buf2, primals_4, 64, grid=grid(64), stream=stream0)
del primals_4
buf3 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
# Topologically Sorted Source Nodes: [o], Original ATen: [aten.mm]
extern_kernels.mm(buf2, reinterpret_tensor(primals_5, (4, 1), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [alpha], Original ATen: [aten._softmax]
triton_poi_fused__softmax_2.run(buf3, buf4, 16, grid=grid(16), stream=stream0)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [alpha], Original ATen: [aten._softmax]
triton_poi_fused__softmax_3.run(buf4, buf5, 16, grid=grid(16), stream=stream0)
buf6 = reinterpret_tensor(buf4, (4, 1, 4), (4, 4, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [bmm], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf5, (4, 1, 4), (4, 0, 1), 0), primals_1, out=buf6)
del buf5
return (reinterpret_tensor(buf6, (4, 4), (4, 1), 0), reinterpret_tensor(buf0, (16, 8), (8, 1), 0), buf2, buf3, reinterpret_tensor(primals_1, (4, 4, 4), (16, 1, 4), 0), primals_5, )
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, 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((1, 4), (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 torch import nn
import torch.nn.functional as F
class Attention(nn.Module):
"""
Applies an attention mechanism on the output features from the decoder.
"""
def __init__(self, dim):
super(Attention, self).__init__()
self.dim = dim
self.linear1 = nn.Linear(dim * 2, dim)
self.linear2 = nn.Linear(dim, 1, bias=False)
def _init_hidden(self):
nn.init.xavier_normal_(self.linear1.weight)
nn.init.xavier_normal_(self.linear2.weight)
def forward(self, hidden_state, encoder_outputs):
"""
Arguments:
hidden_state {Variable} -- batch_size x dim
encoder_outputs {Variable} -- batch_size x seq_len x dim
Returns:
Variable -- context vector of size batch_size x dim
"""
batch_size, seq_len, _ = encoder_outputs.size()
hidden_state = hidden_state.unsqueeze(1).repeat(1, seq_len, 1)
inputs = torch.cat((encoder_outputs, hidden_state), 2).view(-1,
self.dim * 2)
o = self.linear2(F.tanh(self.linear1(inputs)))
e = o.view(batch_size, seq_len)
alpha = F.softmax(e, dim=1)
context = torch.bmm(alpha.unsqueeze(1), encoder_outputs).squeeze(1)
return context
def get_inputs():
return [torch.rand([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
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_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
x3 = xindex // 8
x2 = xindex // 32
x4 = 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 * x3 + 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 * x2 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x4, tmp10, xmask)
@triton.jit
def triton_poi_fused_tanh_1(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_2(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
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 = 16
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):
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, 4), (4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (1, 4), (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.mm(reinterpret_tensor(buf0, (16, 8), (8, 1), 0),
reinterpret_tensor(primals_3, (8, 4), (1, 8), 0), out=buf1)
del primals_3
buf2 = buf1
del buf1
triton_poi_fused_tanh_1[grid(64)](buf2, primals_4, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
extern_kernels.mm(buf2, reinterpret_tensor(primals_5, (4, 1), (1, 4
), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__softmax_2[grid(16)](buf3, buf4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__softmax_3[grid(16)](buf4, buf5, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf6 = reinterpret_tensor(buf4, (4, 1, 4), (4, 4, 1), 0)
del buf4
extern_kernels.bmm(reinterpret_tensor(buf5, (4, 1, 4), (4, 0, 1), 0
), primals_1, out=buf6)
del buf5
return reinterpret_tensor(buf6, (4, 4), (4, 1), 0), reinterpret_tensor(buf0
, (16, 8), (8, 1), 0), buf2, buf3, reinterpret_tensor(primals_1, (4,
4, 4), (16, 1, 4), 0), primals_5
class AttentionNew(nn.Module):
"""
Applies an attention mechanism on the output features from the decoder.
"""
def __init__(self, dim):
super(AttentionNew, self).__init__()
self.dim = dim
self.linear1 = nn.Linear(dim * 2, dim)
self.linear2 = nn.Linear(dim, 1, bias=False)
def _init_hidden(self):
nn.init.xavier_normal_(self.linear1.weight)
nn.init.xavier_normal_(self.linear2.weight)
def forward(self, input_0, input_1):
primals_3 = self.linear1.weight
primals_4 = self.linear1.bias
primals_5 = self.linear2.weight
primals_2 = input_0
primals_1 = input_1
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| salmon7ish/Video-Captioning | Attention | false | 4,378 | [
"MIT"
] | 0 | 08359b1824195a7f5eac5b58982efd19ebc6db01 | https://github.com/salmon7ish/Video-Captioning/tree/08359b1824195a7f5eac5b58982efd19ebc6db01 | import torch
from torch import nn
import torch.nn.functional as F
class Model(nn.Module):
"""
Applies an attention mechanism on the output features from the decoder.
"""
def __init__(self, dim):
super().__init__()
self.dim = dim
self.linear1 = nn.Linear(dim * 2, dim)
self.linear2 = nn.Linear(dim, 1, bias=False)
def _init_hidden(self):
nn.init.xavier_normal_(self.linear1.weight)
nn.init.xavier_normal_(self.linear2.weight)
def forward(self, hidden_state, encoder_outputs):
"""
Arguments:
hidden_state {Variable} -- batch_size x dim
encoder_outputs {Variable} -- batch_size x seq_len x dim
Returns:
Variable -- context vector of size batch_size x dim
"""
batch_size, seq_len, _ = encoder_outputs.size()
hidden_state = hidden_state.unsqueeze(1).repeat(1, seq_len, 1)
inputs = torch.cat((encoder_outputs, hidden_state), 2).view(-1,
self.dim * 2)
o = self.linear2(F.tanh(self.linear1(inputs)))
e = o.view(batch_size, seq_len)
alpha = F.softmax(e, dim=1)
context = torch.bmm(alpha.unsqueeze(1), encoder_outputs).squeeze(1)
return context
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4]
|
PartitionedTransformerEncoderLayer | # 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_7/inductor_cache/cc/cccjs22xobelgxtqlsn2fibmbazwulpl6qiesp7clmtct5t76w6i.py
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# qkv_c => 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_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 = 48
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = (xindex // 6) % 4
x2 = (xindex // 24)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (6*x2) + (12*x1)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/va/cva5gwn5lddb4hdqcni22rre3lrnt4hodmwd6zzu7u23zcobkg5i.py
# Topologically Sorted Source Nodes: [cat, q], Original ATen: [aten.cat, aten.mul]
# Source node to ATen node mapping:
# cat => cat
# q => mul
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze, %squeeze_3], -1), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%cat, 0.5), kwargs = {})
triton_poi_fused_cat_mul_1 = async_compile.triton('triton_poi_fused_cat_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=[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_cat_mul_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_mul_1(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 % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + ((6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + ((6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = 0.5
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/q4/cq4j3evfcbzbqvsc5o6spd3cwbk5ruwxtlfpvyrijy74rzhdizzf.py
# Topologically Sorted Source Nodes: [k], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# k => cat_1
# Graph fragment:
# %cat_1 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze_1, %squeeze_4], -1), 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=[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_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 = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2 + (6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + (2 + (6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/s4/cs472yivvl3yzse325afzknsz7ua5dqrqzmwls3lwujk3hte6xkl.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%view_11, [-1], 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_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 = 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_7/inductor_cache/zh/czh6tw7ngffcygnivwvcjex5edxy3ms4t27ymyn2hemxlpspxzq7.py
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# probs => 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_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 = 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_7/inductor_cache/wy/cwyd2c5eqe6dmcriegfsxxijhdetgyb5fhlqse4jkwrdzjzz4yra.py
# Topologically Sorted Source Nodes: [v], Original ATen: [aten.cat]
# Source node to ATen node mapping:
# v => cat_2
# Graph fragment:
# %cat_2 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%squeeze_2, %squeeze_5], -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=[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_cat_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_cat_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 % 4
x4 = (xindex // 4)
x1 = (xindex // 4) % 4
x2 = (xindex // 16) % 4
x3 = (xindex // 64)
tmp0 = x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 + (6*x4) + x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 4, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tl.load(in_ptr1 + (4 + (6*x4) + ((-2) + x0)), tmp6 & xmask, eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + (4*x2) + (16*x1) + (64*x3)), tmp10, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rb/crbo7mnyz6jhctfcs6irltofws4absfehq57gbubuwmkgqremvir.py
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_c => clone_3
# Graph fragment:
# %clone_3 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_22,), 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=[32, 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 = 32
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
x3 = xindex
y0 = yindex % 2
y1 = (yindex // 2) % 4
y2 = (yindex // 8)
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + (4*y1) + (16*x3) + (64*y2)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + (4*y4)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/uf/cufwzthblw5bpxm6to23u5x5pyxdqityp5l4pgpwkcw24uzkh5ok.py
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_c => clone_4
# Graph fragment:
# %clone_4 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_23,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_7 = async_compile.triton('triton_poi_fused_clone_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: 'i32'}, 'device': 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_7', '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_7(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
x0 = xindex % 2
x1 = (xindex // 2) % 4
x2 = (xindex // 8)
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + (2*x2) + (4*x1)), xmask)
tl.store(out_ptr0 + (x3), tmp0, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/in/cinchtemkwlurwioe377jflj2qj5ijghnp4kjugwuldc3qsv4n6b.py
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
# Source node to ATen node mapping:
# out_p => clone_5
# Graph fragment:
# %clone_5 : [num_users=1] = call_function[target=torch.ops.aten.clone.default](args = (%permute_27,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_8 = async_compile.triton('triton_poi_fused_clone_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=[32, 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_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_clone_8(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = (yindex // 2) % 4
y2 = (yindex // 8)
y4 = yindex
tmp0 = tl.load(in_ptr0 + (2 + y0 + (4*y1) + (16*x3) + (64*y2)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + (4*y4)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/pw/cpwq2dftl7mquhip6p5ogkguqiqaywd2dfdf3vppi43r3xnef24d.py
# Topologically Sorted Source Nodes: [residual, add], Original ATen: [aten.cat, aten.add]
# Source node to ATen node mapping:
# add => add
# residual => cat_3
# Graph fragment:
# %cat_3 : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%view_19, %view_23], -1), kwargs = {})
# %add : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_1, %cat_3), kwargs = {})
triton_poi_fused_add_cat_9 = async_compile.triton('triton_poi_fused_add_cat_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_cat_9', '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_9(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
x0 = xindex % 4
x1 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x2), xmask)
tmp1 = x0
tmp2 = tl.full([1], 0, tl.int64)
tmp3 = tmp1 >= tmp2
tmp4 = tl.full([1], 2, tl.int64)
tmp5 = tmp1 < tmp4
tmp6 = tl.load(in_ptr1 + ((2*x1) + x0), tmp5 & xmask, eviction_policy='evict_last', other=0.0)
tmp7 = tmp1 >= tmp4
tmp8 = tl.full([1], 4, tl.int64)
tmp9 = tmp1 < tmp8
tmp10 = tl.load(in_ptr2 + ((2*x1) + ((-2) + x0)), tmp7 & xmask, eviction_policy='evict_last', other=0.0)
tmp11 = tl.where(tmp5, tmp6, tmp10)
tmp12 = tmp0 + tmp11
tl.store(out_ptr0 + (x2), tmp12, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/nn/cnnxhu2nhctulajy3rzg5wf2f3gd62kwlmtnwu5433qpihin4y26.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => add_1, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add, [2]), kwargs = {correction: 0, keepdim: True})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_10, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_1,), kwargs = {})
triton_poi_fused_native_layer_norm_10 = async_compile.triton('triton_poi_fused_native_layer_norm_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=[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_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_native_layer_norm_10(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_7/inductor_cache/73/c73z2hfes5izl473wn57vaku4rt2ae7swkdamlriywh5x5xt7g3z.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => add_1, add_2, mul_1, mul_2, rsqrt, sub_1, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add, [2]), kwargs = {correction: 0, keepdim: True})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_10, 1e-05), kwargs = {})
# %rsqrt : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_1,), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add, %getitem_11), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub_1, %rsqrt), kwargs = {})
# %mul_2 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%mul_1, %primals_6), kwargs = {})
# %add_2 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_2, %primals_7), kwargs = {})
triton_poi_fused_native_layer_norm_11 = async_compile.triton('triton_poi_fused_native_layer_norm_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=[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_11', '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_11(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_7/inductor_cache/fh/cfhr5stscm2juerq4j7icwweho5x7pvy3hrm3un3lpvqdfbvvwdg.py
# Topologically Sorted Source Nodes: [out_c_1, x_c_2], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# out_c_1 => add_3
# x_c_2 => relu
# Graph fragment:
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_25, %primals_9), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_3,), kwargs = {})
# %le_1 : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu, 0), kwargs = {})
triton_poi_fused_add_relu_threshold_backward_12 = async_compile.triton('triton_poi_fused_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.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, 3), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_relu_threshold_backward_12', '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_add_relu_threshold_backward_12(in_out_ptr0, 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
x0 = xindex % 2
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, 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, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_3, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_4, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_5, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_6, (4, ), (1, ))
assert_size_stride(primals_7, (4, ), (1, ))
assert_size_stride(primals_8, (2, 2), (2, 1))
assert_size_stride(primals_9, (2, ), (1, ))
assert_size_stride(primals_10, (2, 2), (2, 1))
assert_size_stride(primals_11, (2, ), (1, ))
assert_size_stride(primals_12, (2, 2), (2, 1))
assert_size_stride(primals_13, (2, ), (1, ))
assert_size_stride(primals_14, (2, 2), (2, 1))
assert_size_stride(primals_15, (2, ), (1, ))
assert_size_stride(primals_16, (4, ), (1, ))
assert_size_stride(primals_17, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((2, 4, 3, 2, 1, 1), (24, 6, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.clone]
stream0 = get_raw_stream(0)
triton_poi_fused_clone_0.run(primals_2, buf0, 48, grid=grid(48), stream=stream0)
del primals_2
buf1 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_c], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4, 1), 0), reinterpret_tensor(buf0, (1, 2, 24), (0, 24, 1), 0), out=buf1)
buf2 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [qkv_p], Original ATen: [aten.clone]
triton_poi_fused_clone_0.run(primals_3, buf2, 48, grid=grid(48), stream=stream0)
del primals_3
buf3 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
# Topologically Sorted Source Nodes: [qkv_p], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4, 1), 2), reinterpret_tensor(buf2, (1, 2, 24), (0, 24, 1), 0), out=buf3)
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [cat, q], Original ATen: [aten.cat, aten.mul]
triton_poi_fused_cat_mul_1.run(buf1, buf3, buf4, 256, grid=grid(256), stream=stream0)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [k], Original ATen: [aten.cat]
triton_poi_fused_cat_2.run(buf1, buf3, buf5, 256, grid=grid(256), stream=stream0)
buf6 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [dots], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf4, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf5, (16, 4, 4), (16, 1, 4), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
triton_poi_fused__softmax_3.run(buf6, buf7, 256, grid=grid(256), stream=stream0)
buf8 = reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0); del buf6 # reuse
# Topologically Sorted Source Nodes: [probs], Original ATen: [aten._softmax]
triton_poi_fused__softmax_4.run(buf7, buf8, 256, grid=grid(256), stream=stream0)
buf9 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [v], Original ATen: [aten.cat]
triton_poi_fused_cat_5.run(buf1, buf3, buf9, 256, grid=grid(256), stream=stream0)
del buf1
del buf3
buf10 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [o], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
triton_poi_fused_clone_6.run(buf10, buf11, 32, 4, grid=grid(32, 4), stream=stream0)
buf12 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(primals_4, buf12, 16, grid=grid(16), stream=stream0)
del primals_4
buf13 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf11, (1, 16, 8), (0, 8, 1), 0), reinterpret_tensor(buf12, (1, 8, 2), (0, 2, 1), 0), out=buf13)
buf14 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
triton_poi_fused_clone_8.run(buf10, buf14, 32, 4, grid=grid(32, 4), stream=stream0)
del buf10
buf15 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(primals_5, buf15, 16, grid=grid(16), stream=stream0)
del primals_5
buf16 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p], Original ATen: [aten.bmm]
extern_kernels.bmm(reinterpret_tensor(buf14, (1, 16, 8), (0, 8, 1), 0), reinterpret_tensor(buf15, (1, 8, 2), (0, 2, 1), 0), out=buf16)
buf17 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [residual, add], Original ATen: [aten.cat, aten.add]
triton_poi_fused_add_cat_9.run(primals_1, buf13, buf16, buf17, 64, grid=grid(64), stream=stream0)
buf18 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf19 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_10.run(buf17, buf18, buf19, 16, grid=grid(16), stream=stream0)
buf20 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_11.run(buf17, buf18, buf19, primals_6, primals_7, buf20, 64, grid=grid(64), stream=stream0)
del primals_7
buf21 = reinterpret_tensor(buf16, (16, 2), (2, 1), 0); del buf16 # reuse
# Topologically Sorted Source Nodes: [out_c_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf20, (16, 2), (4, 1), 0), reinterpret_tensor(primals_8, (2, 2), (1, 2), 0), out=buf21)
buf22 = reinterpret_tensor(buf13, (16, 2), (2, 1), 0); del buf13 # reuse
# Topologically Sorted Source Nodes: [out_p_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf20, (16, 2), (4, 1), 2), reinterpret_tensor(primals_10, (2, 2), (1, 2), 0), out=buf22)
buf23 = reinterpret_tensor(buf21, (4, 4, 2), (8, 2, 1), 0); del buf21 # reuse
buf32 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.bool)
# Topologically Sorted Source Nodes: [out_c_1, x_c_2], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_12.run(buf23, primals_9, buf32, 32, grid=grid(32), stream=stream0)
del primals_9
buf24 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_c_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_13, reinterpret_tensor(buf23, (16, 2), (2, 1), 0), reinterpret_tensor(primals_12, (2, 2), (1, 2), 0), alpha=1, beta=1, out=buf24)
del primals_13
buf25 = reinterpret_tensor(buf22, (4, 4, 2), (8, 2, 1), 0); del buf22 # reuse
buf31 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.bool)
# Topologically Sorted Source Nodes: [out_p_1, x_p_2], Original ATen: [aten.add, aten.relu, aten.threshold_backward]
triton_poi_fused_add_relu_threshold_backward_12.run(buf25, primals_11, buf31, 32, grid=grid(32), stream=stream0)
del primals_11
buf26 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_p_2], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_15, reinterpret_tensor(buf25, (16, 2), (2, 1), 0), reinterpret_tensor(primals_14, (2, 2), (1, 2), 0), alpha=1, beta=1, out=buf26)
del primals_15
buf27 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [residual_2, add_1], Original ATen: [aten.cat, aten.add]
triton_poi_fused_add_cat_9.run(buf20, buf24, buf26, buf27, 64, grid=grid(64), stream=stream0)
del buf24
del buf26
buf28 = buf19; del buf19 # reuse
buf29 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_10.run(buf27, buf28, buf29, 16, grid=grid(16), stream=stream0)
buf30 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x_1], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_11.run(buf27, buf28, buf29, primals_16, primals_17, buf30, 64, grid=grid(64), stream=stream0)
del buf28
del buf29
del primals_17
return (buf30, primals_6, primals_16, buf8, buf17, reinterpret_tensor(buf20, (16, 2), (4, 1), 0), reinterpret_tensor(buf20, (16, 2), (4, 1), 2), reinterpret_tensor(buf23, (16, 2), (2, 1), 0), reinterpret_tensor(buf25, (16, 2), (2, 1), 0), buf27, primals_14, primals_12, buf31, buf32, primals_10, primals_8, reinterpret_tensor(buf14, (1, 8, 16), (128, 1, 8), 0), reinterpret_tensor(buf15, (1, 2, 8), (16, 1, 2), 0), reinterpret_tensor(buf11, (1, 8, 16), (128, 1, 8), 0), reinterpret_tensor(buf12, (1, 2, 8), (16, 1, 2), 0), reinterpret_tensor(buf9, (16, 4, 4), (16, 1, 4), 0), reinterpret_tensor(buf4, (16, 4, 4), (16, 1, 4), 0), reinterpret_tensor(buf5, (16, 4, 4), (16, 4, 1), 0), reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 2), reinterpret_tensor(primals_1, (1, 2, 16), (64, 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, 2, 3, 2), (12, 6, 2, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((4, 2, 3, 2), (12, 6, 2, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 2, 2), (4, 2, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((4, 2, 2), (4, 2, 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)
primals_8 = rand_strided((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_9 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_14 = rand_strided((2, 2), (2, 1), device='cuda:0', dtype=torch.float32)
primals_15 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_16 = rand_strided((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 math
import torch
import torch.nn as nn
import torch.nn.functional as F
class FeatureDropoutFunction(torch.autograd.function.InplaceFunction):
@staticmethod
def forward(ctx, input, p=0.5, train=False, inplace=False):
if p < 0 or p > 1:
raise ValueError(
'dropout probability has to be between 0 and 1, but got {}'
.format(p))
ctx.p = p
ctx.train = train
ctx.inplace = inplace
if ctx.inplace:
ctx.mark_dirty(input)
output = input
else:
output = input.clone()
if ctx.p > 0 and ctx.train:
ctx.noise = torch.empty((input.size(0), input.size(-1)), dtype=
input.dtype, layout=input.layout, device=input.device)
if ctx.p == 1:
ctx.noise.fill_(0)
else:
ctx.noise.bernoulli_(1 - ctx.p).div_(1 - ctx.p)
ctx.noise = ctx.noise[:, None, :]
output.mul_(ctx.noise)
return output
@staticmethod
def backward(ctx, grad_output):
if ctx.p > 0 and ctx.train:
return grad_output.mul(ctx.noise), None, None, None
else:
return grad_output, None, None, None
class FeatureDropout(nn.Dropout):
"""
Feature-level dropout: takes an input of size len x num_features and drops
each feature with probabibility p. A feature is dropped across the full
portion of the input that corresponds to a single batch element.
"""
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
x_c = FeatureDropoutFunction.apply(x_c, self.p, self.training,
self.inplace)
x_p = FeatureDropoutFunction.apply(x_p, self.p, self.training,
self.inplace)
return x_c, x_p
else:
return FeatureDropoutFunction.apply(x, self.p, self.training,
self.inplace)
class PartitionedLinear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
out_c = self.linear_c(x_c)
out_p = self.linear_p(x_p)
return out_c, out_p
class PartitionedMultiHeadAttention(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, x, mask=None):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
qkv_c = torch.einsum('btf,hfca->bhtca', x_c, self.w_qkv_c)
qkv_p = torch.einsum('btf,hfca->bhtca', x_p, self.w_qkv_p)
q_c, k_c, v_c = [c.squeeze(dim=3) for c in torch.chunk(qkv_c, 3, dim=3)
]
q_p, k_p, v_p = [c.squeeze(dim=3) for c in torch.chunk(qkv_p, 3, dim=3)
]
q = torch.cat([q_c, q_p], dim=-1) * self.scaling_factor
k = torch.cat([k_c, k_p], dim=-1)
v = torch.cat([v_c, v_p], dim=-1)
dots = torch.einsum('bhqa,bhka->bhqk', q, k)
if mask is not None:
dots.data.masked_fill_(~mask[:, None, None, :], -float('inf'))
probs = F.softmax(dots, dim=-1)
probs = self.dropout(probs)
o = torch.einsum('bhqk,bhka->bhqa', probs, v)
o_c, o_p = torch.chunk(o, 2, dim=-1)
out_c = torch.einsum('bhta,haf->btf', o_c, self.w_o_c)
out_p = torch.einsum('bhta,haf->btf', o_p, self.w_o_p)
return out_c, out_p
class PartitionedReLU(nn.ReLU):
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
return super().forward(x_c), super().forward(x_p)
class PartitionedTransformerEncoderLayer(nn.Module):
def __init__(self, d_model, n_head, d_qkv, d_ff, ff_dropout=0.1,
residual_dropout=0.1, attention_dropout=0.1, activation=
PartitionedReLU()):
super().__init__()
self.self_attn = PartitionedMultiHeadAttention(d_model, n_head,
d_qkv, attention_dropout=attention_dropout)
self.linear1 = PartitionedLinear(d_model, d_ff)
self.ff_dropout = FeatureDropout(ff_dropout)
self.linear2 = PartitionedLinear(d_ff, d_model)
self.norm_attn = nn.LayerNorm(d_model)
self.norm_ff = nn.LayerNorm(d_model)
self.residual_dropout_attn = FeatureDropout(residual_dropout)
self.residual_dropout_ff = FeatureDropout(residual_dropout)
self.activation = activation
def forward(self, x, mask=None):
residual = self.self_attn(x, mask=mask)
residual = torch.cat(residual, dim=-1)
residual = self.residual_dropout_attn(residual)
x = self.norm_attn(x + residual)
residual = self.linear2(self.ff_dropout(self.activation(self.
linear1(x))))
residual = torch.cat(residual, dim=-1)
residual = self.residual_dropout_ff(residual)
x = self.norm_ff(x + residual)
return x
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4, 'n_head': 4, 'd_qkv': 4, 'd_ff': 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 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
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 = 48
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 6
x1 = xindex // 6 % 4
x2 = xindex // 24
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 6 * x2 + 12 * x1), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_poi_fused_cat_mul_1(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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (6 * x4 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = 0.5
tmp12 = tmp10 * tmp11
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp12, xmask)
@triton.jit
def triton_poi_fused_cat_2(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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2 + 6 * x4 + x0), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (2 + 6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp10, 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 = 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_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 = 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_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 % 4
x4 = xindex // 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 + 6 * x4 + x0), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 + 6 * x4 + (-2 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), tmp10, xmask)
@triton.jit
def triton_poi_fused_clone_6(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = yindex // 2 % 4
y2 = yindex // 8
y4 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * y1 + 16 * x3 + 64 * y2), xmask &
ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + 4 * y4), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_7(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
x0 = xindex % 2
x1 = xindex // 2 % 4
x2 = xindex // 8
x3 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 2 * x2 + 4 * x1), xmask)
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_poi_fused_clone_8(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 32
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
x3 = xindex
y0 = yindex % 2
y1 = yindex // 2 % 4
y2 = yindex // 8
y4 = yindex
tmp0 = tl.load(in_ptr0 + (2 + y0 + 4 * y1 + 16 * x3 + 64 * y2), xmask &
ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (x3 + 4 * y4), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_cat_9(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
x0 = xindex % 4
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = x0
tl.full([1], 0, tl.int64)
tmp4 = tl.full([1], 2, tl.int64)
tmp5 = tmp1 < tmp4
tmp6 = tl.load(in_ptr1 + (2 * x1 + x0), tmp5 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp7 = tmp1 >= tmp4
tl.full([1], 4, tl.int64)
tmp10 = tl.load(in_ptr2 + (2 * x1 + (-2 + x0)), tmp7 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = tl.where(tmp5, tmp6, tmp10)
tmp12 = tmp0 + tmp11
tl.store(out_ptr0 + x2, tmp12, xmask)
@triton.jit
def triton_poi_fused_native_layer_norm_10(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_11(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_add_relu_threshold_backward_12(in_out_ptr0, 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
x0 = xindex % 2
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, 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, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_3, (4, 2, 3, 2), (12, 6, 2, 1))
assert_size_stride(primals_4, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_5, (4, 2, 2), (4, 2, 1))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (2, 2), (2, 1))
assert_size_stride(primals_9, (2,), (1,))
assert_size_stride(primals_10, (2, 2), (2, 1))
assert_size_stride(primals_11, (2,), (1,))
assert_size_stride(primals_12, (2, 2), (2, 1))
assert_size_stride(primals_13, (2,), (1,))
assert_size_stride(primals_14, (2, 2), (2, 1))
assert_size_stride(primals_15, (2,), (1,))
assert_size_stride(primals_16, (4,), (1,))
assert_size_stride(primals_17, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((2, 4, 3, 2, 1, 1), (24, 6, 2, 1, 1, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(48)](primals_2, buf0, 48, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_2
buf1 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4,
1), 0), reinterpret_tensor(buf0, (1, 2, 24), (0, 24, 1), 0),
out=buf1)
buf2 = buf0
del buf0
triton_poi_fused_clone_0[grid(48)](primals_3, buf2, 48, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_3
buf3 = empty_strided_cuda((1, 16, 24), (384, 24, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(primals_1, (1, 16, 2), (0, 4,
1), 2), reinterpret_tensor(buf2, (1, 2, 24), (0, 24, 1), 0),
out=buf3)
del buf2
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_cat_mul_1[grid(256)](buf1, buf3, buf4, 256, XBLOCK
=256, num_warps=4, num_stages=1)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_cat_2[grid(256)](buf1, buf3, buf5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf6 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf4, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf5, (16, 4, 4), (16, 1, 4), 0), out=buf6)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_3[grid(256)](buf6, buf7, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf8 = reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf6
triton_poi_fused__softmax_4[grid(256)](buf7, buf8, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf9 = buf7
del buf7
triton_poi_fused_cat_5[grid(256)](buf1, buf3, buf9, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf1
del buf3
buf10 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf8, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf9, (16, 4, 4), (16, 4, 1), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch
.float32)
triton_poi_fused_clone_6[grid(32, 4)](buf10, buf11, 32, 4, XBLOCK=4,
YBLOCK=32, num_warps=4, num_stages=1)
buf12 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.
float32)
triton_poi_fused_clone_7[grid(16)](primals_4, buf12, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_4
buf13 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf11, (1, 16, 8), (0, 8, 1),
0), reinterpret_tensor(buf12, (1, 8, 2), (0, 2, 1), 0), out=buf13)
buf14 = empty_strided_cuda((4, 4, 2, 4, 1), (32, 8, 4, 1, 1), torch
.float32)
triton_poi_fused_clone_8[grid(32, 4)](buf10, buf14, 32, 4, XBLOCK=4,
YBLOCK=32, num_warps=4, num_stages=1)
del buf10
buf15 = empty_strided_cuda((2, 4, 2, 1, 1), (8, 2, 1, 1, 1), torch.
float32)
triton_poi_fused_clone_7[grid(16)](primals_5, buf15, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_5
buf16 = empty_strided_cuda((1, 16, 2), (32, 2, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf14, (1, 16, 8), (0, 8, 1),
0), reinterpret_tensor(buf15, (1, 8, 2), (0, 2, 1), 0), out=buf16)
buf17 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_cat_9[grid(64)](primals_1, buf13, buf16, buf17,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf18 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf19 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused_native_layer_norm_10[grid(16)](buf17, buf18, buf19,
16, XBLOCK=16, num_warps=1, num_stages=1)
buf20 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_native_layer_norm_11[grid(64)](buf17, buf18, buf19,
primals_6, primals_7, buf20, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del primals_7
buf21 = reinterpret_tensor(buf16, (16, 2), (2, 1), 0)
del buf16
extern_kernels.mm(reinterpret_tensor(buf20, (16, 2), (4, 1), 0),
reinterpret_tensor(primals_8, (2, 2), (1, 2), 0), out=buf21)
buf22 = reinterpret_tensor(buf13, (16, 2), (2, 1), 0)
del buf13
extern_kernels.mm(reinterpret_tensor(buf20, (16, 2), (4, 1), 2),
reinterpret_tensor(primals_10, (2, 2), (1, 2), 0), out=buf22)
buf23 = reinterpret_tensor(buf21, (4, 4, 2), (8, 2, 1), 0)
del buf21
buf32 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.bool)
triton_poi_fused_add_relu_threshold_backward_12[grid(32)](buf23,
primals_9, buf32, 32, XBLOCK=32, num_warps=1, num_stages=1)
del primals_9
buf24 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
extern_kernels.addmm(primals_13, reinterpret_tensor(buf23, (16, 2),
(2, 1), 0), reinterpret_tensor(primals_12, (2, 2), (1, 2), 0),
alpha=1, beta=1, out=buf24)
del primals_13
buf25 = reinterpret_tensor(buf22, (4, 4, 2), (8, 2, 1), 0)
del buf22
buf31 = empty_strided_cuda((4, 4, 2), (8, 2, 1), torch.bool)
triton_poi_fused_add_relu_threshold_backward_12[grid(32)](buf25,
primals_11, buf31, 32, XBLOCK=32, num_warps=1, num_stages=1)
del primals_11
buf26 = empty_strided_cuda((16, 2), (2, 1), torch.float32)
extern_kernels.addmm(primals_15, reinterpret_tensor(buf25, (16, 2),
(2, 1), 0), reinterpret_tensor(primals_14, (2, 2), (1, 2), 0),
alpha=1, beta=1, out=buf26)
del primals_15
buf27 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_add_cat_9[grid(64)](buf20, buf24, buf26, buf27, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del buf24
del buf26
buf28 = buf19
del buf19
buf29 = buf18
del buf18
triton_poi_fused_native_layer_norm_10[grid(16)](buf27, buf28, buf29,
16, XBLOCK=16, num_warps=1, num_stages=1)
buf30 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_native_layer_norm_11[grid(64)](buf27, buf28, buf29,
primals_16, primals_17, buf30, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del buf28
del buf29
del primals_17
return (buf30, primals_6, primals_16, buf8, buf17, reinterpret_tensor(
buf20, (16, 2), (4, 1), 0), reinterpret_tensor(buf20, (16, 2), (4,
1), 2), reinterpret_tensor(buf23, (16, 2), (2, 1), 0),
reinterpret_tensor(buf25, (16, 2), (2, 1), 0), buf27, primals_14,
primals_12, buf31, buf32, primals_10, primals_8, reinterpret_tensor
(buf14, (1, 8, 16), (128, 1, 8), 0), reinterpret_tensor(buf15, (1,
2, 8), (16, 1, 2), 0), reinterpret_tensor(buf11, (1, 8, 16), (128,
1, 8), 0), reinterpret_tensor(buf12, (1, 2, 8), (16, 1, 2), 0),
reinterpret_tensor(buf9, (16, 4, 4), (16, 1, 4), 0),
reinterpret_tensor(buf4, (16, 4, 4), (16, 1, 4), 0),
reinterpret_tensor(buf5, (16, 4, 4), (16, 4, 1), 0),
reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 2),
reinterpret_tensor(primals_1, (1, 2, 16), (64, 1, 4), 0))
class FeatureDropoutFunction(torch.autograd.function.InplaceFunction):
@staticmethod
def forward(ctx, input, p=0.5, train=False, inplace=False):
if p < 0 or p > 1:
raise ValueError(
'dropout probability has to be between 0 and 1, but got {}'
.format(p))
ctx.p = p
ctx.train = train
ctx.inplace = inplace
if ctx.inplace:
ctx.mark_dirty(input)
output = input
else:
output = input.clone()
if ctx.p > 0 and ctx.train:
ctx.noise = torch.empty((input.size(0), input.size(-1)), dtype=
input.dtype, layout=input.layout, device=input.device)
if ctx.p == 1:
ctx.noise.fill_(0)
else:
ctx.noise.bernoulli_(1 - ctx.p).div_(1 - ctx.p)
ctx.noise = ctx.noise[:, None, :]
output.mul_(ctx.noise)
return output
@staticmethod
def backward(ctx, grad_output):
if ctx.p > 0 and ctx.train:
return grad_output.mul(ctx.noise), None, None, None
else:
return grad_output, None, None, None
class FeatureDropout(nn.Dropout):
"""
Feature-level dropout: takes an input of size len x num_features and drops
each feature with probabibility p. A feature is dropped across the full
portion of the input that corresponds to a single batch element.
"""
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
x_c = FeatureDropoutFunction.apply(x_c, self.p, self.training,
self.inplace)
x_p = FeatureDropoutFunction.apply(x_p, self.p, self.training,
self.inplace)
return x_c, x_p
else:
return FeatureDropoutFunction.apply(x, self.p, self.training,
self.inplace)
class PartitionedLinear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
out_c = self.linear_c(x_c)
out_p = self.linear_p(x_p)
return out_c, out_p
class PartitionedMultiHeadAttention(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, x, mask=None):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
qkv_c = torch.einsum('btf,hfca->bhtca', x_c, self.w_qkv_c)
qkv_p = torch.einsum('btf,hfca->bhtca', x_p, self.w_qkv_p)
q_c, k_c, v_c = [c.squeeze(dim=3) for c in torch.chunk(qkv_c, 3, dim=3)
]
q_p, k_p, v_p = [c.squeeze(dim=3) for c in torch.chunk(qkv_p, 3, dim=3)
]
q = torch.cat([q_c, q_p], dim=-1) * self.scaling_factor
k = torch.cat([k_c, k_p], dim=-1)
v = torch.cat([v_c, v_p], dim=-1)
dots = torch.einsum('bhqa,bhka->bhqk', q, k)
if mask is not None:
dots.data.masked_fill_(~mask[:, None, None, :], -float('inf'))
probs = F.softmax(dots, dim=-1)
probs = self.dropout(probs)
o = torch.einsum('bhqk,bhka->bhqa', probs, v)
o_c, o_p = torch.chunk(o, 2, dim=-1)
out_c = torch.einsum('bhta,haf->btf', o_c, self.w_o_c)
out_p = torch.einsum('bhta,haf->btf', o_p, self.w_o_p)
return out_c, out_p
class PartitionedReLU(nn.ReLU):
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
return super().forward(x_c), super().forward(x_p)
class PartitionedTransformerEncoderLayerNew(nn.Module):
def __init__(self, d_model, n_head, d_qkv, d_ff, ff_dropout=0.1,
residual_dropout=0.1, attention_dropout=0.1, activation=
PartitionedReLU()):
super().__init__()
self.self_attn = PartitionedMultiHeadAttention(d_model, n_head,
d_qkv, attention_dropout=attention_dropout)
self.linear1 = PartitionedLinear(d_model, d_ff)
self.ff_dropout = FeatureDropout(ff_dropout)
self.linear2 = PartitionedLinear(d_ff, d_model)
self.norm_attn = nn.LayerNorm(d_model)
self.norm_ff = nn.LayerNorm(d_model)
self.residual_dropout_attn = FeatureDropout(residual_dropout)
self.residual_dropout_ff = FeatureDropout(residual_dropout)
self.activation = activation
def forward(self, input_0):
primals_2 = self.self_attn.w_qkv_c
primals_3 = self.self_attn.w_qkv_p
primals_4 = self.self_attn.w_o_c
primals_5 = self.self_attn.w_o_p
primals_8 = self.linear1.linear_c.weight
primals_9 = self.linear1.linear_c.bias
primals_10 = self.linear1.linear_p.weight
primals_11 = self.linear1.linear_p.bias
primals_12 = self.linear2.linear_c.weight
primals_13 = self.linear2.linear_c.bias
primals_14 = self.linear2.linear_p.weight
primals_15 = self.linear2.linear_p.bias
primals_6 = self.norm_attn.weight
primals_7 = self.norm_attn.bias
primals_16 = self.norm_ff.weight
primals_17 = self.norm_ff.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,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17])
return output[0]
| skulick/self-attentive-parser | PartitionedTransformerEncoderLayer | false | 4,379 | [
"MIT"
] | 0 | 04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | https://github.com/skulick/self-attentive-parser/tree/04a91e80cc05bcfe8f48145517f58e85f0c8ade6 | import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class FeatureDropoutFunction(torch.autograd.function.InplaceFunction):
@staticmethod
def forward(ctx, input, p=0.5, train=False, inplace=False):
if p < 0 or p > 1:
raise ValueError(
'dropout probability has to be between 0 and 1, but got {}'
.format(p))
ctx.p = p
ctx.train = train
ctx.inplace = inplace
if ctx.inplace:
ctx.mark_dirty(input)
output = input
else:
output = input.clone()
if ctx.p > 0 and ctx.train:
ctx.noise = torch.empty((input.size(0), input.size(-1)), dtype=
input.dtype, layout=input.layout, device=input.device)
if ctx.p == 1:
ctx.noise.fill_(0)
else:
ctx.noise.bernoulli_(1 - ctx.p).div_(1 - ctx.p)
ctx.noise = ctx.noise[:, None, :]
output.mul_(ctx.noise)
return output
@staticmethod
def backward(ctx, grad_output):
if ctx.p > 0 and ctx.train:
return grad_output.mul(ctx.noise), None, None, None
else:
return grad_output, None, None, None
class FeatureDropout(nn.Dropout):
"""
Feature-level dropout: takes an input of size len x num_features and drops
each feature with probabibility p. A feature is dropped across the full
portion of the input that corresponds to a single batch element.
"""
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
x_c = FeatureDropoutFunction.apply(x_c, self.p, self.training,
self.inplace)
x_p = FeatureDropoutFunction.apply(x_p, self.p, self.training,
self.inplace)
return x_c, x_p
else:
return FeatureDropoutFunction.apply(x, self.p, self.training,
self.inplace)
class PartitionedLinear(nn.Module):
def __init__(self, in_features, out_features, bias=True):
super().__init__()
self.linear_c = nn.Linear(in_features // 2, out_features // 2, bias)
self.linear_p = nn.Linear(in_features // 2, out_features // 2, bias)
def forward(self, x):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
out_c = self.linear_c(x_c)
out_p = self.linear_p(x_p)
return out_c, out_p
class PartitionedMultiHeadAttention(nn.Module):
def __init__(self, d_model, n_head, d_qkv, attention_dropout=0.1,
initializer_range=0.02):
super().__init__()
self.w_qkv_c = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_qkv_p = nn.Parameter(torch.Tensor(n_head, d_model // 2, 3,
d_qkv // 2))
self.w_o_c = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
self.w_o_p = nn.Parameter(torch.Tensor(n_head, d_qkv // 2, d_model //
2))
bound = math.sqrt(3.0) * initializer_range
for param in [self.w_qkv_c, self.w_qkv_p, self.w_o_c, self.w_o_p]:
nn.init.uniform_(param, -bound, bound)
self.scaling_factor = 1 / d_qkv ** 0.5
self.dropout = nn.Dropout(attention_dropout)
def forward(self, x, mask=None):
if isinstance(x, tuple):
x_c, x_p = x
else:
x_c, x_p = torch.chunk(x, 2, dim=-1)
qkv_c = torch.einsum('btf,hfca->bhtca', x_c, self.w_qkv_c)
qkv_p = torch.einsum('btf,hfca->bhtca', x_p, self.w_qkv_p)
q_c, k_c, v_c = [c.squeeze(dim=3) for c in torch.chunk(qkv_c, 3, dim=3)
]
q_p, k_p, v_p = [c.squeeze(dim=3) for c in torch.chunk(qkv_p, 3, dim=3)
]
q = torch.cat([q_c, q_p], dim=-1) * self.scaling_factor
k = torch.cat([k_c, k_p], dim=-1)
v = torch.cat([v_c, v_p], dim=-1)
dots = torch.einsum(
# ... truncated (>4000 chars) for memory efficiency |
mlp_model | # 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_7/inductor_cache/tf/ctfobpckmiv3kkga3a6gzs6unuclcnxpb4xc2h5r3udgxgix4ip5.py
# Topologically Sorted Source Nodes: [out_2], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# out_2 => relu
# Graph fragment:
# %add_tensor_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_2, %primals_3), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_2,), 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')
# kernel path: runs/run_shard_7/inductor_cache/i5/ci5f4nyelvfg4yf2o65ompoikj7ejkd32vb6hqtyrgycc5eswrpx.py
# Topologically Sorted Source Nodes: [out_4], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# out_4 => relu_1
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_5), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_1 = async_compile.triton('triton_poi_fused_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=[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_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_relu_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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')
# kernel path: runs/run_shard_7/inductor_cache/dm/cdmk5acvohhdeyjsc6lsgsurbqis3xmgjftzfmanusy52dkygd23.py
# Topologically Sorted Source Nodes: [out_6], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# out_6 => 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_2 = async_compile.triton('triton_poi_fused_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=[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_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_relu_2(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, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, ), (1, ))
assert_size_stride(primals_4, (128, 4), (4, 1))
assert_size_stride(primals_5, (128, ), (1, ))
assert_size_stride(primals_6, (64, 128), (128, 1))
assert_size_stride(primals_7, (64, ), (1, ))
assert_size_stride(primals_8, (4, 64), (64, 1))
assert_size_stride(primals_9, (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: [out_2], 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, 128), (128, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf1, reinterpret_tensor(primals_4, (4, 128), (1, 4), 0), out=buf2)
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [out_4], Original ATen: [aten.relu]
triton_poi_fused_relu_1.run(buf3, primals_5, 512, grid=grid(512), stream=stream0)
del primals_5
buf4 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf3, reinterpret_tensor(primals_6, (128, 64), (1, 128), 0), out=buf4)
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [out_6], Original ATen: [aten.relu]
triton_poi_fused_relu_2.run(buf5, primals_7, 256, grid=grid(256), stream=stream0)
del primals_7
buf6 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [out_7], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_9, buf5, reinterpret_tensor(primals_8, (64, 4), (1, 64), 0), alpha=1, beta=1, out=buf6)
del primals_9
return (buf6, primals_1, buf1, buf3, buf5, 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, 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((128, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((128, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((64, 128), (128, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((64, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_8 = rand_strided((4, 64), (64, 1), device='cuda:0', dtype=torch.float32)
primals_9 = 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])
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 mlp_model(nn.Module):
def __init__(self, input_dim, hidden_dim, output_dim):
super(mlp_model, self).__init__()
self.fc1 = nn.Linear(input_dim, hidden_dim)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(hidden_dim, 128)
self.relu2 = nn.ReLU()
self.fc3 = nn.Linear(128, 64)
self.relu3 = nn.ReLU()
self.fc4 = nn.Linear(64, output_dim)
def forward(self, x):
out = x.view(-1, self.num_flat_features(x))
out = self.fc1(out)
out = self.relu1(out)
out = self.fc2(out)
out = self.relu2(out)
out = self.fc3(out)
out = self.relu3(out)
out = self.fc4(out)
return out
def num_flat_features(self, x):
size = x.size()[1:]
num_features = 1
for s in size:
num_features *= s
return num_features
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'input_dim': 4, 'hidden_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
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_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)
@triton.jit
def triton_poi_fused_relu_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
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)
@triton.jit
def triton_poi_fused_relu_2(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, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (128, 4), (4, 1))
assert_size_stride(primals_5, (128,), (1,))
assert_size_stride(primals_6, (64, 128), (128, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (4, 64), (64, 1))
assert_size_stride(primals_9, (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, 128), (128, 1), torch.float32)
extern_kernels.mm(buf1, reinterpret_tensor(primals_4, (4, 128), (1,
4), 0), out=buf2)
buf3 = buf2
del buf2
triton_poi_fused_relu_1[grid(512)](buf3, primals_5, 512, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
extern_kernels.mm(buf3, reinterpret_tensor(primals_6, (128, 64), (1,
128), 0), out=buf4)
buf5 = buf4
del buf4
triton_poi_fused_relu_2[grid(256)](buf5, primals_7, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_9, buf5, reinterpret_tensor(primals_8,
(64, 4), (1, 64), 0), alpha=1, beta=1, out=buf6)
del primals_9
return buf6, primals_1, buf1, buf3, buf5, primals_8, primals_6, primals_4
class mlp_modelNew(nn.Module):
def __init__(self, input_dim, hidden_dim, output_dim):
super(mlp_modelNew, self).__init__()
self.fc1 = nn.Linear(input_dim, hidden_dim)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(hidden_dim, 128)
self.relu2 = nn.ReLU()
self.fc3 = nn.Linear(128, 64)
self.relu3 = nn.ReLU()
self.fc4 = nn.Linear(64, output_dim)
def num_flat_features(self, x):
size = x.size()[1:]
num_features = 1
for s in size:
num_features *= s
return num_features
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_3 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.bias
primals_8 = self.fc4.weight
primals_9 = self.fc4.bias
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]
| st186/complementary_label_learning | mlp_model | false | 4,380 | [
"MIT"
] | 0 | 5d22ea638e9e6c087cc5bba7797c1c201679ba12 | https://github.com/st186/complementary_label_learning/tree/5d22ea638e9e6c087cc5bba7797c1c201679ba12 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, input_dim, hidden_dim, output_dim):
super().__init__()
self.fc1 = nn.Linear(input_dim, hidden_dim)
self.relu1 = nn.ReLU()
self.fc2 = nn.Linear(hidden_dim, 128)
self.relu2 = nn.ReLU()
self.fc3 = nn.Linear(128, 64)
self.relu3 = nn.ReLU()
self.fc4 = nn.Linear(64, output_dim)
def forward(self, x):
out = x.view(-1, self.num_flat_features(x))
out = self.fc1(out)
out = self.relu1(out)
out = self.fc2(out)
out = self.relu2(out)
out = self.fc3(out)
out = self.relu3(out)
out = self.fc4(out)
return out
def num_flat_features(self, x):
size = x.size()[1:]
num_features = 1
for s in size:
num_features *= s
return num_features
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
PrimaryCaps | # 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_7/inductor_cache/jo/cjoo56yo7awtt6oev6cvdfwedqercpy7m6xu45f3pq6md6tems26.py
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
# Source node to ATen node mapping:
# out => convolution
# Graph fragment:
# %convolution : [num_users=2] = call_function[target=torch.ops.aten.convolution.default](args = (%primals_1, %primals_2, %primals_3, [2, 2], [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=[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_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 = 50176
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 784) % 16
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_7/inductor_cache/pr/cprnjg45payb2mej3jdaa4qcbs5tr6mjibqm3lgom4km7bghpd24.py
# Topologically Sorted Source Nodes: [pow_1, v_length_sq, v_length, add, truediv, scaling_factor, out_4], Original ATen: [aten.pow, aten.sum, aten.sqrt, aten.add, aten.div, aten.mul]
# Source node to ATen node mapping:
# add => add
# out_4 => mul
# pow_1 => pow_1
# scaling_factor => div_1
# truediv => div
# v_length => sqrt
# v_length_sq => sum_1
# Graph fragment:
# %pow_1 : [num_users=1] = call_function[target=torch.ops.aten.pow.Tensor_Scalar](args = (%view_1, 2), kwargs = {})
# %sum_1 : [num_users=3] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%pow_1, [2], True), kwargs = {})
# %sqrt : [num_users=1] = call_function[target=torch.ops.aten.sqrt.default](args = (%sum_1,), 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 = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%div, %sqrt), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, %div_1), kwargs = {})
triton_poi_fused_add_div_mul_pow_sqrt_sum_1 = async_compile.triton('triton_poi_fused_add_div_mul_pow_sqrt_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=[16384, 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_div_mul_pow_sqrt_sum_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_add_div_mul_pow_sqrt_sum_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK : tl.constexpr, XBLOCK : tl.constexpr):
ynumel = 12544
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 % 3136
y1 = (yindex // 3136)
y3 = yindex
tmp0 = tl.load(in_ptr0 + ((784*x2) + (3136*(y0 // 784)) + (12544*y1) + (y0 % 784)), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + ((3136*(y0 // 784)) + (12544*y1) + (y0 % 784)), ymask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (784 + (3136*(y0 // 784)) + (12544*y1) + (y0 % 784)), ymask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1568 + (3136*(y0 // 784)) + (12544*y1) + (y0 % 784)), ymask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (2352 + (3136*(y0 // 784)) + (12544*y1) + (y0 % 784)), ymask, 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 = 1.0
tmp13 = tmp11 + tmp12
tmp14 = tmp11 / tmp13
tmp15 = libdevice.sqrt(tmp11)
tmp16 = tmp14 / tmp15
tmp17 = tmp0 * tmp16
tl.store(out_ptr0 + (x2 + (4*y3)), tmp17, xmask & ymask)
''', 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, 64, 64), (16384, 4096, 64, 1))
assert_size_stride(primals_2, (16, 4, 9, 9), (324, 81, 9, 1))
assert_size_stride(primals_3, (16, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
buf0 = extern_kernels.convolution(primals_1, primals_2, stride=(2, 2), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 28, 28), (12544, 784, 28, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [out], Original ATen: [aten.convolution]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_0.run(buf1, primals_3, 50176, grid=grid(50176), stream=stream0)
del primals_3
buf2 = empty_strided_cuda((4, 3136, 4), (12544, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [pow_1, v_length_sq, v_length, add, truediv, scaling_factor, out_4], Original ATen: [aten.pow, aten.sum, aten.sqrt, aten.add, aten.div, aten.mul]
triton_poi_fused_add_div_mul_pow_sqrt_sum_1.run(buf1, buf2, 12544, 4, grid=grid(12544, 4), stream=stream0)
return (buf2, primals_1, primals_2, 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, 64, 64), (16384, 4096, 64, 1), device='cuda:0', dtype=torch.float32)
primals_2 = rand_strided((16, 4, 9, 9), (324, 81, 9, 1), device='cuda:0', dtype=torch.float32)
primals_3 = rand_strided((16, ), (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 squash(x, dim=2):
v_length_sq = x.pow(2).sum(dim=dim, keepdim=True)
v_length = torch.sqrt(v_length_sq)
scaling_factor = v_length_sq / (1 + v_length_sq) / v_length
return x * scaling_factor
class PrimaryCaps(nn.Module):
"""
PrimaryCaps layers.
"""
def __init__(self, in_channels, out_capsules, out_capsule_dim,
kernel_size=9, stride=2):
super(PrimaryCaps, self).__init__()
self.in_channels = in_channels
self.out_capsules = out_capsules
self.out_capsule_dim = out_capsule_dim
self.capsules = nn.Conv2d(in_channels=in_channels, out_channels=
out_capsules * out_capsule_dim, kernel_size=kernel_size, stride
=stride, bias=True)
def forward(self, x):
"""
Revise based on adambielski's implementation.
ref: https://github.com/adambielski/CapsNet-pytorch/blob/master/net.py
"""
batch_size = x.size(0)
out = self.capsules(x)
_, _C, H, W = out.size()
out = out.view(batch_size, self.out_capsules, self.out_capsule_dim,
H, W)
out = out.permute(0, 1, 3, 4, 2).contiguous()
out = out.view(out.size(0), -1, out.size(4))
out = squash(out, dim=2)
return out
def get_inputs():
return [torch.rand([4, 4, 64, 64])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_capsules': 4, 'out_capsule_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 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_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 50176
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 784 % 16
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_add_div_mul_pow_sqrt_sum_1(in_ptr0, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 12544
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 % 3136
y1 = yindex // 3136
y3 = yindex
tmp0 = tl.load(in_ptr0 + (784 * x2 + 3136 * (y0 // 784) + 12544 * y1 +
y0 % 784), xmask & ymask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (3136 * (y0 // 784) + 12544 * y1 + y0 % 784),
ymask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (784 + 3136 * (y0 // 784) + 12544 * y1 + y0 %
784), ymask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1568 + 3136 * (y0 // 784) + 12544 * y1 + y0 %
784), ymask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (2352 + 3136 * (y0 // 784) + 12544 * y1 + y0 %
784), ymask, 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 = 1.0
tmp13 = tmp11 + tmp12
tmp14 = tmp11 / tmp13
tmp15 = libdevice.sqrt(tmp11)
tmp16 = tmp14 / tmp15
tmp17 = tmp0 * tmp16
tl.store(out_ptr0 + (x2 + 4 * y3), tmp17, xmask & ymask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 64, 64), (16384, 4096, 64, 1))
assert_size_stride(primals_2, (16, 4, 9, 9), (324, 81, 9, 1))
assert_size_stride(primals_3, (16,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_1, primals_2, stride=(2,
2), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 28, 28), (12544, 784, 28, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(50176)](buf1, primals_3, 50176,
XBLOCK=512, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((4, 3136, 4), (12544, 4, 1), torch.float32)
triton_poi_fused_add_div_mul_pow_sqrt_sum_1[grid(12544, 4)](buf1,
buf2, 12544, 4, XBLOCK=4, YBLOCK=256, num_warps=4, num_stages=1)
return buf2, primals_1, primals_2, buf1
def squash(x, dim=2):
v_length_sq = x.pow(2).sum(dim=dim, keepdim=True)
v_length = torch.sqrt(v_length_sq)
scaling_factor = v_length_sq / (1 + v_length_sq) / v_length
return x * scaling_factor
class PrimaryCapsNew(nn.Module):
"""
PrimaryCaps layers.
"""
def __init__(self, in_channels, out_capsules, out_capsule_dim,
kernel_size=9, stride=2):
super(PrimaryCapsNew, self).__init__()
self.in_channels = in_channels
self.out_capsules = out_capsules
self.out_capsule_dim = out_capsule_dim
self.capsules = nn.Conv2d(in_channels=in_channels, out_channels=
out_capsules * out_capsule_dim, kernel_size=kernel_size, stride
=stride, bias=True)
def forward(self, input_0):
primals_2 = self.capsules.weight
primals_3 = self.capsules.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| spikefairway/CapsNet-PyTorch | PrimaryCaps | false | 4,381 | [
"MIT"
] | 0 | 76aaabaad01283333a5f73a564cb1461449b4449 | https://github.com/spikefairway/CapsNet-PyTorch/tree/76aaabaad01283333a5f73a564cb1461449b4449 | import torch
import torch.nn as nn
def squash(x, dim=2):
v_length_sq = x.pow(2).sum(dim=dim, keepdim=True)
v_length = torch.sqrt(v_length_sq)
scaling_factor = v_length_sq / (1 + v_length_sq) / v_length
return x * scaling_factor
class Model(nn.Module):
"""
PrimaryCaps layers.
"""
def __init__(self, in_channels, out_capsules, out_capsule_dim,
kernel_size=9, stride=2):
super().__init__()
self.in_channels = in_channels
self.out_capsules = out_capsules
self.out_capsule_dim = out_capsule_dim
self.capsules = nn.Conv2d(in_channels=in_channels, out_channels=
out_capsules * out_capsule_dim, kernel_size=kernel_size, stride
=stride, bias=True)
def forward(self, x):
"""
Revise based on adambielski's implementation.
ref: https://github.com/adambielski/CapsNet-pytorch/blob/master/net.py
"""
batch_size = x.size(0)
out = self.capsules(x)
_, _C, H, W = out.size()
out = out.view(batch_size, self.out_capsules, self.out_capsule_dim,
H, W)
out = out.permute(0, 1, 3, 4, 2).contiguous()
out = out.view(out.size(0), -1, out.size(4))
out = squash(out, dim=2)
return out
def get_inputs():
return [torch.rand([4, 4, 64, 64])]
def get_init_inputs():
return [4, 4, 4]
|
DownRightShiftedConv2d | # 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_7/inductor_cache/2a/c2aucoxj5ek5nl3pl5n67fq4vdl55x2rw66tvv2aq23fsb3jvf5q.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, [3, 0, 3, 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=[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 = 784
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 7) % 7
x0 = xindex % 7
x2 = (xindex // 49)
x4 = xindex
tmp0 = (-3) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = (-3) + x0
tmp4 = tmp3 >= tmp1
tmp5 = tmp2 & tmp4
tmp6 = tl.load(in_ptr0 + ((-15) + x0 + (4*x1) + (16*x2)), tmp5 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp6, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/32/c32v7egt4mupqssam3gmac2qgv3ujprjybthsgweflmot256qqw7.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 = (%constant_pad_nd, %primals_2, %primals_3, [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=[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_convolution_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_1(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
x3 = xindex
x1 = (xindex // 16) % 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, 4, 4), (64, 16, 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, 7, 7), (196, 49, 7, 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, 784, grid=grid(784), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [conv2d], 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 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
triton_poi_fused_convolution_1.run(buf2, primals_3, 256, grid=grid(256), stream=stream0)
del primals_3
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, 4), (64, 16, 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)
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 DownRightShiftedConv2d(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((self.kernel_size[1] - 1, 0, self
.kernel_size[0] - 1, 0), 0.0)
def forward(self, x):
x = self.shift_pad(x)
return super().forward(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
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, xnumel, XBLOCK:
tl.constexpr):
xnumel = 784
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 7 % 7
x0 = xindex % 7
x2 = xindex // 49
x4 = xindex
tmp0 = -3 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = -3 + x0
tmp4 = tmp3 >= tmp1
tmp5 = tmp2 & tmp4
tmp6 = tl.load(in_ptr0 + (-15 + x0 + 4 * x1 + 16 * x2), tmp5 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp6, xmask)
@triton.jit
def triton_poi_fused_convolution_1(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
x3 = xindex
x1 = xindex // 16 % 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, 4, 4), (64, 16, 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, 7, 7), (196, 49, 7, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(784)](primals_1, buf0, 784,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_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 = buf1
del buf1
triton_poi_fused_convolution_1[grid(256)](buf2, primals_3, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_3
return buf2, primals_2, buf0
class DownRightShiftedConv2dNew(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((self.kernel_size[1] - 1, 0, self
.kernel_size[0] - 1, 0), 0.0)
def forward(self, input_0):
primals_1 = self.weight
primals_3 = self.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| stankevich-mipt/pixiv-tags-to-image | DownRightShiftedConv2d | false | 4,382 | [
"MIT"
] | 0 | 220a157956296c8a5b183ffe219e7c1929342c39 | https://github.com/stankevich-mipt/pixiv-tags-to-image/tree/220a157956296c8a5b183ffe219e7c1929342c39 | import torch
import torch.nn as nn
class Model(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((self.kernel_size[1] - 1, 0, self
.kernel_size[0] - 1, 0), 0.0)
def forward(self, x):
x = self.shift_pad(x)
return super().forward(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
OhemLoss | # 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_7/inductor_cache/f6/cf6msygnufcklcpvjj2djhdclyyqh6bvy7v6bxcvwogb4rkzv4nj.py
# Topologically Sorted Source Nodes: [loss], Original ATen: [aten.binary_cross_entropy]
# Source node to ATen node mapping:
# loss => full_default, full_default_1, log, log1p, maximum, maximum_1, mean, mul, mul_1, neg, sub, sub_1
# Graph fragment:
# %sub : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%view_1, 1), kwargs = {})
# %neg : [num_users=1] = call_function[target=torch.ops.aten.neg.default](args = (%view,), kwargs = {})
# %log1p : [num_users=1] = call_function[target=torch.ops.aten.log1p.default](args = (%neg,), kwargs = {})
# %full_default : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], -100), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %maximum : [num_users=1] = call_function[target=torch.ops.aten.maximum.default](args = (%log1p, %full_default), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%sub, %maximum), kwargs = {})
# %log : [num_users=1] = call_function[target=torch.ops.aten.log.default](args = (%view,), kwargs = {})
# %full_default_1 : [num_users=1] = call_function[target=torch.ops.aten.full.default](args = ([], -100), kwargs = {dtype: torch.float32, layout: torch.strided, device: cuda:0, pin_memory: False})
# %maximum_1 : [num_users=1] = call_function[target=torch.ops.aten.maximum.default](args = (%log, %full_default_1), kwargs = {})
# %mul_1 : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%view_1, %maximum_1), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%mul, %mul_1), kwargs = {})
# %mean : [num_users=1] = call_function[target=torch.ops.aten.mean.default](args = (%sub_1,), kwargs = {})
triton_per_fused_binary_cross_entropy_0 = async_compile.triton('triton_per_fused_binary_cross_entropy_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_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_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 = tmp0 - tmp1
tmp4 = -tmp3
tmp5 = libdevice.log1p(tmp4)
tmp6 = -100.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp2 * tmp7
tmp9 = tl_math.log(tmp3)
tmp10 = triton_helpers.maximum(tmp9, tmp6)
tmp11 = tmp0 * tmp10
tmp12 = tmp8 - 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: [loss], Original ATen: [aten.binary_cross_entropy]
stream0 = get_raw_stream(0)
triton_per_fused_binary_cross_entropy_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
class OhemLoss(nn.Module):
def __init__(self):
super(OhemLoss, self).__init__()
self.criteria = nn.BCELoss()
def forward(self, label_p, label_t):
label_p = label_p.view(-1)
label_t = label_t.view(-1)
loss = self.criteria(label_p, label_t)
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_per_fused_binary_cross_entropy_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 = tmp0 - tmp1
tmp4 = -tmp3
tmp5 = libdevice.log1p(tmp4)
tmp6 = -100.0
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp2 * tmp7
tmp9 = tl_math.log(tmp3)
tmp10 = triton_helpers.maximum(tmp9, tmp6)
tmp11 = tmp0 * tmp10
tmp12 = tmp8 - 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_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 OhemLossNew(nn.Module):
def __init__(self):
super(OhemLossNew, self).__init__()
self.criteria = nn.BCELoss()
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
| suifengwangshi/MotifC | OhemLoss | false | 4,383 | [
"Apache-2.0"
] | 0 | 34117a6bfb7dacd5a84da3abd5b8a339ae73cc76 | https://github.com/suifengwangshi/MotifC/tree/34117a6bfb7dacd5a84da3abd5b8a339ae73cc76 | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self):
super().__init__()
self.criteria = nn.BCELoss()
def forward(self, label_p, label_t):
label_p = label_p.view(-1)
label_t = label_t.view(-1)
loss = self.criteria(label_p, label_t)
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
EncoderBlock | # 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_7/inductor_cache/td/ctdybbibnws4d7ukbk3fpn35zkgapxylowdhzwx7vgsllncbdrxa.py
# Topologically Sorted Source Nodes: [input_1, input_2], Original ATen: [aten.convolution, aten.relu]
# Source node to ATen node mapping:
# input_1 => convolution
# input_2 => 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=[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_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 = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 16) % 4
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_7/inductor_cache/ha/chax2xzsjk5gzzvesrwcdefbdprendceelemsii2xegjpkxxd34i.py
# Topologically Sorted Source Nodes: [input_7, input_8, mul, add], Original ATen: [aten.convolution, aten.relu, aten.mul, aten.add, aten.threshold_backward]
# Source node to ATen node mapping:
# add => add
# input_7 => convolution_3
# input_8 => relu_3
# mul => mul
# Graph fragment:
# %convolution_3 : [num_users=1] = call_function[target=torch.ops.aten.convolution.default](args = (%relu_2, %primals_8, %primals_9, [1, 1], [1, 1], [1, 1], False, [0, 0], 1), kwargs = {})
# %relu_3 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%convolution_3,), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%relu_3, 1.0), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %mul), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_3, 0), kwargs = {})
triton_poi_fused_add_convolution_mul_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_add_convolution_mul_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=[256],
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_mul_relu_threshold_backward_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_convolution_mul_relu_threshold_backward_1(in_ptr0, in_ptr1, in_ptr2, 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
x3 = xindex
x1 = (xindex // 16) % 4
tmp0 = tl.load(in_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr1 + (x3), xmask)
tmp2 = tl.load(in_ptr2 + (x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 1.0
tmp7 = tmp5 * tmp6
tmp8 = tmp0 + tmp7
tmp9 = 0.0
tmp10 = tmp5 <= tmp9
tl.store(out_ptr0 + (x3), tmp8, xmask)
tl.store(out_ptr1 + (x3), 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, primals_8, primals_9 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 9, 3, 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, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (4, ), (1, ))
assert_size_stride(primals_6, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_7, (4, ), (1, ))
assert_size_stride(primals_8, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_9, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
# Topologically Sorted Source Nodes: [input_1], 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, 4, 4), (64, 16, 4, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [input_1, input_2], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 256, grid=grid(256), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [input_3], 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, 4, 4, 4), (64, 16, 4, 1))
buf3 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [input_3, input_4], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf3, primals_5, 256, grid=grid(256), stream=stream0)
del primals_5
# Topologically Sorted Source Nodes: [input_5], 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, 4, 4, 4), (64, 16, 4, 1))
buf5 = buf4; del buf4 # reuse
# Topologically Sorted Source Nodes: [input_5, input_6], Original ATen: [aten.convolution, aten.relu]
triton_poi_fused_convolution_relu_0.run(buf5, primals_7, 256, grid=grid(256), stream=stream0)
del primals_7
# Topologically Sorted Source Nodes: [input_7], 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, 4, 4, 4), (64, 16, 4, 1))
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [input_7, input_8, mul, add], Original ATen: [aten.convolution, aten.relu, aten.mul, aten.add, aten.threshold_backward]
triton_poi_fused_add_convolution_mul_relu_threshold_backward_1.run(primals_3, buf6, primals_9, buf7, buf8, 256, grid=grid(256), stream=stream0)
del buf6
del primals_9
return (buf7, primals_1, primals_3, primals_4, primals_6, primals_8, buf1, buf3, buf5, 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((4, 4, 3, 3), (36, 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)
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((4, 4, 3, 3), (36, 9, 3, 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, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_9 = 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])
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 collections import OrderedDict
class EncoderBlock(nn.Module):
def __init__(self, n_in, n_out, n_layers):
super().__init__()
self.n_in = n_in
self.n_out = n_out
self.n_hid = self.n_out
self.n_layers = n_layers
self.post_gain = 1.0
self.id_path = nn.Conv2d(self.n_in, self.n_out, 1
) if self.n_in != self.n_out else nn.Identity()
self.res_path = nn.Sequential(OrderedDict([('conv_1', nn.Conv2d(
self.n_in, self.n_hid, kernel_size=3, padding=1)), ('relu_1',
nn.ReLU()), ('conv_2', nn.Conv2d(self.n_hid, self.n_hid,
kernel_size=3, padding=1)), ('relu_2', nn.ReLU()), ('conv_3',
nn.Conv2d(self.n_hid, self.n_hid, kernel_size=3, padding=1)), (
'relu_3', nn.ReLU()), ('conv_4', nn.Conv2d(self.n_hid, self.
n_out, kernel_size=3, padding=1)), ('relu_4', nn.ReLU())]))
def forward(self, x: 'torch.Tensor') ->torch.Tensor:
return self.id_path(x) + self.post_gain * self.res_path(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'n_in': 4, 'n_out': 4, 'n_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 import triton_helpers
import torch.nn as nn
from collections import OrderedDict
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):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
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_add_convolution_mul_relu_threshold_backward_1(in_ptr0,
in_ptr1, in_ptr2, 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
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + x3, xmask)
tmp2 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 1.0
tmp7 = tmp5 * tmp6
tmp8 = tmp0 + tmp7
tmp9 = 0.0
tmp10 = tmp5 <= tmp9
tl.store(out_ptr0 + x3, tmp8, xmask)
tl.store(out_ptr1 + x3, tmp10, 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, 4, 3, 3), (36, 9, 3, 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, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_9, (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, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(256)](buf1, primals_2, 256,
XBLOCK=128, 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, 4, 4, 4), (64, 16, 4, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_0[grid(256)](buf3, primals_5, 256,
XBLOCK=128, 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, 4, 4, 4), (64, 16, 4, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_0[grid(256)](buf5, primals_7, 256,
XBLOCK=128, 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, 4, 4, 4), (64, 16, 4, 1))
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_convolution_mul_relu_threshold_backward_1[grid
(256)](primals_3, buf6, primals_9, buf7, buf8, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf6
del primals_9
return (buf7, primals_1, primals_3, primals_4, primals_6, primals_8,
buf1, buf3, buf5, buf8)
class EncoderBlockNew(nn.Module):
def __init__(self, n_in, n_out, n_layers):
super().__init__()
self.n_in = n_in
self.n_out = n_out
self.n_hid = self.n_out
self.n_layers = n_layers
self.post_gain = 1.0
self.id_path = nn.Conv2d(self.n_in, self.n_out, 1
) if self.n_in != self.n_out else nn.Identity()
self.res_path = nn.Sequential(OrderedDict([('conv_1', nn.Conv2d(
self.n_in, self.n_hid, kernel_size=3, padding=1)), ('relu_1',
nn.ReLU()), ('conv_2', nn.Conv2d(self.n_hid, self.n_hid,
kernel_size=3, padding=1)), ('relu_2', nn.ReLU()), ('conv_3',
nn.Conv2d(self.n_hid, self.n_hid, kernel_size=3, padding=1)), (
'relu_3', nn.ReLU()), ('conv_4', nn.Conv2d(self.n_hid, self.
n_out, kernel_size=3, padding=1)), ('relu_4', nn.ReLU())]))
def forward(self, input_0):
primals_1 = self.res_path.conv_1.weight
primals_2 = self.res_path.conv_1.bias
primals_4 = self.res_path.conv_2.weight
primals_5 = self.res_path.conv_2.bias
primals_6 = self.res_path.conv_3.weight
primals_7 = self.res_path.conv_3.bias
primals_8 = self.res_path.conv_4.weight
primals_9 = self.res_path.conv_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])
return output[0]
| stankevich-mipt/pixiv-tags-to-image | EncoderBlock | false | 4,384 | [
"MIT"
] | 0 | 220a157956296c8a5b183ffe219e7c1929342c39 | https://github.com/stankevich-mipt/pixiv-tags-to-image/tree/220a157956296c8a5b183ffe219e7c1929342c39 | import torch
import torch.nn as nn
from collections import OrderedDict
class Model(nn.Module):
def __init__(self, n_in, n_out, n_layers):
super().__init__()
self.n_in = n_in
self.n_out = n_out
self.n_hid = self.n_out
self.n_layers = n_layers
self.post_gain = 1.0
self.id_path = nn.Conv2d(self.n_in, self.n_out, 1
) if self.n_in != self.n_out else nn.Identity()
self.res_path = nn.Sequential(OrderedDict([('conv_1', nn.Conv2d(
self.n_in, self.n_hid, kernel_size=3, padding=1)), ('relu_1',
nn.ReLU()), ('conv_2', nn.Conv2d(self.n_hid, self.n_hid,
kernel_size=3, padding=1)), ('relu_2', nn.ReLU()), ('conv_3',
nn.Conv2d(self.n_hid, self.n_hid, kernel_size=3, padding=1)), (
'relu_3', nn.ReLU()), ('conv_4', nn.Conv2d(self.n_hid, self.
n_out, kernel_size=3, padding=1)), ('relu_4', nn.ReLU())]))
def forward(self, x: 'torch.Tensor') ->torch.Tensor:
return self.id_path(x) + self.post_gain * self.res_path(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 1]
|
DownShiftedConv2d | # 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_7/inductor_cache/ns/cnsfp7p7oqys7oiccw5stwjuopxm6r5oeetrumxribuqt3olqnp6.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, 3, 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=[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 = 672
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = (xindex // 6) % 7
x0 = xindex % 6
x2 = (xindex // 42)
x4 = xindex
tmp0 = (-3) + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = (-1) + x0
tmp4 = tmp3 >= tmp1
tmp5 = tl.full([1], 4, tl.int64)
tmp6 = tmp3 < tmp5
tmp7 = tmp2 & tmp4
tmp8 = tmp7 & tmp6
tmp9 = tl.load(in_ptr0 + ((-13) + x0 + (4*x1) + (16*x2)), tmp8 & xmask, other=0.0)
tl.store(out_ptr0 + (x4), tmp9, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ty/ctyiqywk6eb3ghqv2mg5w772jefcoyl2pv657xgca54xkil5ukrm.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 = (%constant_pad_nd, %primals_2, %primals_3, [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=[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_convolution_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_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 12) % 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, 4, 4), (64, 16, 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, 7, 6), (168, 42, 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, 672, grid=grid(672), stream=stream0)
del primals_1
# Topologically Sorted Source Nodes: [conv2d], 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, 3), (48, 12, 3, 1))
buf2 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [conv2d], Original ATen: [aten.convolution]
triton_poi_fused_convolution_1.run(buf2, primals_3, 192, grid=grid(192), stream=stream0)
del primals_3
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, 4), (64, 16, 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)
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 DownShiftedConv2d(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((int((self.kernel_size[1] - 1) //
2), int((self.kernel_size[1] - 1) // 2), self.kernel_size[0] -
1, 0), 0.0)
def forward(self, x):
x = self.shift_pad(x)
return super().forward(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
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, xnumel, XBLOCK:
tl.constexpr):
xnumel = 672
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 6 % 7
x0 = xindex % 6
x2 = xindex // 42
x4 = xindex
tmp0 = -3 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = -1 + x0
tmp4 = tmp3 >= tmp1
tmp5 = tl.full([1], 4, tl.int64)
tmp6 = tmp3 < tmp5
tmp7 = tmp2 & tmp4
tmp8 = tmp7 & tmp6
tmp9 = tl.load(in_ptr0 + (-13 + x0 + 4 * x1 + 16 * x2), tmp8 & xmask,
other=0.0)
tl.store(out_ptr0 + x4, tmp9, xmask)
@triton.jit
def triton_poi_fused_convolution_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 12 % 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, 4, 4), (64, 16, 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, 7, 6), (168, 42, 6, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(672)](primals_1, buf0, 672,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_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, 3), (48, 12, 3, 1))
buf2 = buf1
del buf1
triton_poi_fused_convolution_1[grid(192)](buf2, primals_3, 192,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_3
return buf2, primals_2, buf0
class DownShiftedConv2dNew(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((int((self.kernel_size[1] - 1) //
2), int((self.kernel_size[1] - 1) // 2), self.kernel_size[0] -
1, 0), 0.0)
def forward(self, input_0):
primals_1 = self.weight
primals_3 = self.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
| stankevich-mipt/pixiv-tags-to-image | DownShiftedConv2d | false | 4,385 | [
"MIT"
] | 0 | 220a157956296c8a5b183ffe219e7c1929342c39 | https://github.com/stankevich-mipt/pixiv-tags-to-image/tree/220a157956296c8a5b183ffe219e7c1929342c39 | import torch
import torch.nn as nn
class Model(nn.Conv2d):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.shift_pad = nn.ConstantPad2d((int((self.kernel_size[1] - 1) //
2), int((self.kernel_size[1] - 1) // 2), self.kernel_size[0] -
1, 0), 0.0)
def forward(self, x):
x = self.shift_pad(x)
return super().forward(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
StatsPool | # 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_7/inductor_cache/eo/ceol4pvopahhfr62gdumixobymkaolmwvai66o7pjybv57v53l22.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 = ([%mean, %sqrt], 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 // 4) % 8
x0 = xindex % 4
x2 = (xindex // 32)
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 = tl.load(in_ptr0 + (4 + x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl.load(in_ptr0 + (8 + x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp9 = tmp7 + tmp8
tmp10 = tl.load(in_ptr0 + (12 + x0 + (16*x1) + (64*x2)), tmp4 & xmask, other=0.0)
tmp11 = tmp9 + tmp10
tmp12 = 4.0
tmp13 = tmp11 / tmp12
tmp14 = tl.full(tmp13.shape, 0.0, tmp13.dtype)
tmp15 = tl.where(tmp4, tmp13, tmp14)
tmp16 = tmp0 >= tmp3
tmp17 = tl.full([1], 8, tl.int64)
tmp18 = tmp0 < tmp17
tmp19 = tl.load(in_ptr0 + (x0 + (16*((-4) + x1)) + (64*x2)), tmp16 & xmask, other=0.0)
tmp20 = tl.load(in_ptr0 + (4 + x0 + (16*((-4) + x1)) + (64*x2)), tmp16 & xmask, other=0.0)
tmp21 = tmp19 + tmp20
tmp22 = tl.load(in_ptr0 + (8 + x0 + (16*((-4) + x1)) + (64*x2)), tmp16 & xmask, other=0.0)
tmp23 = tmp21 + tmp22
tmp24 = tl.load(in_ptr0 + (12 + x0 + (16*((-4) + x1)) + (64*x2)), tmp16 & xmask, other=0.0)
tmp25 = tmp23 + tmp24
tmp26 = tmp25 / tmp12
tmp27 = tmp19 - tmp26
tmp28 = tmp27 * tmp27
tmp29 = tmp20 - tmp26
tmp30 = tmp29 * tmp29
tmp31 = tmp28 + tmp30
tmp32 = tmp22 - tmp26
tmp33 = tmp32 * tmp32
tmp34 = tmp31 + tmp33
tmp35 = tmp24 - tmp26
tmp36 = tmp35 * tmp35
tmp37 = tmp34 + tmp36
tmp38 = 3.0
tmp39 = tmp37 / tmp38
tmp40 = libdevice.sqrt(tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp16, tmp40, tmp41)
tmp43 = tl.where(tmp4, tmp15, tmp42)
tl.store(out_ptr0 + (x3), tmp43, 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, 8, 4), (32, 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 warnings
import torch.nn as nn
from typing import Optional
import torch.optim
import torch.nn.functional as F
class StatsPool(nn.Module):
"""Statistics pooling
Compute temporal mean and (unbiased) standard deviation
and returns their concatenation.
Reference
---------
https://en.wikipedia.org/wiki/Weighted_arithmetic_mean
"""
def forward(self, sequences: 'torch.Tensor', weights:
'Optional[torch.Tensor]'=None) ->torch.Tensor:
"""Forward pass
Parameters
----------
sequences : (batch, channel, frames) torch.Tensor
Sequences.
weights : (batch, frames) torch.Tensor, optional
When provided, compute weighted mean and standard deviation.
Returns
-------
output : (batch, 2 * channel) torch.Tensor
Concatenation of mean and (unbiased) standard deviation.
"""
if weights is None:
mean = sequences.mean(dim=2)
std = sequences.std(dim=2, unbiased=True)
else:
weights = weights.unsqueeze(dim=1)
num_frames = sequences.shape[2]
num_weights = weights.shape[2]
if num_frames != num_weights:
warnings.warn(
f'Mismatch between frames ({num_frames}) and weights ({num_weights}) numbers.'
)
weights = F.interpolate(weights, size=num_frames, mode=
'linear', align_corners=False)
v1 = weights.sum(dim=2)
mean = torch.sum(sequences * weights, dim=2) / v1
dx2 = torch.square(sequences - mean.unsqueeze(2))
v2 = torch.square(weights).sum(dim=2)
var = torch.sum(dx2 * weights, dim=2) / (v1 - v2 / v1)
std = torch.sqrt(var)
return torch.cat([mean, std], 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.triton_helpers import libdevice
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
@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 // 4 % 8
x0 = xindex % 4
x2 = xindex // 32
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 = tl.load(in_ptr0 + (4 + x0 + 16 * x1 + 64 * x2), tmp4 & xmask,
other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl.load(in_ptr0 + (8 + x0 + 16 * x1 + 64 * x2), tmp4 & xmask,
other=0.0)
tmp9 = tmp7 + tmp8
tmp10 = tl.load(in_ptr0 + (12 + x0 + 16 * x1 + 64 * x2), tmp4 & xmask,
other=0.0)
tmp11 = tmp9 + tmp10
tmp12 = 4.0
tmp13 = tmp11 / tmp12
tmp14 = tl.full(tmp13.shape, 0.0, tmp13.dtype)
tmp15 = tl.where(tmp4, tmp13, tmp14)
tmp16 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp19 = tl.load(in_ptr0 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp16 &
xmask, other=0.0)
tmp20 = tl.load(in_ptr0 + (4 + x0 + 16 * (-4 + x1) + 64 * x2), tmp16 &
xmask, other=0.0)
tmp21 = tmp19 + tmp20
tmp22 = tl.load(in_ptr0 + (8 + x0 + 16 * (-4 + x1) + 64 * x2), tmp16 &
xmask, other=0.0)
tmp23 = tmp21 + tmp22
tmp24 = tl.load(in_ptr0 + (12 + x0 + 16 * (-4 + x1) + 64 * x2), tmp16 &
xmask, other=0.0)
tmp25 = tmp23 + tmp24
tmp26 = tmp25 / tmp12
tmp27 = tmp19 - tmp26
tmp28 = tmp27 * tmp27
tmp29 = tmp20 - tmp26
tmp30 = tmp29 * tmp29
tmp31 = tmp28 + tmp30
tmp32 = tmp22 - tmp26
tmp33 = tmp32 * tmp32
tmp34 = tmp31 + tmp33
tmp35 = tmp24 - tmp26
tmp36 = tmp35 * tmp35
tmp37 = tmp34 + tmp36
tmp38 = 3.0
tmp39 = tmp37 / tmp38
tmp40 = libdevice.sqrt(tmp39)
tmp41 = tl.full(tmp40.shape, 0.0, tmp40.dtype)
tmp42 = tl.where(tmp16, tmp40, tmp41)
tmp43 = tl.where(tmp4, tmp15, tmp42)
tl.store(out_ptr0 + x3, tmp43, 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, 8, 4), (32, 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 StatsPoolNew(nn.Module):
"""Statistics pooling
Compute temporal mean and (unbiased) standard deviation
and returns their concatenation.
Reference
---------
https://en.wikipedia.org/wiki/Weighted_arithmetic_mean
"""
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| suissemaxx/pyannote-audio-develop_colab | StatsPool | false | 4,386 | [
"MIT"
] | 0 | e9499372a1771c21e1604424a6dd041337111093 | https://github.com/suissemaxx/pyannote-audio-develop_colab/tree/e9499372a1771c21e1604424a6dd041337111093 | import torch
import warnings
import torch.nn as nn
from typing import Optional
import torch.optim
import torch.nn.functional as F
class Model(nn.Module):
"""Statistics pooling
Compute temporal mean and (unbiased) standard deviation
and returns their concatenation.
Reference
---------
https://en.wikipedia.org/wiki/Weighted_arithmetic_mean
"""
def forward(self, sequences: 'torch.Tensor', weights:
'Optional[torch.Tensor]'=None) ->torch.Tensor:
"""Forward pass
Parameters
----------
sequences : (batch, channel, frames) torch.Tensor
Sequences.
weights : (batch, frames) torch.Tensor, optional
When provided, compute weighted mean and standard deviation.
Returns
-------
output : (batch, 2 * channel) torch.Tensor
Concatenation of mean and (unbiased) standard deviation.
"""
if weights is None:
mean = sequences.mean(dim=2)
std = sequences.std(dim=2, unbiased=True)
else:
weights = weights.unsqueeze(dim=1)
num_frames = sequences.shape[2]
num_weights = weights.shape[2]
if num_frames != num_weights:
warnings.warn(
f'Mismatch between frames ({num_frames}) and weights ({num_weights}) numbers.'
)
weights = F.interpolate(weights, size=num_frames, mode=
'linear', align_corners=False)
v1 = weights.sum(dim=2)
mean = torch.sum(sequences * weights, dim=2) / v1
dx2 = torch.square(sequences - mean.unsqueeze(2))
v2 = torch.square(weights).sum(dim=2)
var = torch.sum(dx2 * weights, dim=2) / (v1 - v2 / v1)
std = torch.sqrt(var)
return torch.cat([mean, std], dim=1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
ConvRelu | # 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_7/inductor_cache/af/cafjp5uw3trbmdrcdgfueohgc3x4wmjo5mg44oojk6wkeq5n6mlf.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.leaky_relu, aten.leaky_relu_backward]
# Source node to ATen node mapping:
# x => convolution
# x_1 => gt, mul, where
# Graph fragment:
# %convolution : [num_users=3] = 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 = {})
# %gt : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%convolution, 0), kwargs = {})
# %mul : [num_users=1] = call_function[target=torch.ops.aten.mul.Tensor](args = (%convolution, 0.01), kwargs = {})
# %where : [num_users=2] = call_function[target=torch.ops.aten.where.self](args = (%gt, %convolution, %mul), kwargs = {})
# %gt_1 : [num_users=1] = call_function[target=torch.ops.aten.gt.Scalar](args = (%where, 0), kwargs = {})
triton_poi_fused_convolution_leaky_relu_leaky_relu_backward_0 = async_compile.triton('triton_poi_fused_convolution_leaky_relu_leaky_relu_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_convolution_leaky_relu_leaky_relu_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_leaky_relu_leaky_relu_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
x3 = xindex
x1 = (xindex // 16) % 4
tmp0 = tl.load(in_out_ptr0 + (x3), xmask)
tmp1 = tl.load(in_ptr0 + (x1), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tmp8 = tmp7 > tmp3
tl.store(in_out_ptr0 + (x3), tmp7, xmask)
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 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 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(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, 4, 4), (64, 16, 4, 1))
buf1 = buf0; del buf0 # reuse
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.leaky_relu, aten.leaky_relu_backward]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_leaky_relu_leaky_relu_backward_0.run(buf1, primals_2, buf2, 256, grid=grid(256), 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, 3, 3), (36, 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
class ConvRelu(nn.Module):
def __init__(self, in_, out):
super().__init__()
self.conv = nn.Conv2d(in_, out, 3, padding=1)
self.activation = nn.LeakyReLU(inplace=True)
def forward(self, x):
x = self.conv(x)
x = self.activation(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_': 4, 'out': 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
@triton.jit
def triton_poi_fused_convolution_leaky_relu_leaky_relu_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
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tmp8 = tmp7 > tmp3
tl.store(in_out_ptr0 + x3, tmp7, xmask)
tl.store(out_ptr0 + x3, tmp8, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 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(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, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_convolution_leaky_relu_leaky_relu_backward_0[grid(256)
](buf1, primals_2, buf2, 256, XBLOCK=256, num_warps=4, num_stages=1
)
del primals_2
return buf1, primals_1, primals_3, buf2
class ConvReluNew(nn.Module):
def __init__(self, in_, out):
super().__init__()
self.conv = nn.Conv2d(in_, out, 3, padding=1)
self.activation = nn.LeakyReLU(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]
| sudonull1/Crack-Segmentation | ConvRelu | false | 4,387 | [
"MIT"
] | 0 | 640f86839ce5d79b48916b176caf8ad83c7355ae | https://github.com/sudonull1/Crack-Segmentation/tree/640f86839ce5d79b48916b176caf8ad83c7355ae | import torch
import torch.nn as nn
class Model(nn.Module):
def __init__(self, in_, out):
super().__init__()
self.conv = nn.Conv2d(in_, out, 3, padding=1)
self.activation = nn.LeakyReLU(inplace=True)
def forward(self, x):
x = self.conv(x)
x = self.activation(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
fire | # 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_7/inductor_cache/td/ctdybbibnws4d7ukbk3fpn35zkgapxylowdhzwx7vgsllncbdrxa.py
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
# 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=3] = 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=[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_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 = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = (xindex // 16) % 4
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_7/inductor_cache/ak/cakz5alakpulryqj4tcuaom3i64qsmr7djmmr6wgqwea5fak5urc.py
# Topologically Sorted Source Nodes: [out, out_1], Original ATen: [aten.cat, aten.relu, aten.threshold_backward]
# Source node to ATen node mapping:
# out => cat
# out_1 => relu_1
# Graph fragment:
# %cat : [num_users=1] = call_function[target=torch.ops.aten.cat.default](args = ([%convolution_1, %convolution_2], 1), kwargs = {})
# %relu_1 : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%cat,), kwargs = {})
# %le : [num_users=1] = call_function[target=torch.ops.aten.le.Scalar](args = (%relu_1, 0), kwargs = {})
triton_poi_fused_cat_relu_threshold_backward_1 = async_compile.triton('triton_poi_fused_cat_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=[256],
filename=__file__,
triton_meta={'signature': {0: '*fp32', 1: '*fp32', 2: '*fp32', 3: '*fp32', 4: '*fp32', 5: '*i1', 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_relu_threshold_backward_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_cat_relu_threshold_backward_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3, 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
x1 = (xindex // 16) % 4
x0 = xindex % 16
x2 = (xindex // 64)
x3 = xindex
tmp0 = x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + (16*x1) + (32*x2)), tmp4 & xmask, other=0.0)
tmp6 = tl.load(in_ptr1 + (x1), tmp4 & xmask, eviction_policy='evict_last', 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], 4, tl.int64)
tmp12 = tmp0 < tmp11
tmp13 = tl.load(in_ptr2 + (x0 + (16*((-2) + x1)) + (32*x2)), tmp10 & xmask, other=0.0)
tmp14 = tl.load(in_ptr3 + ((-2) + x1), tmp10 & xmask, eviction_policy='evict_last', 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.full([1], 0, tl.int32)
tmp20 = triton_helpers.maximum(tmp19, tmp18)
tmp21 = 0.0
tmp22 = tmp20 <= tmp21
tl.store(out_ptr0 + (x3), tmp20, xmask)
tl.store(out_ptr1 + (x3), tmp22, 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, 1, 1), (4, 1, 1, 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, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (2, ), (1, ))
assert_size_stride(primals_6, (2, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_7, (2, ), (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, 4, 4), (64, 16, 4, 1))
buf1 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x, x_1], Original ATen: [aten.convolution, aten.relu]
stream0 = get_raw_stream(0)
triton_poi_fused_convolution_relu_0.run(buf1, primals_2, 256, grid=grid(256), stream=stream0)
del primals_2
# Topologically Sorted Source Nodes: [out1], Original ATen: [aten.convolution]
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 2, 4, 4), (32, 16, 4, 1))
# Topologically Sorted Source Nodes: [out2], Original ATen: [aten.convolution]
buf3 = extern_kernels.convolution(buf1, primals_6, stride=(1, 1), padding=(1, 1), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 2, 4, 4), (32, 16, 4, 1))
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
# Topologically Sorted Source Nodes: [out, out_1], Original ATen: [aten.cat, aten.relu, aten.threshold_backward]
triton_poi_fused_cat_relu_threshold_backward_1.run(buf2, primals_5, buf3, primals_7, buf4, buf5, 256, grid=grid(256), stream=stream0)
del buf2
del buf3
del primals_5
del primals_7
return (buf4, primals_1, primals_3, primals_4, primals_6, buf1, 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, 1, 1), (4, 1, 1, 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((2, 4, 1, 1), (4, 1, 1, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((2, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((2, 4, 3, 3), (36, 9, 3, 1), device='cuda:0', dtype=torch.float32)
primals_7 = rand_strided((2, ), (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 itertools import product as product
import torch.nn as nn
class fire(nn.Module):
def __init__(self, inplanes, squeeze_planes, expand_planes, st=1):
super(fire, self).__init__()
self.conv1 = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1,
stride=1)
self.relu1 = nn.ReLU(inplace=True)
self.conv2 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=1, stride=st)
self.conv3 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=3, stride=st, padding=1)
self.relu2 = nn.ReLU(inplace=True)
def forward(self, x):
x = self.conv1(x)
x = self.relu1(x)
out1 = self.conv2(x)
out2 = self.conv3(x)
out = torch.cat([out1, out2], 1)
out = self.relu2(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'inplanes': 4, 'squeeze_planes': 4, 'expand_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 itertools import product as product
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_convolution_relu_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
x3 = xindex
x1 = xindex // 16 % 4
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_cat_relu_threshold_backward_1(in_ptr0, in_ptr1,
in_ptr2, in_ptr3, 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
x1 = xindex // 16 % 4
x0 = xindex % 16
x2 = xindex // 64
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 2, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 32 * x2), tmp4 & xmask, other=0.0)
tmp6 = tl.load(in_ptr1 + x1, tmp4 & xmask, eviction_policy='evict_last',
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], 4, tl.int64)
tmp13 = tl.load(in_ptr2 + (x0 + 16 * (-2 + x1) + 32 * x2), tmp10 &
xmask, other=0.0)
tmp14 = tl.load(in_ptr3 + (-2 + x1), tmp10 & xmask, eviction_policy=
'evict_last', 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.full([1], 0, tl.int32)
tmp20 = triton_helpers.maximum(tmp19, tmp18)
tmp21 = 0.0
tmp22 = tmp20 <= tmp21
tl.store(out_ptr0 + x3, tmp20, xmask)
tl.store(out_ptr1 + x3, tmp22, 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, 1, 1), (4, 1, 1, 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, (2, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_5, (2,), (1,))
assert_size_stride(primals_6, (2, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_7, (2,), (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, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(256)](buf1, primals_2, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 2, 4, 4), (32, 16, 4, 1))
buf3 = extern_kernels.convolution(buf1, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 2, 4, 4), (32, 16, 4, 1))
buf4 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_cat_relu_threshold_backward_1[grid(256)](buf2,
primals_5, buf3, primals_7, buf4, buf5, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del buf2
del buf3
del primals_5
del primals_7
return buf4, primals_1, primals_3, primals_4, primals_6, buf1, buf5
class fireNew(nn.Module):
def __init__(self, inplanes, squeeze_planes, expand_planes, st=1):
super(fireNew, self).__init__()
self.conv1 = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1,
stride=1)
self.relu1 = nn.ReLU(inplace=True)
self.conv2 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=1, stride=st)
self.conv3 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=3, stride=st, padding=1)
self.relu2 = 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.conv3.weight
primals_7 = self.conv3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
| suiguoxin/Pytorch_Retinaface | fire | false | 4,388 | [
"MIT"
] | 0 | d9393bad43103635261b4ec5b03f20e79931d0da | https://github.com/suiguoxin/Pytorch_Retinaface/tree/d9393bad43103635261b4ec5b03f20e79931d0da | import torch
from itertools import product as product
import torch.nn as nn
class Model(nn.Module):
def __init__(self, inplanes, squeeze_planes, expand_planes, st=1):
super().__init__()
self.conv1 = nn.Conv2d(inplanes, squeeze_planes, kernel_size=1,
stride=1)
self.relu1 = nn.ReLU(inplace=True)
self.conv2 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=1, stride=st)
self.conv3 = nn.Conv2d(squeeze_planes, int(expand_planes / 2),
kernel_size=3, stride=st, padding=1)
self.relu2 = nn.ReLU(inplace=True)
def forward(self, x):
x = self.conv1(x)
x = self.relu1(x)
out1 = self.conv2(x)
out2 = self.conv3(x)
out = torch.cat([out1, out2], 1)
out = self.relu2(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
Attn | # 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_7/inductor_cache/r6/cr6neze6yovkog6kjrk5k2db63h47ozkojywfys6karxe7dlumrz.py
# Topologically Sorted Source Nodes: [attn_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# attn_weights => amax, exp, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%bmm, [2], 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_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_7/inductor_cache/kj/ckjtlefzavjukjsytvkak6ek26zmzexpcbnlwelx4k5kascjxlf3.py
# Topologically Sorted Source Nodes: [attn_weights], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# attn_weights => div, sum_1
# Graph fragment:
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [2], True), kwargs = {})
# %div : [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=[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):
primals_1, primals_2, primals_3, primals_4 = 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), (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: [energy], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_3, reinterpret_tensor(primals_4, (16, 4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf0)
del primals_2
del primals_3
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [attn_energies], Original ATen: [aten.bmm]
extern_kernels.bmm(primals_1, reinterpret_tensor(buf0, (4, 4, 4), (16, 1, 4), 0), out=buf1)
buf2 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [attn_weights], Original ATen: [aten._softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__softmax_0.run(buf1, buf2, 64, grid=grid(64), stream=stream0)
buf3 = buf1; del buf1 # reuse
# Topologically Sorted Source Nodes: [attn_weights], Original ATen: [aten._softmax]
triton_poi_fused__softmax_1.run(buf2, buf3, 64, grid=grid(64), stream=stream0)
buf4 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [context], Original ATen: [aten.bmm]
extern_kernels.bmm(buf3, primals_4, out=buf4)
return (buf4, buf3, primals_4, buf3, reinterpret_tensor(primals_1, (4, 4, 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, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, 4, 4), (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
from torch import nn
class Attn(torch.nn.Module):
"""
Attention:
feature_dim: dimension of feature embedding
method: method to calculate attention, (general, dot, concat)
input_dim: dimension of input embedding, default is the same as feature_dim; method dot is only available when input_dim == feature_dim
Inputs:
inputs: batch of inputs; the inp_size is optional; shape=(batch_size, inp_size, feature_dim)
targets: batch of targets to pay attention; shape=(batch_size, tgt_size, feature_dim)
mask: optional target binary mask to avoid paying attention to padding item; shape=(batch_size, tgt_size)
Outputs:
context: context vector computed as the weighted average of all the encoder outputs; inp_size is optional;shape=(batch_size, inp_size, feature_dim)
attntion: attention weight paid to the targets; shape=(batch_size, inp_size, tgt_size)
"""
def __init__(self, feature_dim, method='general', input_dim=None):
super(Attn, self).__init__()
self.method = method
if not input_dim:
input_dim = feature_dim
if self.method not in ['dot', 'general', 'concat']:
raise ValueError(self.method,
'is not an appropriate attention method.')
elif self.method == 'dot' and input_dim != feature_dim:
raise ValueError(
'dot does not work when input_dim does not equals to feature_dim'
)
if self.method == 'general':
self.attn = torch.nn.Linear(feature_dim, input_dim)
elif self.method == 'concat':
self.attn = torch.nn.Linear(input_dim + feature_dim, feature_dim)
self.v = torch.nn.Parameter(torch.FloatTensor(feature_dim))
def score(self, inputs, targets):
if self.method == 'dot':
return inputs.bmm(targets.transpose(1, 2))
elif self.method == 'general':
energy = self.attn(targets)
return inputs.bmm(energy.transpose(1, 2))
elif self.method == 'concat':
inp_size = inputs.size(1)
tgt_size = targets.size(1)
inputs_exp = inputs.unsqueeze(2).expand(-1, -1, tgt_size, -1)
targets_exp = targets.unsqueeze(1).expand(-1, inp_size, -1, -1)
combined = torch.cat((inputs_exp, targets_exp), 3)
energy = self.attn(combined).tanh()
return torch.sum(self.v * energy, dim=3)
def forward(self, inputs, targets, mask=None):
inp_shape = inputs.size()
if len(inp_shape) == 2:
inputs = inputs.view(inp_shape[0], 1, inp_shape[-1])
attn_energies = self.score(inputs, targets)
if mask is not None:
mask = mask.unsqueeze(1).expand(-1, attn_energies.size(1), -1)
attn_energies = attn_energies.masked_fill(~mask, float('-inf'))
attn_weights = nn.functional.softmax(attn_energies, dim=2)
context = attn_weights.bmm(targets)
if len(inp_shape) == 2:
context = context.squeeze(1)
attn_weights = attn_weights.squeeze(1)
return context, attn_weights
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'feature_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
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)
def call(args):
primals_1, primals_2, primals_3, primals_4 = 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), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(primals_4, (16,
4), (4, 1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_2
del primals_3
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(primals_1, reinterpret_tensor(buf0, (4, 4, 4), (
16, 1, 4), 0), out=buf1)
buf2 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(64)](buf1, buf2, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf3 = buf1
del buf1
triton_poi_fused__softmax_1[grid(64)](buf2, buf3, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf4 = buf2
del buf2
extern_kernels.bmm(buf3, primals_4, out=buf4)
return buf4, buf3, primals_4, buf3, reinterpret_tensor(primals_1, (4, 4,
4), (16, 1, 4), 0)
class AttnNew(torch.nn.Module):
"""
Attention:
feature_dim: dimension of feature embedding
method: method to calculate attention, (general, dot, concat)
input_dim: dimension of input embedding, default is the same as feature_dim; method dot is only available when input_dim == feature_dim
Inputs:
inputs: batch of inputs; the inp_size is optional; shape=(batch_size, inp_size, feature_dim)
targets: batch of targets to pay attention; shape=(batch_size, tgt_size, feature_dim)
mask: optional target binary mask to avoid paying attention to padding item; shape=(batch_size, tgt_size)
Outputs:
context: context vector computed as the weighted average of all the encoder outputs; inp_size is optional;shape=(batch_size, inp_size, feature_dim)
attntion: attention weight paid to the targets; shape=(batch_size, inp_size, tgt_size)
"""
def __init__(self, feature_dim, method='general', input_dim=None):
super(AttnNew, self).__init__()
self.method = method
if not input_dim:
input_dim = feature_dim
if self.method not in ['dot', 'general', 'concat']:
raise ValueError(self.method,
'is not an appropriate attention method.')
elif self.method == 'dot' and input_dim != feature_dim:
raise ValueError(
'dot does not work when input_dim does not equals to feature_dim'
)
if self.method == 'general':
self.attn = torch.nn.Linear(feature_dim, input_dim)
elif self.method == 'concat':
self.attn = torch.nn.Linear(input_dim + feature_dim, feature_dim)
self.v = torch.nn.Parameter(torch.FloatTensor(feature_dim))
def score(self, inputs, targets):
if self.method == 'dot':
return inputs.bmm(targets.transpose(1, 2))
elif self.method == 'general':
energy = self.attn(targets)
return inputs.bmm(energy.transpose(1, 2))
elif self.method == 'concat':
inp_size = inputs.size(1)
tgt_size = targets.size(1)
inputs_exp = inputs.unsqueeze(2).expand(-1, -1, tgt_size, -1)
targets_exp = targets.unsqueeze(1).expand(-1, inp_size, -1, -1)
combined = torch.cat((inputs_exp, targets_exp), 3)
energy = self.attn(combined).tanh()
return torch.sum(self.v * energy, dim=3)
def forward(self, input_0, input_1):
primals_2 = self.attn.weight
primals_3 = self.attn.bias
primals_1 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0], output[1]
| stillarrow/NRT-Lite | Attn | false | 4,389 | [
"MIT"
] | 0 | ba0f091ebfeae19325ce713e11bc426ff63402cd | https://github.com/stillarrow/NRT-Lite/tree/ba0f091ebfeae19325ce713e11bc426ff63402cd | import torch
from torch import nn
class Model(torch.nn.Module):
"""
Attention:
feature_dim: dimension of feature embedding
method: method to calculate attention, (general, dot, concat)
input_dim: dimension of input embedding, default is the same as feature_dim; method dot is only available when input_dim == feature_dim
Inputs:
inputs: batch of inputs; the inp_size is optional; shape=(batch_size, inp_size, feature_dim)
targets: batch of targets to pay attention; shape=(batch_size, tgt_size, feature_dim)
mask: optional target binary mask to avoid paying attention to padding item; shape=(batch_size, tgt_size)
Outputs:
context: context vector computed as the weighted average of all the encoder outputs; inp_size is optional;shape=(batch_size, inp_size, feature_dim)
attntion: attention weight paid to the targets; shape=(batch_size, inp_size, tgt_size)
"""
def __init__(self, feature_dim, method='general', input_dim=None):
super().__init__()
self.method = method
if not input_dim:
input_dim = feature_dim
if self.method not in ['dot', 'general', 'concat']:
raise ValueError(self.method,
'is not an appropriate attention method.')
elif self.method == 'dot' and input_dim != feature_dim:
raise ValueError(
'dot does not work when input_dim does not equals to feature_dim'
)
if self.method == 'general':
self.attn = torch.nn.Linear(feature_dim, input_dim)
elif self.method == 'concat':
self.attn = torch.nn.Linear(input_dim + feature_dim, feature_dim)
self.v = torch.nn.Parameter(torch.FloatTensor(feature_dim))
def score(self, inputs, targets):
if self.method == 'dot':
return inputs.bmm(targets.transpose(1, 2))
elif self.method == 'general':
energy = self.attn(targets)
return inputs.bmm(energy.transpose(1, 2))
elif self.method == 'concat':
inp_size = inputs.size(1)
tgt_size = targets.size(1)
inputs_exp = inputs.unsqueeze(2).expand(-1, -1, tgt_size, -1)
targets_exp = targets.unsqueeze(1).expand(-1, inp_size, -1, -1)
combined = torch.cat((inputs_exp, targets_exp), 3)
energy = self.attn(combined).tanh()
return torch.sum(self.v * energy, dim=3)
def forward(self, inputs, targets, mask=None):
inp_shape = inputs.size()
if len(inp_shape) == 2:
inputs = inputs.view(inp_shape[0], 1, inp_shape[-1])
attn_energies = self.score(inputs, targets)
if mask is not None:
mask = mask.unsqueeze(1).expand(-1, attn_energies.size(1), -1)
attn_energies = attn_energies.masked_fill(~mask, float('-inf'))
attn_weights = nn.functional.softmax(attn_energies, dim=2)
context = attn_weights.bmm(targets)
if len(inp_shape) == 2:
context = context.squeeze(1)
attn_weights = attn_weights.squeeze(1)
return context, attn_weights
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4]
|
TransformerEncoderLayer | # 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_7/inductor_cache/6s/c6sstbvcita246hkfqwdeatnmsh3e6vlcncrzcwlsoqg7dmxvabp.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => add, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_1, [1]), 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=[4],
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_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 = 4
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_7/inductor_cache/zv/czv3tzezwxkylzsgkrivaldxprnr7tvjr5iihe4mbc7bzdev5lsj.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => 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_1, [1]), 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_1, %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_2), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_1, %primals_3), 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=[16],
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 = 16
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_7/inductor_cache/ah/cahpqo3o7hv3q647n5lretlqvfljlubj4ic7gscxws4yvkm5jzff.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# multi_head_attention_forward => mul_2
# Graph fragment:
# %mul_2 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_3, 1.0), 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=[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_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_mul_2(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_7/inductor_cache/7s/c7spagnqvsgjrukyw5jujzjmswxuigeuvpyhxgdob766q2gfvgzr.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_1
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%bmm, [-1], True), kwargs = {})
# %sub_1 : [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_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=[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_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 = 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_7/inductor_cache/dw/cdwqsjnh2osfmjr2utzzaqdg2vrfivzkuhareq3urgidllj2bsvr.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_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=[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_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 = 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_7/inductor_cache/y5/cy5gjrtl7netbzcjhig66pdorub2vbq2qvwmv3tamld2ehimmlz7.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_5 = async_compile.triton('triton_poi_fused_clone_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=[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_5', '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_5(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_7/inductor_cache/ji/cjikooh3unjvssdwbmc5bbgrf7argvwkpdjikzfpajfrzpotlkhf.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_2
# x_3 => var_mean_1
# Graph fragment:
# %add_2 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_1, %squeeze), kwargs = {})
# %var_mean_1 : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add_2, [1]), kwargs = {correction: 0, keepdim: True})
triton_poi_fused_add_native_layer_norm_6 = async_compile.triton('triton_poi_fused_add_native_layer_norm_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: '*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_native_layer_norm_6', '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_6(in_ptr0, in_ptr1, out_ptr0, out_ptr1, 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 + (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_7/inductor_cache/j4/cj4vucbv6vxdldbfg73k3ixw2brnd6f754oxugjq3s7syrcrb4qe.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_2
# x_3 => 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_1, %squeeze), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_8, 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_9), 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_2), kwargs = {})
# %add_4 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %primals_3), kwargs = {})
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: '*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_7', '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_7(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, 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
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_7/inductor_cache/x5/cx5pihzzf3a6eymyu3ib45lia74ugyzdsxkzmca3dxgwtf2oyqhq.py
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# x_4 => relu
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_9), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_8 = async_compile.triton('triton_poi_fused_relu_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: 'i32'}, 'device': 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_8', '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_8(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 % 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)
tl.store(in_out_ptr0 + (x2), tmp4, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/44/c444sh6bryz652bk24ocru63kbqhe67iwwzctt3isl7imfgv5iaa.py
# Topologically Sorted Source Nodes: [x_2, x_8], Original ATen: [aten.add]
# Source node to ATen node mapping:
# x_2 => add_2
# x_8 => add_5
# Graph fragment:
# %add_2 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_1, %squeeze), kwargs = {})
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_11), kwargs = {})
# %add_5 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, %add_tensor), kwargs = {})
triton_poi_fused_add_9 = async_compile.triton('triton_poi_fused_add_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=[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_9', '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_9(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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_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 = 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, ), (1, ))
assert_size_stride(primals_4, (12, 4), (4, 1))
assert_size_stride(primals_5, (12, ), (1, ))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, ), (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((4, 1), (1, 4), torch.float32)
buf1 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_poi_fused_native_layer_norm_0.run(primals_1, buf0, buf1, 4, grid=grid(4), stream=stream0)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_1.run(primals_1, buf0, buf1, primals_2, primals_3, buf2, 16, grid=grid(16), stream=stream0)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf3)
buf4 = 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_5, (4, ), (1, ), 4), buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4), 16), alpha=1, beta=1, out=buf4)
buf5 = 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_5, (4, ), (1, ), 8), buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4), 32), alpha=1, beta=1, out=buf5)
buf6 = reinterpret_tensor(buf3, (4, 4, 1), (1, 4, 16), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
triton_poi_fused_mul_2.run(buf6, primals_5, 16, grid=grid(16), stream=stream0)
del primals_5
buf7 = 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(buf6, reinterpret_tensor(buf4, (4, 1, 4), (1, 1, 4), 0), out=buf7)
buf8 = 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_3.run(buf7, buf8, 64, grid=grid(64), stream=stream0)
buf9 = buf7; del buf7 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten._softmax]
triton_poi_fused__softmax_4.run(buf8, buf9, 64, grid=grid(64), stream=stream0)
buf10 = 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(buf9, reinterpret_tensor(buf5, (4, 4, 1), (1, 4, 1), 0), out=buf10)
buf11 = 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_5.run(buf10, buf11, 4, 4, grid=grid(4, 4), stream=stream0)
buf12 = reinterpret_tensor(buf10, (4, 4), (4, 1), 0); del buf10 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_7, reinterpret_tensor(buf11, (4, 4), (4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf12)
del primals_7
buf13 = buf1; del buf1 # reuse
buf14 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [x_2, x_3], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_6.run(primals_1, buf12, buf13, buf14, 4, grid=grid(4), stream=stream0)
buf15 = empty_strided_cuda((4, 4), (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_7.run(primals_1, buf12, buf13, buf14, primals_2, primals_3, buf15, 16, grid=grid(16), stream=stream0)
del buf13
del buf14
del primals_3
buf16 = reinterpret_tensor(buf8, (4, 16), (16, 1), 0); del buf8 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf15, reinterpret_tensor(primals_8, (4, 16), (1, 4), 0), out=buf16)
buf17 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [x_4], Original ATen: [aten.relu]
triton_poi_fused_relu_8.run(buf17, primals_9, 64, grid=grid(64), stream=stream0)
del primals_9
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf17, reinterpret_tensor(primals_10, (16, 4), (1, 16), 0), out=buf18)
buf19 = buf18; del buf18 # reuse
# Topologically Sorted Source Nodes: [x_2, x_8], Original ATen: [aten.add]
triton_poi_fused_add_9.run(buf19, primals_1, buf12, primals_11, 16, grid=grid(16), stream=stream0)
del primals_11
return (buf19, primals_1, primals_2, buf2, buf9, reinterpret_tensor(buf11, (4, 4), (4, 1), 0), buf12, buf15, buf17, primals_10, primals_8, primals_6, reinterpret_tensor(buf5, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf6, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf4, (4, 4, 1), (1, 4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (4, 1), 32), reinterpret_tensor(primals_4, (4, 4), (4, 1), 16), reinterpret_tensor(primals_4, (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, 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((12, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_5 = rand_strided((12, ), (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((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 torch
from torch import nn
def fill_with_neg_inf(t):
"""FP16-compatible function that fills a tensor with -inf."""
return t.float().fill_(float('-inf')).type_as(t)
def buffered_future_mask(tensor1, tensor2, device):
dim1 = dim2 = tensor1.size()
if tensor2 is not None:
dim2 = tensor2.size()
future_mask = torch.triu(fill_with_neg_inf(torch.ones(dim1, dim2)), 1 +
abs(dim2 - dim1))
future_mask
return future_mask[:dim1, :dim2]
class TransformerEncoderLayer(nn.Module):
def __init__(self, embed_dim, num_heads=4, attn_dropout=0.1,
relu_dropout=0.1, res_dropout=0.1, attn_mask=False):
super().__init__()
self.embed_dim = embed_dim
self.num_heads = num_heads
self.attn_dropout = attn_dropout
self.self_attn = nn.MultiheadAttention(embed_dim=self.embed_dim,
num_heads=self.num_heads, dropout=self.attn_dropout)
self.attn_mask = attn_mask
self.relu_dropout = nn.Dropout(p=relu_dropout)
self.res_dropout = nn.Dropout(p=res_dropout)
self.relu = nn.ReLU()
self.fc1 = nn.Linear(self.embed_dim, 4 * self.embed_dim)
self.fc2 = nn.Linear(4 * self.embed_dim, self.embed_dim)
self.layer_norm = nn.LayerNorm(embed_dim)
def forward(self, x, x_k=None, x_v=None, src_key_padding_mask=None):
residual = x
x = self.layer_norm(x)
mask = buffered_future_mask(x, x_k) if self.attn_mask else None
if x_k is None and x_v is None:
x, _ = self.self_attn(query=x, key=x, value=x, key_padding_mask
=src_key_padding_mask, attn_mask=mask)
else:
x_k = self.layer_norm(x_k)
x_v = self.layer_norm(x_v)
x, _ = self.self_attn(query=x, key=x_k, value=x_v,
key_padding_mask=src_key_padding_mask, attn_mask=mask)
x = self.res_dropout(x)
x = residual + x
residual = x
x = self.layer_norm(x)
x = self.relu(self.fc1(x))
x = self.relu_dropout(x)
x = self.fc2(x)
x = self.res_dropout(x)
x = residual + x
return x
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'embed_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
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_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1,
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 + 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 = 16
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_mul_2(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_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
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_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
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_5(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_native_layer_norm_6(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, 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 + 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_7(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, 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
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_relu_8(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 % 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)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
@triton.jit
def triton_poi_fused_add_9(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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_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) = 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,), (1,))
assert_size_stride(primals_4, (12, 4), (4, 1))
assert_size_stride(primals_5, (12,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (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((4, 1), (1, 4), torch.float32)
buf1 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(4)](primals_1, buf0, buf1,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_native_layer_norm_1[grid(16)](primals_1, buf0,
buf1, primals_2, primals_3, buf2, 16, XBLOCK=16, num_warps=1,
num_stages=1)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4
), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(reinterpret_tensor(primals_5, (4,), (1,), 4),
buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4), 16), alpha=
1, beta=1, out=buf4)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(reinterpret_tensor(primals_5, (4,), (1,), 8),
buf2, reinterpret_tensor(primals_4, (4, 4), (1, 4), 32), alpha=
1, beta=1, out=buf5)
buf6 = reinterpret_tensor(buf3, (4, 4, 1), (1, 4, 16), 0)
del buf3
triton_poi_fused_mul_2[grid(16)](buf6, primals_5, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_5
buf7 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf6, reinterpret_tensor(buf4, (4, 1, 4), (1, 1,
4), 0), out=buf7)
buf8 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_3[grid(64)](buf7, buf8, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf9 = buf7
del buf7
triton_poi_fused__softmax_4[grid(64)](buf8, buf9, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf10 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
extern_kernels.bmm(buf9, reinterpret_tensor(buf5, (4, 4, 1), (1, 4,
1), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
triton_poi_fused_clone_5[grid(4, 4)](buf10, buf11, 4, 4, XBLOCK=4,
YBLOCK=4, num_warps=1, num_stages=1)
buf12 = reinterpret_tensor(buf10, (4, 4), (4, 1), 0)
del buf10
extern_kernels.addmm(primals_7, reinterpret_tensor(buf11, (4, 4), (
4, 1), 0), reinterpret_tensor(primals_6, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf12)
del primals_7
buf13 = buf1
del buf1
buf14 = buf0
del buf0
triton_poi_fused_add_native_layer_norm_6[grid(4)](primals_1, buf12,
buf13, buf14, 4, XBLOCK=4, num_warps=1, num_stages=1)
buf15 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_add_native_layer_norm_7[grid(16)](primals_1, buf12,
buf13, buf14, primals_2, primals_3, buf15, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf13
del buf14
del primals_3
buf16 = reinterpret_tensor(buf8, (4, 16), (16, 1), 0)
del buf8
extern_kernels.mm(buf15, reinterpret_tensor(primals_8, (4, 16), (1,
4), 0), out=buf16)
buf17 = buf16
del buf16
triton_poi_fused_relu_8[grid(64)](buf17, primals_9, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_9
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf17, reinterpret_tensor(primals_10, (16, 4), (1,
16), 0), out=buf18)
buf19 = buf18
del buf18
triton_poi_fused_add_9[grid(16)](buf19, primals_1, buf12,
primals_11, 16, XBLOCK=16, num_warps=1, num_stages=1)
del primals_11
return (buf19, primals_1, primals_2, buf2, buf9, reinterpret_tensor(
buf11, (4, 4), (4, 1), 0), buf12, buf15, buf17, primals_10,
primals_8, primals_6, reinterpret_tensor(buf5, (4, 1, 4), (1, 1, 4),
0), reinterpret_tensor(buf6, (4, 1, 4), (1, 1, 4), 0),
reinterpret_tensor(buf4, (4, 4, 1), (1, 4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (4, 1), 32),
reinterpret_tensor(primals_4, (4, 4), (4, 1), 16),
reinterpret_tensor(primals_4, (4, 4), (4, 1), 0))
def fill_with_neg_inf(t):
"""FP16-compatible function that fills a tensor with -inf."""
return t.float().fill_(float('-inf')).type_as(t)
def buffered_future_mask(tensor1, tensor2, device):
dim1 = dim2 = tensor1.size()
if tensor2 is not None:
dim2 = tensor2.size()
future_mask = torch.triu(fill_with_neg_inf(torch.ones(dim1, dim2)), 1 +
abs(dim2 - dim1))
future_mask
return future_mask[:dim1, :dim2]
class TransformerEncoderLayerNew(nn.Module):
def __init__(self, embed_dim, num_heads=4, attn_dropout=0.1,
relu_dropout=0.1, res_dropout=0.1, attn_mask=False):
super().__init__()
self.embed_dim = embed_dim
self.num_heads = num_heads
self.attn_dropout = attn_dropout
self.self_attn = nn.MultiheadAttention(embed_dim=self.embed_dim,
num_heads=self.num_heads, dropout=self.attn_dropout)
self.attn_mask = attn_mask
self.relu_dropout = nn.Dropout(p=relu_dropout)
self.res_dropout = nn.Dropout(p=res_dropout)
self.relu = nn.ReLU()
self.fc1 = nn.Linear(self.embed_dim, 4 * self.embed_dim)
self.fc2 = nn.Linear(4 * self.embed_dim, self.embed_dim)
self.layer_norm = nn.LayerNorm(embed_dim)
def forward(self, input_0):
primals_4 = self.self_attn.in_proj_weight
primals_5 = self.self_attn.in_proj_bias
primals_1 = self.self_attn.out_proj.weight
primals_2 = self.self_attn.out_proj.bias
primals_8 = self.fc1.weight
primals_9 = self.fc1.bias
primals_10 = self.fc2.weight
primals_3 = self.fc2.bias
primals_7 = self.layer_norm.weight
primals_11 = self.layer_norm.bias
primals_6 = 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])
return output[0]
| sreekanth-sreekumar/daiz-woz-nlp-project | TransformerEncoderLayer | false | 4,390 | [
"MIT"
] | 0 | 9971f752aee6a850e265f15e97a3a1ef2dacd323 | https://github.com/sreekanth-sreekumar/daiz-woz-nlp-project/tree/9971f752aee6a850e265f15e97a3a1ef2dacd323 | import torch
from torch import nn
def fill_with_neg_inf(t):
"""FP16-compatible function that fills a tensor with -inf."""
return t.float().fill_(float('-inf')).type_as(t)
def buffered_future_mask(tensor1, tensor2, device):
dim1 = dim2 = tensor1.size()
if tensor2 is not None:
dim2 = tensor2.size()
future_mask = torch.triu(fill_with_neg_inf(torch.ones(dim1, dim2)), 1 +
abs(dim2 - dim1))
future_mask
return future_mask[:dim1, :dim2]
class Model(nn.Module):
def __init__(self, embed_dim, num_heads=4, attn_dropout=0.1,
relu_dropout=0.1, res_dropout=0.1, attn_mask=False):
super().__init__()
self.embed_dim = embed_dim
self.num_heads = num_heads
self.attn_dropout = attn_dropout
self.self_attn = nn.MultiheadAttention(embed_dim=self.embed_dim,
num_heads=self.num_heads, dropout=self.attn_dropout)
self.attn_mask = attn_mask
self.relu_dropout = nn.Dropout(p=relu_dropout)
self.res_dropout = nn.Dropout(p=res_dropout)
self.relu = nn.ReLU()
self.fc1 = nn.Linear(self.embed_dim, 4 * self.embed_dim)
self.fc2 = nn.Linear(4 * self.embed_dim, self.embed_dim)
self.layer_norm = nn.LayerNorm(embed_dim)
def forward(self, x, x_k=None, x_v=None, src_key_padding_mask=None):
residual = x
x = self.layer_norm(x)
mask = buffered_future_mask(x, x_k) if self.attn_mask else None
if x_k is None and x_v is None:
x, _ = self.self_attn(query=x, key=x, value=x, key_padding_mask
=src_key_padding_mask, attn_mask=mask)
else:
x_k = self.layer_norm(x_k)
x_v = self.layer_norm(x_v)
x, _ = self.self_attn(query=x, key=x_k, value=x_v,
key_padding_mask=src_key_padding_mask, attn_mask=mask)
x = self.res_dropout(x)
x = residual + x
residual = x
x = self.layer_norm(x)
x = self.relu(self.fc1(x))
x = self.relu_dropout(x)
x = self.fc2(x)
x = self.res_dropout(x)
x = residual + x
return x
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [4]
|
IdentityPadding | # 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_7/inductor_cache/i5/ci5jtsingp2qnrodx6wlbyak3tjanrom5rpybyruspgrcwq6flhn.py
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.max_pool2d_with_indices]
# Source node to ATen node mapping:
# out_1 => 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': 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_max_pool2d_with_indices_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):
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: [out_1], Original ATen: [aten.max_pool2d_with_indices]
stream0 = get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_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
import torch.nn.functional as F
class IdentityPadding(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super(IdentityPadding, self).__init__()
self.identity = nn.MaxPool2d(1, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, x):
out = F.pad(x, (0, 0, 0, 0, 0, self.num_zeros))
out = self.identity(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_filters': 4, 'channels_in': 4, 'stride': 1}]
| 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_poi_fused_max_pool2d_with_indices_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):
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_max_pool2d_with_indices_0[grid(256)](arg0_1, buf0,
256, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class IdentityPaddingNew(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super(IdentityPaddingNew, self).__init__()
self.identity = nn.MaxPool2d(1, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| sunqcc/Pytorch-HW-CIFAR10 | IdentityPadding | false | 4,391 | [
"MIT"
] | 0 | 33a55a5a832474083820b65c46f809ac98f8b109 | https://github.com/sunqcc/Pytorch-HW-CIFAR10/tree/33a55a5a832474083820b65c46f809ac98f8b109 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super().__init__()
self.identity = nn.MaxPool2d(1, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, x):
out = F.pad(x, (0, 0, 0, 0, 0, self.num_zeros))
out = self.identity(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 1]
|
SoftCrossEntropyLoss2d | # 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_7/inductor_cache/td/ctdj5kazgiki6gdaadhqtp2x7tq2ee5ey5hqqdcoqmp54jyhf74f.py
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
# Source node to ATen node mapping:
# log_softmax => amax, sub
# Graph fragment:
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%arg0_1, [1], True), kwargs = {})
# %sub : [num_users=2] = call_function[target=torch.ops.aten.sub.Tensor](args = (%arg0_1, %amax), kwargs = {})
triton_poi_fused__log_softmax_0 = async_compile.triton('triton_poi_fused__log_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__log_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__log_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
tl.store(out_ptr0 + (x3), tmp8, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/ne/cneuoe5ed43ex5ojv524lcm6efihmzp4tx5sn3qedrcityjvt6pz.py
# Topologically Sorted Source Nodes: [log_softmax, inputs], Original ATen: [aten._log_softmax, aten.neg]
# Source node to ATen node mapping:
# inputs => neg
# 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 = {})
# %neg : [num_users=4] = call_function[target=torch.ops.aten.neg.default](args = (%sub_1,), kwargs = {})
triton_poi_fused__log_softmax_neg_1 = async_compile.triton('triton_poi_fused__log_softmax_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=[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__log_softmax_neg_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__log_softmax_neg_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)
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 = 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
tmp14 = -tmp13
tl.store(out_ptr0 + (x3), tmp14, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/hm/chmjikblxn7fogqton2wplz4p5ubizfnslw2y2jskup5crqiglsi.py
# Topologically Sorted Source Nodes: [getitem], Original ATen: [aten.index]
# Source node to ATen node mapping:
# getitem => index
# Graph fragment:
# %index : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%neg, [%full_default]), kwargs = {})
triton_poi_fused_index_2 = async_compile.triton('triton_poi_fused_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=[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_index_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_index_2(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 + (x1 + (16*y0)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (4*x1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/xg/cxgyo7qw6gisgut6u4bocabj7jkiaw3bydso4hzgts7qjaia74xe.py
# Topologically Sorted Source Nodes: [getitem_2], Original ATen: [aten.index]
# Source node to ATen node mapping:
# getitem_2 => index_2
# Graph fragment:
# %index_2 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%neg, [%full_default_2]), kwargs = {})
triton_poi_fused_index_3 = async_compile.triton('triton_poi_fused_index_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, 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_index_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_index_3(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 + (64 + x1 + (16*y0)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (4*x1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/iq/ciq2g5midlhlq7225k24quvfyfki7oejp5vufoszhy2i3ktzwl66.py
# Topologically Sorted Source Nodes: [getitem_4], Original ATen: [aten.index]
# Source node to ATen node mapping:
# getitem_4 => index_4
# Graph fragment:
# %index_4 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%neg, [%full_default_4]), kwargs = {})
triton_poi_fused_index_4 = async_compile.triton('triton_poi_fused_index_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=[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_index_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_index_4(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 + (128 + x1 + (16*y0)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (4*x1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/uw/cuwcbkvcc6zyei2rc3jzebdq4rl7s3nyyw3nqvslxd7f2o7txngx.py
# Topologically Sorted Source Nodes: [getitem_6], Original ATen: [aten.index]
# Source node to ATen node mapping:
# getitem_6 => index_6
# Graph fragment:
# %index_6 : [num_users=1] = call_function[target=torch.ops.aten.index.Tensor](args = (%neg, [%full_default_6]), kwargs = {})
triton_poi_fused_index_5 = async_compile.triton('triton_poi_fused_index_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=[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_index_5', '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_index_5(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 + (192 + x1 + (16*y0)), xmask & ymask, eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + (4*x1)), tmp0, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/zz/czzlytqgjwzprkdc572dfyzk5vpiocqqctt74eadeo2aq54ezkih.py
# Topologically Sorted Source Nodes: [truediv, loss, truediv_1, loss_1, truediv_2, loss_2, truediv_3, loss_3], Original ATen: [aten.div, aten.add]
# Source node to ATen node mapping:
# loss => add
# loss_1 => add_1
# loss_2 => add_2
# loss_3 => add_3
# truediv => div
# truediv_1 => div_1
# truediv_2 => div_2
# truediv_3 => div_3
# Graph fragment:
# %div : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%convolution, 16), kwargs = {})
# %add : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%div, 0), kwargs = {})
# %div_1 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%convolution_1, 16), kwargs = {})
# %add_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add, %div_1), kwargs = {})
# %div_2 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%convolution_2, 16), kwargs = {})
# %add_2 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_1, %div_2), kwargs = {})
# %div_3 : [num_users=1] = call_function[target=torch.ops.aten.div.Tensor](args = (%convolution_3, 16), kwargs = {})
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_2, %div_3), kwargs = {})
triton_poi_fused_add_div_6 = async_compile.triton('triton_poi_fused_add_div_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=[1],
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': {4: 1}, 'configs': [AttrsDescriptor(divisible_by_16=(0, 1, 2, 3), equal_to_1=(4,))]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_div_6', '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_div_6(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, 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_ptr0 + (0))
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp6 = tl.load(in_ptr1 + (0))
tmp7 = tl.broadcast_to(tmp6, [XBLOCK])
tmp10 = tl.load(in_out_ptr0 + (0))
tmp11 = tl.broadcast_to(tmp10, [XBLOCK])
tmp14 = tl.load(in_ptr2 + (0))
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp2 = 0.0625
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = tmp3 + tmp4
tmp8 = tmp7 * tmp2
tmp9 = tmp5 + tmp8
tmp12 = tmp11 * tmp2
tmp13 = tmp9 + tmp12
tmp16 = tmp15 * tmp2
tmp17 = tmp13 + tmp16
tl.store(in_out_ptr0 + (tl.full([XBLOCK], 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((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [log_softmax], Original ATen: [aten._log_softmax]
stream0 = get_raw_stream(0)
triton_poi_fused__log_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: [log_softmax, inputs], Original ATen: [aten._log_softmax, aten.neg]
triton_poi_fused__log_softmax_neg_1.run(buf0, buf1, 256, grid=grid(256), stream=stream0)
del buf0
buf2 = empty_strided_cuda((1, 4, 4, 4), (64, 1, 16, 4), torch.float32)
# Topologically Sorted Source Nodes: [getitem], Original ATen: [aten.index]
triton_poi_fused_index_2.run(buf1, buf2, 4, 16, grid=grid(4, 16), stream=stream0)
buf3 = empty_strided_cuda((1, 4, 4, 4), (64, 1, 16, 4), torch.float32)
# Topologically Sorted Source Nodes: [getitem_1], Original ATen: [aten.index]
triton_poi_fused_index_2.run(arg1_1, buf3, 4, 16, grid=grid(4, 16), stream=stream0)
# Topologically Sorted Source Nodes: [getitem, getitem_1, conv2d], Original ATen: [aten.index, aten.convolution]
buf4 = extern_kernels.convolution(buf2, buf3, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (1, 1, 1, 1), (1, 1, 1, 1))
buf5 = buf3; del buf3 # reuse
# Topologically Sorted Source Nodes: [getitem_2], Original ATen: [aten.index]
triton_poi_fused_index_3.run(buf1, buf5, 4, 16, grid=grid(4, 16), stream=stream0)
buf6 = buf2; del buf2 # reuse
# Topologically Sorted Source Nodes: [getitem_3], Original ATen: [aten.index]
triton_poi_fused_index_3.run(arg1_1, buf6, 4, 16, grid=grid(4, 16), stream=stream0)
# Topologically Sorted Source Nodes: [getitem_2, getitem_3, conv2d_1], Original ATen: [aten.index, aten.convolution]
buf7 = extern_kernels.convolution(buf5, buf6, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (1, 1, 1, 1), (1, 1, 1, 1))
buf8 = buf6; del buf6 # reuse
# Topologically Sorted Source Nodes: [getitem_4], Original ATen: [aten.index]
triton_poi_fused_index_4.run(buf1, buf8, 4, 16, grid=grid(4, 16), stream=stream0)
buf9 = buf5; del buf5 # reuse
# Topologically Sorted Source Nodes: [getitem_5], Original ATen: [aten.index]
triton_poi_fused_index_4.run(arg1_1, buf9, 4, 16, grid=grid(4, 16), stream=stream0)
# Topologically Sorted Source Nodes: [getitem_4, getitem_5, conv2d_2], Original ATen: [aten.index, aten.convolution]
buf10 = extern_kernels.convolution(buf8, buf9, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (1, 1, 1, 1), (1, 1, 1, 1))
buf11 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [getitem_6], Original ATen: [aten.index]
triton_poi_fused_index_5.run(buf1, buf11, 4, 16, grid=grid(4, 16), stream=stream0)
del buf1
buf12 = buf8; del buf8 # reuse
# Topologically Sorted Source Nodes: [getitem_7], Original ATen: [aten.index]
triton_poi_fused_index_5.run(arg1_1, buf12, 4, 16, grid=grid(4, 16), stream=stream0)
del arg1_1
# Topologically Sorted Source Nodes: [getitem_6, getitem_7, conv2d_3], Original ATen: [aten.index, aten.convolution]
buf13 = extern_kernels.convolution(buf11, buf12, stride=(1, 1), padding=(0, 0), dilation=(1, 1), transposed=False, output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf13, (1, 1, 1, 1), (1, 1, 1, 1))
del buf11
del buf12
buf14 = buf10; del buf10 # reuse
# Topologically Sorted Source Nodes: [truediv, loss, truediv_1, loss_1, truediv_2, loss_2, truediv_3, loss_3], Original ATen: [aten.div, aten.add]
triton_poi_fused_add_div_6.run(buf14, buf4, buf7, buf13, 1, grid=grid(1), stream=stream0)
del buf13
del buf4
del buf7
return (buf14, )
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
from torch import nn
class SoftCrossEntropyLoss2d(nn.Module):
def forward(self, inputs, targets):
loss = 0
inputs = -F.log_softmax(inputs, dim=1)
for index in range(inputs.size()[0]):
loss += F.conv2d(inputs[range(index, index + 1)], targets[range
(index, index + 1)]) / (targets.size()[2] * targets.size()[3])
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
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
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__log_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
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_softmax_neg_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)
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 = 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
tmp14 = -tmp13
tl.store(out_ptr0 + x3, tmp14, xmask)
@triton.jit
def triton_poi_fused_index_2(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 + (x1 + 16 * y0), xmask & ymask, eviction_policy
='evict_last')
tl.store(out_ptr0 + (y0 + 4 * x1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_index_3(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 + (64 + x1 + 16 * y0), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + 4 * x1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_index_4(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 + (128 + x1 + 16 * y0), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + 4 * x1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_index_5(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 + (192 + x1 + 16 * y0), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (y0 + 4 * x1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_div_6(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2,
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_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp6 = tl.load(in_ptr1 + 0)
tmp7 = tl.broadcast_to(tmp6, [XBLOCK])
tmp10 = tl.load(in_out_ptr0 + 0)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK])
tmp14 = tl.load(in_ptr2 + 0)
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp2 = 0.0625
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = tmp3 + tmp4
tmp8 = tmp7 * tmp2
tmp9 = tmp5 + tmp8
tmp12 = tmp11 * tmp2
tmp13 = tmp9 + tmp12
tmp16 = tmp15 * tmp2
tmp17 = tmp13 + tmp16
tl.store(in_out_ptr0 + tl.full([XBLOCK], 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((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_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__log_softmax_neg_1[grid(256)](buf0, buf1, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del buf0
buf2 = empty_strided_cuda((1, 4, 4, 4), (64, 1, 16, 4), torch.float32)
triton_poi_fused_index_2[grid(4, 16)](buf1, buf2, 4, 16, XBLOCK=16,
YBLOCK=4, num_warps=1, num_stages=1)
buf3 = empty_strided_cuda((1, 4, 4, 4), (64, 1, 16, 4), torch.float32)
triton_poi_fused_index_2[grid(4, 16)](arg1_1, buf3, 4, 16, XBLOCK=
16, YBLOCK=4, num_warps=1, num_stages=1)
buf4 = extern_kernels.convolution(buf2, buf3, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (1, 1, 1, 1), (1, 1, 1, 1))
buf5 = buf3
del buf3
triton_poi_fused_index_3[grid(4, 16)](buf1, buf5, 4, 16, XBLOCK=16,
YBLOCK=4, num_warps=1, num_stages=1)
buf6 = buf2
del buf2
triton_poi_fused_index_3[grid(4, 16)](arg1_1, buf6, 4, 16, XBLOCK=
16, YBLOCK=4, num_warps=1, num_stages=1)
buf7 = extern_kernels.convolution(buf5, buf6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (1, 1, 1, 1), (1, 1, 1, 1))
buf8 = buf6
del buf6
triton_poi_fused_index_4[grid(4, 16)](buf1, buf8, 4, 16, XBLOCK=16,
YBLOCK=4, num_warps=1, num_stages=1)
buf9 = buf5
del buf5
triton_poi_fused_index_4[grid(4, 16)](arg1_1, buf9, 4, 16, XBLOCK=
16, YBLOCK=4, num_warps=1, num_stages=1)
buf10 = extern_kernels.convolution(buf8, buf9, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (1, 1, 1, 1), (1, 1, 1, 1))
buf11 = buf9
del buf9
triton_poi_fused_index_5[grid(4, 16)](buf1, buf11, 4, 16, XBLOCK=16,
YBLOCK=4, num_warps=1, num_stages=1)
del buf1
buf12 = buf8
del buf8
triton_poi_fused_index_5[grid(4, 16)](arg1_1, buf12, 4, 16, XBLOCK=
16, YBLOCK=4, num_warps=1, num_stages=1)
del arg1_1
buf13 = extern_kernels.convolution(buf11, buf12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf13, (1, 1, 1, 1), (1, 1, 1, 1))
del buf11
del buf12
buf14 = buf10
del buf10
triton_poi_fused_add_div_6[grid(1)](buf14, buf4, buf7, buf13, 1,
XBLOCK=1, num_warps=1, num_stages=1)
del buf13
del buf4
del buf7
return buf14,
class SoftCrossEntropyLoss2dNew(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]
| sudohainguyen/GLNet-pytorch | SoftCrossEntropyLoss2d | false | 4,392 | [
"Apache-2.0"
] | 0 | 91454831fac6e27f894d55d320dd3bcec946ac0f | https://github.com/sudohainguyen/GLNet-pytorch/tree/91454831fac6e27f894d55d320dd3bcec946ac0f | import torch
import torch.nn.functional as F
from torch import nn
class Model(nn.Module):
def forward(self, inputs, targets):
loss = 0
inputs = -F.log_softmax(inputs, dim=1)
for index in range(inputs.size()[0]):
loss += F.conv2d(inputs[range(index, index + 1)], targets[range
(index, index + 1)]) / (targets.size()[2] * targets.size()[3])
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return []
|
TransformerDecoderLayer | # 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_7/inductor_cache/6s/c6sstbvcita246hkfqwdeatnmsh3e6vlcncrzcwlsoqg7dmxvabp.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => add, rsqrt, var_mean
# Graph fragment:
# %var_mean : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%primals_3, [1]), 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=[4],
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_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 = 4
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_7/inductor_cache/zv/czv3tzezwxkylzsgkrivaldxprnr7tvjr5iihe4mbc7bzdev5lsj.py
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
# Source node to ATen node mapping:
# x => 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, [1]), 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=[16],
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 = 16
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_7/inductor_cache/nh/cnhzogcsc23357d6mz66i2tbiakefor7jmjbuknlxd2pukwzsqvy.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_4
# Graph fragment:
# %add_4 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_5, %getitem_7), 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=[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_add_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_add_2(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 + (8 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/b2/cb27uzzqmelnhn5rizrrnrquph7emgjshbdfed4tkdvm4cgmlb2b.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
# Source node to ATen node mapping:
# multi_head_attention_forward => mul_2
# Graph fragment:
# %mul_2 : [num_users=2] = call_function[target=torch.ops.aten.mul.Tensor](args = (%permute_4, 1.0), 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=[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_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_mul_3(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_7/inductor_cache/ty/ctyn3cgabgvf3tfboxeqvqs2polv3ds4sd4xq43nkhpfgnumz36a.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_3
# Graph fragment:
# %add_3 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%view_3, %getitem_6), 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: 'i32'}, 'device': 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_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_add_4(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 + (4 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + (x2), tmp2, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/cj/ccjtb53vi2enhyltbi75pqjo7acm2c3wbd3vecte5jgk4346tlkg.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.baddbmm, aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_5, amax, exp, sub_1, sum_1
# Graph fragment:
# %add_5 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%unsqueeze_3, %bmm), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_5, [-1], True), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_5, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_1,), kwargs = {})
# %sum_1 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp, [-1], True), kwargs = {})
triton_poi_fused__softmax_baddbmm_5 = async_compile.triton('triton_poi_fused__softmax_baddbmm_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=[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__softmax_baddbmm_5', '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_baddbmm_5(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 % 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (4*x2), 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*x2)), 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*x2)), 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*x2)), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tl.store(out_ptr0 + (x2), tmp14, xmask)
tl.store(out_ptr1 + (x2), tmp25, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/cx/ccxtljuw2xcyrbo62aj6yebpzwarnqixht7ivpkvanfonpxoxtpm.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.baddbmm, aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward => add_5, amax, div, exp, sub_1
# Graph fragment:
# %add_5 : [num_users=2] = call_function[target=torch.ops.aten.add.Tensor](args = (%unsqueeze_3, %bmm), kwargs = {})
# %amax : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%add_5, [-1], True), kwargs = {})
# %sub_1 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_5, %amax), kwargs = {})
# %exp : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_1,), kwargs = {})
# %div : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp, %sum_1), kwargs = {})
triton_poi_fused__softmax_baddbmm_6 = async_compile.triton('triton_poi_fused__softmax_baddbmm_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=[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__softmax_baddbmm_6', '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__softmax_baddbmm_6(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
x3 = xindex % 16
x4 = xindex
x5 = (xindex // 4)
tmp0 = tl.load(in_ptr0 + (x3), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_out_ptr0 + (x4), xmask)
tmp3 = tl.load(in_ptr1 + (x5), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr2 + (x5), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp5 = tl_math.exp(tmp4)
tmp7 = tmp5 / tmp6
tl.store(in_out_ptr0 + (x4), tmp7, xmask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/x4/cx4e42dqckymblkx3gacocdjlcbrlkvigu7oawnafti3mshwlw7z.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_8,), kwargs = {memory_format: torch.contiguous_format})
triton_poi_fused_clone_7 = async_compile.triton('triton_poi_fused_clone_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=[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_7', '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_7(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_7/inductor_cache/ig/cigozvvqtyrz4ol6g7z5mt7uuxfxudhygi33iembcjwnz37tpgxk.py
# Topologically Sorted Source Nodes: [input_1, x_3], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# input_1 => add_6
# x_3 => var_mean_1
# Graph fragment:
# %add_6 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %permute_10), kwargs = {})
# %var_mean_1 : [num_users=2] = call_function[target=torch.ops.aten.var_mean.correction](args = (%add_6, [1]), kwargs = {correction: 0, keepdim: True})
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=[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_native_layer_norm_8', '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_8(in_ptr0, in_ptr1, out_ptr0, out_ptr1, 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 + (4*x0), xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x0), xmask)
tmp3 = tl.load(in_ptr0 + (1 + (4*x0)), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (4 + x0), xmask)
tmp7 = tl.load(in_ptr0 + (2 + (4*x0)), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (8 + x0), xmask)
tmp11 = tl.load(in_ptr0 + (3 + (4*x0)), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (12 + x0), xmask)
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_7/inductor_cache/6w/c6wmk2qrhn7x46yaqhwiq3cu7hyilbqe6zekmvx5m2am2noka73i.py
# Topologically Sorted Source Nodes: [input_1, x_3], Original ATen: [aten.add, aten.native_layer_norm]
# Source node to ATen node mapping:
# input_1 => add_6
# x_3 => add_7, add_8, mul_3, mul_4, rsqrt_1, sub_2
# Graph fragment:
# %add_6 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %permute_10), kwargs = {})
# %add_7 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%getitem_8, 1e-05), kwargs = {})
# %rsqrt_1 : [num_users=1] = call_function[target=torch.ops.aten.rsqrt.default](args = (%add_7,), kwargs = {})
# %sub_2 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%add_6, %getitem_9), 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_9), kwargs = {})
# %add_8 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mul_4, %primals_10), kwargs = {})
triton_poi_fused_add_native_layer_norm_9 = async_compile.triton('triton_poi_fused_add_native_layer_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.pointwise(
size_hints=[4, 4], tile_hint=TileHint.DEFAULT,
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_poi_fused_add_native_layer_norm_9', '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_9(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, 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 + (x1 + (4*y0)), xmask & ymask)
tmp1 = tl.load(in_ptr1 + (y0 + (4*x1)), xmask & ymask)
tmp3 = tl.load(in_ptr2 + (y0), ymask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + (y0), ymask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + (x1), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + (x1), 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 + (x1 + (4*y0)), tmp13, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/cq/ccqjwtivgffj3fdrhqc5qea4z3k5ktjfmaa7nlhpe6umeiobmnoz.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward_1 => amax_1, exp_1, sub_3
# Graph fragment:
# %amax_1 : [num_users=1] = call_function[target=torch.ops.aten.amax.default](args = (%bmm_2, [-1], True), kwargs = {})
# %sub_3 : [num_users=1] = call_function[target=torch.ops.aten.sub.Tensor](args = (%bmm_2, %amax_1), kwargs = {})
# %exp_1 : [num_users=2] = call_function[target=torch.ops.aten.exp.default](args = (%sub_3,), kwargs = {})
triton_poi_fused__softmax_10 = async_compile.triton('triton_poi_fused__softmax_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: 'i32'}, 'device': 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_10', '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_10(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_7/inductor_cache/bb/cbb2v4rxyqxrj7p636aeq2f4a47hddydr723lathv4zacx4aighi.py
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten._softmax]
# Source node to ATen node mapping:
# multi_head_attention_forward_1 => div_1, sum_2
# Graph fragment:
# %sum_2 : [num_users=1] = call_function[target=torch.ops.aten.sum.dim_IntList](args = (%exp_1, [-1], True), kwargs = {})
# %div_1 : [num_users=2] = call_function[target=torch.ops.aten.div.Tensor](args = (%exp_1, %sum_2), kwargs = {})
triton_poi_fused__softmax_11 = async_compile.triton('triton_poi_fused__softmax_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=[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_11', '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_11(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_7/inductor_cache/ue/cue4vol3zjpfonntu7u2y53xk2aejnv6m5m7hnffiuqf7b7ak6xo.py
# Topologically Sorted Source Nodes: [input_1, input_2], Original ATen: [aten.add]
# Source node to ATen node mapping:
# input_1 => add_6
# input_2 => add_12
# Graph fragment:
# %add_6 : [num_users=3] = call_function[target=torch.ops.aten.add.Tensor](args = (%primals_3, %permute_10), kwargs = {})
# %add_12 : [num_users=4] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_6, %permute_23), kwargs = {})
triton_poi_fused_add_12 = async_compile.triton('triton_poi_fused_add_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=[4, 4], tile_hint=TileHint.DEFAULT,
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), equal_to_1=())]},
inductor_meta={'autotune_hints': set(), 'kernel_name': 'triton_poi_fused_add_12', '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_add_12(in_ptr0, in_ptr1, in_ptr2, in_ptr3, 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 + (x1 + (4*y0)), xmask & ymask)
tmp1 = tl.load(in_ptr1 + (y0 + (4*x1)), xmask & ymask)
tmp3 = tl.load(in_ptr2 + (y0 + (4*x1)), xmask & ymask)
tmp4 = tl.load(in_ptr3 + (y0), ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tl.store(out_ptr0 + (x1 + (4*y0)), tmp6, xmask & ymask)
''', device_str='cuda')
# kernel path: runs/run_shard_7/inductor_cache/rh/crhjgkmrrze54eudfkj5yh6iiuvh4czuvpabouzuvslbmeubyjqd.py
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
# Source node to ATen node mapping:
# relu => relu
# Graph fragment:
# %add_tensor_1 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default_1, %primals_19), kwargs = {})
# %relu : [num_users=2] = call_function[target=torch.ops.aten.relu.default](args = (%add_tensor_1,), kwargs = {})
triton_poi_fused_relu_13 = async_compile.triton('triton_poi_fused_relu_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=[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_13', '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_13(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')
# kernel path: runs/run_shard_7/inductor_cache/dh/cdholtmgflhsxrlmxtz3uiixsj6agz2bd7ktwu5hfoomdtouzfef.py
# Topologically Sorted Source Nodes: [input_3], Original ATen: [aten.add]
# Source node to ATen node mapping:
# input_3 => add_15
# Graph fragment:
# %add_tensor : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%mm_default, %primals_21), kwargs = {})
# %add_15 : [num_users=1] = call_function[target=torch.ops.aten.add.Tensor](args = (%add_12, %add_tensor), kwargs = {})
triton_poi_fused_add_14 = async_compile.triton('triton_poi_fused_add_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: 'i32'}, 'device': 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_14', '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_14(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, 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, (4, ), (1, ))
assert_size_stride(primals_2, (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, (12, 4), (4, 1))
assert_size_stride(primals_6, (12, ), (1, ))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, ), (1, ))
assert_size_stride(primals_9, (4, ), (1, ))
assert_size_stride(primals_10, (4, ), (1, ))
assert_size_stride(primals_11, (4, 4), (4, 1))
assert_size_stride(primals_12, (12, 4), (4, 1))
assert_size_stride(primals_13, (12, ), (1, ))
assert_size_stride(primals_14, (4, 4), (4, 1))
assert_size_stride(primals_15, (4, ), (1, ))
assert_size_stride(primals_16, (4, ), (1, ))
assert_size_stride(primals_17, (4, ), (1, ))
assert_size_stride(primals_18, (4, 4), (4, 1))
assert_size_stride(primals_19, (4, ), (1, ))
assert_size_stride(primals_20, (4, 4), (4, 1))
assert_size_stride(primals_21, (4, ), (1, ))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
buf1 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
stream0 = get_raw_stream(0)
triton_poi_fused_native_layer_norm_0.run(primals_3, buf0, buf1, 4, grid=grid(4), stream=stream0)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [x], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_1.run(primals_3, buf0, buf1, primals_1, primals_2, buf2, 16, grid=grid(16), stream=stream0)
del primals_1
del primals_2
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 16), out=buf4)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 32), out=buf5)
buf6 = reinterpret_tensor(buf5, (4, 1, 4), (4, 4, 1), 0); del buf5 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
triton_poi_fused_add_2.run(buf6, primals_6, 16, grid=grid(16), stream=stream0)
buf7 = reinterpret_tensor(buf3, (4, 4, 1), (1, 4, 16), 0); del buf3 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.mul]
triton_poi_fused_mul_3.run(buf7, primals_6, 16, grid=grid(16), stream=stream0)
buf8 = reinterpret_tensor(buf4, (4, 1, 4), (4, 4, 1), 0); del buf4 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.add]
triton_poi_fused_add_4.run(buf8, primals_6, 16, grid=grid(16), stream=stream0)
del primals_6
buf9 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.baddbmm]
extern_kernels.bmm(buf7, reinterpret_tensor(buf8, (4, 1, 4), (1, 0, 4), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf11 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.baddbmm, aten._softmax]
triton_poi_fused__softmax_baddbmm_5.run(primals_4, buf9, buf10, buf11, 16, grid=grid(16), stream=stream0)
buf12 = buf9; del buf9 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.baddbmm, aten._softmax]
triton_poi_fused__softmax_baddbmm_6.run(buf12, primals_4, buf10, buf11, 64, grid=grid(64), stream=stream0)
del primals_4
buf13 = reinterpret_tensor(buf11, (4, 4, 1), (4, 1, 1), 0); del buf11 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.bmm]
extern_kernels.bmm(buf12, reinterpret_tensor(buf6, (4, 4, 1), (1, 4, 0), 0), out=buf13)
buf14 = reinterpret_tensor(buf10, (4, 4, 1), (4, 1, 1), 0); del buf10 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(buf13, buf14, 4, 4, grid=grid(4, 4), stream=stream0)
buf15 = reinterpret_tensor(buf13, (4, 4), (4, 1), 0); del buf13 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward], Original ATen: [aten.addmm]
extern_kernels.addmm(primals_8, reinterpret_tensor(buf14, (4, 4), (4, 1), 0), reinterpret_tensor(primals_7, (4, 4), (1, 4), 0), alpha=1, beta=1, out=buf15)
del primals_8
buf16 = buf1; del buf1 # reuse
buf17 = buf0; del buf0 # reuse
# Topologically Sorted Source Nodes: [input_1, x_3], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_8.run(primals_3, buf15, buf16, buf17, 4, grid=grid(4), stream=stream0)
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [input_1, x_3], Original ATen: [aten.add, aten.native_layer_norm]
triton_poi_fused_add_native_layer_norm_9.run(primals_3, buf15, buf16, buf17, primals_9, primals_10, buf18, 4, 4, grid=grid(4, 4), stream=stream0)
del primals_10
buf19 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(buf18, (4, 4), (1, 4), 0), reinterpret_tensor(primals_12, (4, 4), (1, 4), 0), out=buf19)
buf20 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_11, (4, 4), (1, 4), 0), reinterpret_tensor(primals_12, (4, 4), (1, 4), 16), out=buf20)
buf21 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_11, (4, 4), (1, 4), 0), reinterpret_tensor(primals_12, (4, 4), (1, 4), 32), out=buf21)
buf22 = reinterpret_tensor(buf21, (4, 1, 4), (4, 4, 1), 0); del buf21 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.add]
triton_poi_fused_add_2.run(buf22, primals_13, 16, grid=grid(16), stream=stream0)
buf23 = reinterpret_tensor(buf19, (4, 4, 1), (1, 4, 16), 0); del buf19 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.mul]
triton_poi_fused_mul_3.run(buf23, primals_13, 16, grid=grid(16), stream=stream0)
buf24 = reinterpret_tensor(buf20, (4, 1, 4), (4, 4, 1), 0); del buf20 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.add]
triton_poi_fused_add_4.run(buf24, primals_13, 16, grid=grid(16), stream=stream0)
del primals_13
buf25 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.bmm]
extern_kernels.bmm(buf23, reinterpret_tensor(buf24, (4, 1, 4), (1, 0, 4), 0), out=buf25)
buf26 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_10.run(buf25, buf26, 64, grid=grid(64), stream=stream0)
buf27 = buf25; del buf25 # reuse
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten._softmax]
triton_poi_fused__softmax_11.run(buf26, buf27, 64, grid=grid(64), stream=stream0)
del buf26
buf28 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.bmm]
extern_kernels.bmm(buf27, reinterpret_tensor(buf22, (4, 4, 1), (1, 4, 0), 0), out=buf28)
buf29 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
# Topologically Sorted Source Nodes: [multi_head_attention_forward_1], Original ATen: [aten.clone]
triton_poi_fused_clone_7.run(buf28, buf29, 4, 4, grid=grid(4, 4), stream=stream0)
buf30 = reinterpret_tensor(buf28, (4, 4), (4, 1), 0); del buf28 # reuse
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(reinterpret_tensor(buf29, (4, 4), (4, 1), 0), reinterpret_tensor(primals_14, (4, 4), (1, 4), 0), out=buf30)
buf31 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [input_1, input_2], Original ATen: [aten.add]
triton_poi_fused_add_12.run(primals_3, buf15, buf30, primals_15, buf31, 4, 4, grid=grid(4, 4), stream=stream0)
del primals_15
buf32 = buf17; del buf17 # reuse
buf33 = buf16; del buf16 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_0.run(buf31, buf32, buf33, 4, grid=grid(4), stream=stream0)
buf34 = buf30; del buf30 # reuse
# Topologically Sorted Source Nodes: [x_6], Original ATen: [aten.native_layer_norm]
triton_poi_fused_native_layer_norm_1.run(buf31, buf32, buf33, primals_16, primals_17, buf34, 16, grid=grid(16), stream=stream0)
del buf32
del buf33
del primals_17
buf35 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf34, reinterpret_tensor(primals_18, (4, 4), (1, 4), 0), out=buf35)
buf36 = buf35; del buf35 # reuse
# Topologically Sorted Source Nodes: [relu], Original ATen: [aten.relu]
triton_poi_fused_relu_13.run(buf36, primals_19, 16, grid=grid(16), stream=stream0)
del primals_19
buf37 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [], Original ATen: []
extern_kernels.mm(buf36, reinterpret_tensor(primals_20, (4, 4), (1, 4), 0), out=buf37)
buf38 = buf37; del buf37 # reuse
# Topologically Sorted Source Nodes: [input_3], Original ATen: [aten.add]
triton_poi_fused_add_14.run(buf38, buf31, primals_21, 16, grid=grid(16), stream=stream0)
del primals_21
return (buf38, primals_3, primals_9, primals_16, reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), reinterpret_tensor(buf2, (4, 4), (1, 4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 16), reinterpret_tensor(primals_5, (4, 4), (1, 4), 32), buf12, reinterpret_tensor(buf14, (4, 4), (4, 1), 0), buf15, reinterpret_tensor(primals_12, (4, 4), (1, 4), 0), reinterpret_tensor(buf18, (4, 4), (1, 4), 0), reinterpret_tensor(primals_11, (4, 4), (1, 4), 0), buf27, reinterpret_tensor(buf29, (4, 4), (4, 1), 0), buf31, buf34, buf36, primals_20, primals_18, primals_14, reinterpret_tensor(buf22, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf23, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf24, (4, 4, 1), (1, 4, 1), 0), primals_7, reinterpret_tensor(buf6, (4, 1, 4), (1, 1, 4), 0), reinterpret_tensor(buf8, (4, 4, 1), (1, 4, 1), 0), reinterpret_tensor(buf7, (4, 1, 4), (1, 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, ), (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, 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((12, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_6 = rand_strided((12, ), (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, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_10 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_11 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_12 = rand_strided((12, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_13 = rand_strided((12, ), (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, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_17 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_18 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_19 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_20 = rand_strided((4, 4), (4, 1), device='cuda:0', dtype=torch.float32)
primals_21 = 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, 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
from torch import nn
import torch.nn.functional as F
def _get_activation_fn(activation):
if activation == 'relu':
return F.relu
raise RuntimeError('activation shud be relu, not {}'.format(activation))
class TransformerDecoderLayer(nn.Module):
def __init__(self, d_model, nhead, dim_feedforward, dropout=0.3,
activation='relu'):
super(TransformerDecoderLayer, self).__init__()
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.norm3 = nn.LayerNorm(d_model)
self.nhead = nhead
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.dropout3 = nn.Dropout(dropout)
self.masked_attn = nn.MultiheadAttention(d_model, nhead, dropout=
dropout)
self.attn = nn.MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, dim_feedforward)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(dim_feedforward, d_model)
self.activation = _get_activation_fn(activation)
def __setstate__(self, state):
if 'activation' not in state:
state['activation'] = F.relu
super(TransformerDecoderLayer, self).__setstate__(state)
def forward(self, input, output, mask):
x = self.norm1(input)
x = torch.transpose(x, 0, 1)
x = self.masked_attn(x, x, x, attn_mask=mask)[0]
x = torch.transpose(x, 0, 1)
input = input + self.dropout1(x)
x = self.norm2(input)
x = torch.transpose(x, 0, 1)
output = torch.transpose(output, 0, 1)
x = self.attn(x, output, output)[0]
output = torch.transpose(output, 0, 1)
x = torch.transpose(x, 0, 1)
input = input + self.dropout2(x)
x = self.norm3(input)
x = self.linear2(self.dropout(self.activation(self.linear1(x))))
input = input + self.dropout3(x)
return input
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'d_model': 4, 'nhead': 4, 'dim_feedforward': 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
from torch import 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
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 = 4
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 = 16
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_add_2(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 + (8 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused_mul_3(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_add_4(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 + (4 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused__softmax_baddbmm_5(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 % 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + 4 * x2, 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 * x2), 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 * x2), 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 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = triton_helpers.maximum(tmp2, tmp5)
tmp9 = tmp7 + tmp8
tmp10 = triton_helpers.maximum(tmp6, tmp9)
tmp13 = tmp11 + tmp12
tmp14 = triton_helpers.maximum(tmp10, tmp13)
tmp15 = tmp2 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tmp5 - tmp14
tmp18 = tl_math.exp(tmp17)
tmp19 = tmp16 + tmp18
tmp20 = tmp9 - tmp14
tmp21 = tl_math.exp(tmp20)
tmp22 = tmp19 + tmp21
tmp23 = tmp13 - tmp14
tmp24 = tl_math.exp(tmp23)
tmp25 = tmp22 + tmp24
tl.store(out_ptr0 + x2, tmp14, xmask)
tl.store(out_ptr1 + x2, tmp25, xmask)
@triton.jit
def triton_poi_fused__softmax_baddbmm_6(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
x3 = xindex % 16
x4 = xindex
x5 = xindex // 4
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_out_ptr0 + x4, xmask)
tmp3 = tl.load(in_ptr1 + x5, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr2 + x5, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp5 = tl_math.exp(tmp4)
tmp7 = tmp5 / tmp6
tl.store(in_out_ptr0 + x4, tmp7, xmask)
@triton.jit
def triton_poi_fused_clone_7(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_native_layer_norm_8(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, 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 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x0, xmask)
tmp3 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (4 + x0), xmask)
tmp7 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (8 + x0), xmask)
tmp11 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp12 = tl.load(in_ptr1 + (12 + x0), xmask)
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_9(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, 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 + (x1 + 4 * y0), xmask & ymask)
tmp1 = tl.load(in_ptr1 + (y0 + 4 * x1), xmask & ymask)
tmp3 = tl.load(in_ptr2 + y0, ymask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + y0, ymask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + x1, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + x1, 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 + (x1 + 4 * y0), tmp13, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_10(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_11(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_add_12(in_ptr0, in_ptr1, in_ptr2, in_ptr3, 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 + (x1 + 4 * y0), xmask & ymask)
tmp1 = tl.load(in_ptr1 + (y0 + 4 * x1), xmask & ymask)
tmp3 = tl.load(in_ptr2 + (y0 + 4 * x1), xmask & ymask)
tmp4 = tl.load(in_ptr3 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tl.store(out_ptr0 + (x1 + 4 * y0), tmp6, xmask & ymask)
@triton.jit
def triton_poi_fused_relu_13(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)
@triton.jit
def triton_poi_fused_add_14(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, 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, (4,), (1,))
assert_size_stride(primals_2, (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, (12, 4), (4, 1))
assert_size_stride(primals_6, (12,), (1,))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4,), (1,))
assert_size_stride(primals_9, (4,), (1,))
assert_size_stride(primals_10, (4,), (1,))
assert_size_stride(primals_11, (4, 4), (4, 1))
assert_size_stride(primals_12, (12, 4), (4, 1))
assert_size_stride(primals_13, (12,), (1,))
assert_size_stride(primals_14, (4, 4), (4, 1))
assert_size_stride(primals_15, (4,), (1,))
assert_size_stride(primals_16, (4,), (1,))
assert_size_stride(primals_17, (4,), (1,))
assert_size_stride(primals_18, (4, 4), (4, 1))
assert_size_stride(primals_19, (4,), (1,))
assert_size_stride(primals_20, (4, 4), (4, 1))
assert_size_stride(primals_21, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
buf1 = empty_strided_cuda((4, 1), (1, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(4)](primals_3, buf0, buf1,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_native_layer_norm_1[grid(16)](primals_3, buf0,
buf1, primals_1, primals_2, buf2, 16, XBLOCK=16, num_warps=1,
num_stages=1)
del primals_1
del primals_2
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 16), out=buf4)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf2, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 32), out=buf5)
buf6 = reinterpret_tensor(buf5, (4, 1, 4), (4, 4, 1), 0)
del buf5
triton_poi_fused_add_2[grid(16)](buf6, primals_6, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf7 = reinterpret_tensor(buf3, (4, 4, 1), (1, 4, 16), 0)
del buf3
triton_poi_fused_mul_3[grid(16)](buf7, primals_6, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf8 = reinterpret_tensor(buf4, (4, 1, 4), (4, 4, 1), 0)
del buf4
triton_poi_fused_add_4[grid(16)](buf8, primals_6, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_6
buf9 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf7, reinterpret_tensor(buf8, (4, 1, 4), (1, 0,
4), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf11 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused__softmax_baddbmm_5[grid(16)](primals_4, buf9,
buf10, buf11, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf12 = buf9
del buf9
triton_poi_fused__softmax_baddbmm_6[grid(64)](buf12, primals_4,
buf10, buf11, 64, XBLOCK=64, num_warps=1, num_stages=1)
del primals_4
buf13 = reinterpret_tensor(buf11, (4, 4, 1), (4, 1, 1), 0)
del buf11
extern_kernels.bmm(buf12, reinterpret_tensor(buf6, (4, 4, 1), (1, 4,
0), 0), out=buf13)
buf14 = reinterpret_tensor(buf10, (4, 4, 1), (4, 1, 1), 0)
del buf10
triton_poi_fused_clone_7[grid(4, 4)](buf13, buf14, 4, 4, XBLOCK=4,
YBLOCK=4, num_warps=1, num_stages=1)
buf15 = reinterpret_tensor(buf13, (4, 4), (4, 1), 0)
del buf13
extern_kernels.addmm(primals_8, reinterpret_tensor(buf14, (4, 4), (
4, 1), 0), reinterpret_tensor(primals_7, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf15)
del primals_8
buf16 = buf1
del buf1
buf17 = buf0
del buf0
triton_poi_fused_add_native_layer_norm_8[grid(4)](primals_3, buf15,
buf16, buf17, 4, XBLOCK=4, num_warps=1, num_stages=1)
buf18 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_add_native_layer_norm_9[grid(4, 4)](primals_3,
buf15, buf16, buf17, primals_9, primals_10, buf18, 4, 4, XBLOCK
=4, YBLOCK=4, num_warps=1, num_stages=1)
del primals_10
buf19 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf18, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_12, (4, 4), (1, 4), 0), out=buf19)
buf20 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_11, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_12, (4, 4), (1, 4), 16), out=buf20)
buf21 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_11, (4, 4), (1, 4), 0),
reinterpret_tensor(primals_12, (4, 4), (1, 4), 32), out=buf21)
buf22 = reinterpret_tensor(buf21, (4, 1, 4), (4, 4, 1), 0)
del buf21
triton_poi_fused_add_2[grid(16)](buf22, primals_13, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf23 = reinterpret_tensor(buf19, (4, 4, 1), (1, 4, 16), 0)
del buf19
triton_poi_fused_mul_3[grid(16)](buf23, primals_13, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf24 = reinterpret_tensor(buf20, (4, 1, 4), (4, 4, 1), 0)
del buf20
triton_poi_fused_add_4[grid(16)](buf24, primals_13, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_13
buf25 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf23, reinterpret_tensor(buf24, (4, 1, 4), (1,
0, 4), 0), out=buf25)
buf26 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_10[grid(64)](buf25, buf26, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf27 = buf25
del buf25
triton_poi_fused__softmax_11[grid(64)](buf26, buf27, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf26
buf28 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
extern_kernels.bmm(buf27, reinterpret_tensor(buf22, (4, 4, 1), (1,
4, 0), 0), out=buf28)
buf29 = empty_strided_cuda((4, 4, 1), (4, 1, 1), torch.float32)
triton_poi_fused_clone_7[grid(4, 4)](buf28, buf29, 4, 4, XBLOCK=4,
YBLOCK=4, num_warps=1, num_stages=1)
buf30 = reinterpret_tensor(buf28, (4, 4), (4, 1), 0)
del buf28
extern_kernels.mm(reinterpret_tensor(buf29, (4, 4), (4, 1), 0),
reinterpret_tensor(primals_14, (4, 4), (1, 4), 0), out=buf30)
buf31 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_add_12[grid(4, 4)](primals_3, buf15, buf30,
primals_15, buf31, 4, 4, XBLOCK=4, YBLOCK=4, num_warps=1,
num_stages=1)
del primals_15
buf32 = buf17
del buf17
buf33 = buf16
del buf16
triton_poi_fused_native_layer_norm_0[grid(4)](buf31, buf32, buf33,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf34 = buf30
del buf30
triton_poi_fused_native_layer_norm_1[grid(16)](buf31, buf32, buf33,
primals_16, primals_17, buf34, 16, XBLOCK=16, num_warps=1,
num_stages=1)
del buf32
del buf33
del primals_17
buf35 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf34, reinterpret_tensor(primals_18, (4, 4), (1,
4), 0), out=buf35)
buf36 = buf35
del buf35
triton_poi_fused_relu_13[grid(16)](buf36, primals_19, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_19
buf37 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf36, reinterpret_tensor(primals_20, (4, 4), (1,
4), 0), out=buf37)
buf38 = buf37
del buf37
triton_poi_fused_add_14[grid(16)](buf38, buf31, primals_21, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_21
return (buf38, primals_3, primals_9, primals_16, reinterpret_tensor(
primals_5, (4, 4), (1, 4), 0), reinterpret_tensor(buf2, (4, 4), (1,
4), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 16),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 32), buf12,
reinterpret_tensor(buf14, (4, 4), (4, 1), 0), buf15,
reinterpret_tensor(primals_12, (4, 4), (1, 4), 0),
reinterpret_tensor(buf18, (4, 4), (1, 4), 0), reinterpret_tensor(
primals_11, (4, 4), (1, 4), 0), buf27, reinterpret_tensor(buf29, (4,
4), (4, 1), 0), buf31, buf34, buf36, primals_20, primals_18,
primals_14, reinterpret_tensor(buf22, (4, 1, 4), (1, 1, 4), 0),
reinterpret_tensor(buf23, (4, 1, 4), (1, 1, 4), 0),
reinterpret_tensor(buf24, (4, 4, 1), (1, 4, 1), 0), primals_7,
reinterpret_tensor(buf6, (4, 1, 4), (1, 1, 4), 0),
reinterpret_tensor(buf8, (4, 4, 1), (1, 4, 1), 0),
reinterpret_tensor(buf7, (4, 1, 4), (1, 1, 4), 0))
def _get_activation_fn(activation):
if activation == 'relu':
return F.relu
raise RuntimeError('activation shud be relu, not {}'.format(activation))
class TransformerDecoderLayerNew(nn.Module):
def __init__(self, d_model, nhead, dim_feedforward, dropout=0.3,
activation='relu'):
super(TransformerDecoderLayerNew, self).__init__()
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.norm3 = nn.LayerNorm(d_model)
self.nhead = nhead
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.dropout3 = nn.Dropout(dropout)
self.masked_attn = nn.MultiheadAttention(d_model, nhead, dropout=
dropout)
self.attn = nn.MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, dim_feedforward)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(dim_feedforward, d_model)
self.activation = _get_activation_fn(activation)
def __setstate__(self, state):
if 'activation' not in state:
state['activation'] = F.relu
super(TransformerDecoderLayerNew, self).__setstate__(state)
def forward(self, input_0, input_1, input_2):
primals_1 = self.norm1.weight
primals_2 = self.norm1.bias
primals_8 = self.norm2.weight
primals_9 = self.norm2.bias
primals_10 = self.norm3.weight
primals_15 = self.norm3.bias
primals_5 = self.masked_attn.in_proj_weight
primals_6 = self.masked_attn.in_proj_bias
primals_3 = self.masked_attn.out_proj.weight
primals_16 = self.masked_attn.out_proj.bias
primals_12 = self.attn.in_proj_weight
primals_13 = self.attn.in_proj_bias
primals_4 = self.attn.out_proj.weight
primals_17 = self.attn.out_proj.bias
primals_7 = self.linear1.weight
primals_19 = self.linear1.bias
primals_11 = self.linear2.weight
primals_21 = self.linear2.bias
primals_14 = input_0
primals_18 = input_1
primals_20 = 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, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21])
return output[0]
| salmon7ish/Video-Captioning | TransformerDecoderLayer | false | 4,393 | [
"MIT"
] | 0 | 08359b1824195a7f5eac5b58982efd19ebc6db01 | https://github.com/salmon7ish/Video-Captioning/tree/08359b1824195a7f5eac5b58982efd19ebc6db01 | import torch
from torch import nn
import torch.nn.functional as F
def _get_activation_fn(activation):
if activation == 'relu':
return F.relu
raise RuntimeError('activation shud be relu, not {}'.format(activation))
class Model(nn.Module):
def __init__(self, d_model, nhead, dim_feedforward, dropout=0.3,
activation='relu'):
super().__init__()
self.norm1 = nn.LayerNorm(d_model)
self.norm2 = nn.LayerNorm(d_model)
self.norm3 = nn.LayerNorm(d_model)
self.nhead = nhead
self.dropout1 = nn.Dropout(dropout)
self.dropout2 = nn.Dropout(dropout)
self.dropout3 = nn.Dropout(dropout)
self.masked_attn = nn.MultiheadAttention(d_model, nhead, dropout=
dropout)
self.attn = nn.MultiheadAttention(d_model, nhead, dropout=dropout)
self.linear1 = nn.Linear(d_model, dim_feedforward)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(dim_feedforward, d_model)
self.activation = _get_activation_fn(activation)
def __setstate__(self, state):
if 'activation' not in state:
state['activation'] = F.relu
super(TransformerDecoderLayer, self).__setstate__(state)
def forward(self, input, output, mask):
x = self.norm1(input)
x = torch.transpose(x, 0, 1)
x = self.masked_attn(x, x, x, attn_mask=mask)[0]
x = torch.transpose(x, 0, 1)
input = input + self.dropout1(x)
x = self.norm2(input)
x = torch.transpose(x, 0, 1)
output = torch.transpose(output, 0, 1)
x = self.attn(x, output, output)[0]
output = torch.transpose(output, 0, 1)
x = torch.transpose(x, 0, 1)
input = input + self.dropout2(x)
x = self.norm3(input)
x = self.linear2(self.dropout(self.activation(self.linear1(x))))
input = input + self.dropout3(x)
return input
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [4, 4, 4]
|
AvgPoolPadding | # 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_7/inductor_cache/gd/cgdmff72zhhf2ovmu434dne6gex5w2xlpp56xqfjp5a2nsjfiyrx.py
# Topologically Sorted Source Nodes: [out_1], Original ATen: [aten.avg_pool2d]
# Source node to ATen node mapping:
# out_1 => avg_pool2d
# Graph fragment:
# %avg_pool2d : [num_users=1] = call_function[target=torch.ops.aten.avg_pool2d.default](args = (%arg0_1, [1, 1], [1, 1]), kwargs = {})
triton_poi_fused_avg_pool2d_0 = async_compile.triton('triton_poi_fused_avg_pool2d_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_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_avg_pool2d_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 = 1.0
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + (x0), tmp2, 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: [out_1], Original ATen: [aten.avg_pool2d]
stream0 = get_raw_stream(0)
triton_poi_fused_avg_pool2d_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
import torch.nn.functional as F
class AvgPoolPadding(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super(AvgPoolPadding, self).__init__()
self.identity = nn.AvgPool2d(stride, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, x):
out = F.pad(x, (0, 0, 0, 0, 0, self.num_zeros))
out = self.identity(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_filters': 4, 'channels_in': 4, 'stride': 1}]
| 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_poi_fused_avg_pool2d_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 = 1.0
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x0, tmp2, 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_avg_pool2d_0[grid(256)](arg0_1, buf0, 256, XBLOCK=
128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class AvgPoolPaddingNew(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super(AvgPoolPaddingNew, self).__init__()
self.identity = nn.AvgPool2d(stride, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| sunqcc/Pytorch-HW-CIFAR10 | AvgPoolPadding | false | 4,394 | [
"MIT"
] | 0 | 33a55a5a832474083820b65c46f809ac98f8b109 | https://github.com/sunqcc/Pytorch-HW-CIFAR10/tree/33a55a5a832474083820b65c46f809ac98f8b109 | import torch
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self, num_filters, channels_in, stride):
super().__init__()
self.identity = nn.AvgPool2d(stride, stride=stride)
self.num_zeros = num_filters - channels_in
def forward(self, x):
out = F.pad(x, (0, 0, 0, 0, 0, self.num_zeros))
out = self.identity(out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4, 1]
|
GrayScaleToRGB | # 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_7/inductor_cache/47/c475jigalfpocuu3zf37chuxynaevtuk62hzapxn3o6voqpn4gg5.py
# Topologically Sorted Source Nodes: [repeat], Original ATen: [aten.repeat]
# Source node to ATen node mapping:
# repeat => repeat
# Graph fragment:
# %repeat : [num_users=1] = call_function[target=torch.ops.aten.repeat.default](args = (%arg0_1, [3, 1, 1]), kwargs = {})
triton_poi_fused_repeat_0 = async_compile.triton('triton_poi_fused_repeat_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_repeat_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_repeat_0(in_ptr0, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 192
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 + (16*(x1 % 4))), xmask)
tl.store(out_ptr0 + (x2), tmp0, 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((12, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [repeat], Original ATen: [aten.repeat]
stream0 = get_raw_stream(0)
triton_poi_fused_repeat_0.run(arg0_1, buf0, 192, grid=grid(192), 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), (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.utils.data
class GrayScaleToRGB(torch.nn.Module):
"""
Applies the transformation on an image to convert grayscale to rgb
"""
def __init__(self):
super().__init__()
def forward(self, sample):
return sample.repeat(3, 1, 1)
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
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_repeat_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 192
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 + 16 * (x1 % 4)), xmask)
tl.store(out_ptr0 + x2, tmp0, 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((12, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_repeat_0[grid(192)](arg0_1, buf0, 192, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class GrayScaleToRGBNew(torch.nn.Module):
"""
Applies the transformation on an image to convert grayscale to rgb
"""
def __init__(self):
super().__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
| saifullah3396/doc_robustness | GrayScaleToRGB | false | 4,395 | [
"Apache-2.0"
] | 0 | 80207fb44709d4b97de826331c074784be9c75ca | https://github.com/saifullah3396/doc_robustness/tree/80207fb44709d4b97de826331c074784be9c75ca | import torch
import torch.utils.data
class Model(torch.nn.Module):
"""
Applies the transformation on an image to convert grayscale to rgb
"""
def __init__(self):
super().__init__()
def forward(self, sample):
return sample.repeat(3, 1, 1)
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return []
|
SineActivation | # 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_7/inductor_cache/26/c264zszw4gigvwnid67zbfdd7jgsuyrkhlvddypypsnqnz7aete6.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 = ([%sin, %add_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=[512],
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_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_cat_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0, xnumel, XBLOCK : tl.constexpr):
xnumel = 512
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 = tl.load(in_ptr1 + (x0), tmp4 & xmask, eviction_policy='evict_last', other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl_math.sin(tmp7)
tmp9 = tl.full(tmp8.shape, 0.0, tmp8.dtype)
tmp10 = tl.where(tmp4, tmp8, tmp9)
tmp11 = tmp0 >= tmp3
tmp12 = tl.full([1], 8, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tl.load(in_ptr2 + ((4*x1) + ((-4) + x0)), tmp11 & xmask, eviction_policy='evict_last', other=0.0)
tmp15 = tl.load(in_ptr3 + ((-4) + x0), tmp11 & xmask, eviction_policy='evict_last', other=0.0)
tmp16 = tmp14 + tmp15
tmp17 = tl.full(tmp16.shape, 0.0, tmp16.dtype)
tmp18 = tl.where(tmp11, tmp16, tmp17)
tmp19 = tl.where(tmp4, tmp10, tmp18)
tl.store(out_ptr0 + (x2), tmp19, 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, 1))
assert_size_stride(primals_4, (4, ), (1, ))
assert_size_stride(primals_5, (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: [matmul], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_5, (64, 4), (4, 1), 0), primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
# Topologically Sorted Source Nodes: [matmul_1], Original ATen: [aten.mm]
extern_kernels.mm(reinterpret_tensor(primals_5, (64, 4), (4, 1), 0), primals_3, out=buf1)
del primals_3
buf2 = empty_strided_cuda((4, 4, 4, 8), (128, 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(buf0, primals_2, buf1, primals_4, buf2, 512, grid=grid(512), stream=stream0)
del buf1
del primals_4
return (buf2, primals_2, buf0, reinterpret_tensor(primals_5, (4, 64), (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, 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, 1), device='cuda:0', dtype=torch.float32)
primals_4 = rand_strided((4, ), (1, ), device='cuda:0', dtype=torch.float32)
primals_5 = 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])
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 t2v(tau, f, weight_linear, bias_linear, weight_periodic, bias_periodic,
arg=None):
if arg:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear, arg)
else:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear)
v2 = torch.matmul(tau, weight_periodic) + bias_periodic
return torch.cat([v1, v2], -1)
class SineActivation(nn.Module):
def __init__(self, in_features, output_features):
super(SineActivation, self).__init__()
self.output_features = output_features
self.weight_linear = nn.parameter.Parameter(torch.randn(in_features,
output_features))
self.bias_linear = nn.parameter.Parameter(torch.randn(output_features))
self.weight_periodic = nn.parameter.Parameter(torch.randn(
in_features, output_features))
self.bias_periodic = nn.parameter.Parameter(torch.randn(
output_features))
self.f = torch.sin
def forward(self, tau):
return t2v(tau, self.f, self.weight_linear, self.bias_linear, self.
weight_periodic, self.bias_periodic)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'output_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 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, in_ptr2, in_ptr3, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 512
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 = tl.load(in_ptr1 + x0, tmp4 & xmask, eviction_policy='evict_last',
other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl_math.sin(tmp7)
tmp9 = tl.full(tmp8.shape, 0.0, tmp8.dtype)
tmp10 = tl.where(tmp4, tmp8, tmp9)
tmp11 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp14 = tl.load(in_ptr2 + (4 * x1 + (-4 + x0)), tmp11 & xmask,
eviction_policy='evict_last', other=0.0)
tmp15 = tl.load(in_ptr3 + (-4 + x0), tmp11 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp16 = tmp14 + tmp15
tmp17 = tl.full(tmp16.shape, 0.0, tmp16.dtype)
tmp18 = tl.where(tmp11, tmp16, tmp17)
tmp19 = tl.where(tmp4, tmp10, tmp18)
tl.store(out_ptr0 + x2, tmp19, 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, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (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_5, (64, 4), (4, 1), 0),
primals_1, out=buf0)
del primals_1
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_5, (64, 4), (4, 1), 0),
primals_3, out=buf1)
del primals_3
buf2 = empty_strided_cuda((4, 4, 4, 8), (128, 32, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(512)](buf0, primals_2, buf1, primals_4,
buf2, 512, XBLOCK=128, num_warps=4, num_stages=1)
del buf1
del primals_4
return buf2, primals_2, buf0, reinterpret_tensor(primals_5, (4, 64), (1,
4), 0)
def t2v(tau, f, weight_linear, bias_linear, weight_periodic, bias_periodic,
arg=None):
if arg:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear, arg)
else:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear)
v2 = torch.matmul(tau, weight_periodic) + bias_periodic
return torch.cat([v1, v2], -1)
class SineActivationNew(nn.Module):
def __init__(self, in_features, output_features):
super(SineActivationNew, self).__init__()
self.output_features = output_features
self.weight_linear = nn.parameter.Parameter(torch.randn(in_features,
output_features))
self.bias_linear = nn.parameter.Parameter(torch.randn(output_features))
self.weight_periodic = nn.parameter.Parameter(torch.randn(
in_features, output_features))
self.bias_periodic = nn.parameter.Parameter(torch.randn(
output_features))
self.f = torch.sin
def forward(self, input_0):
primals_1 = self.weight_linear
primals_2 = self.bias_linear
primals_3 = self.weight_periodic
primals_4 = self.bias_periodic
primals_5 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
| sungreong/PyTimeSeries | SineActivation | false | 4,396 | [
"MIT"
] | 0 | d5321c1226fc7fb6a45fec7009843894be417594 | https://github.com/sungreong/PyTimeSeries/tree/d5321c1226fc7fb6a45fec7009843894be417594 | import torch
import torch.nn as nn
def t2v(tau, f, weight_linear, bias_linear, weight_periodic, bias_periodic,
arg=None):
if arg:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear, arg)
else:
v1 = f(torch.matmul(tau, weight_linear) + bias_linear)
v2 = torch.matmul(tau, weight_periodic) + bias_periodic
return torch.cat([v1, v2], -1)
class Model(nn.Module):
def __init__(self, in_features, output_features):
super().__init__()
self.output_features = output_features
self.weight_linear = nn.parameter.Parameter(torch.randn(in_features,
output_features))
self.bias_linear = nn.parameter.Parameter(torch.randn(output_features))
self.weight_periodic = nn.parameter.Parameter(torch.randn(
in_features, output_features))
self.bias_periodic = nn.parameter.Parameter(torch.randn(
output_features))
self.f = torch.sin
def forward(self, tau):
return t2v(tau, self.f, self.weight_linear, self.bias_linear, self.
weight_periodic, self.bias_periodic)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [4, 4]
|
GraphConvolution | # 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_7/inductor_cache/ko/ckoyo74xxqfts3weuv4uuiyaxxpdybgkr3f5hnr3e7qk2z5coapt.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 = (%view_1, %view_3), 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=[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_0', '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_0(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_out_ptr0 + (x2), xmask)
tmp1 = tl.load(in_ptr0 + (x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (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 = 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), (16, 4, 1))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (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 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
# Topologically Sorted Source Nodes: [bmm], Original ATen: [aten.bmm]
extern_kernels.bmm(primals_4, primals_3, 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(buf1, (16, 4), (4, 1), 0), reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf2)
del primals_5
buf3 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0); del buf0 # reuse
# Topologically Sorted Source Nodes: [add], Original ATen: [aten.add]
stream0 = get_raw_stream(0)
triton_poi_fused_add_0.run(buf3, primals_2, buf2, primals_6, 64, grid=grid(64), stream=stream0)
del buf2
del primals_2
del primals_6
return (buf3, reinterpret_tensor(primals_3, (16, 4), (4, 1), 0), reinterpret_tensor(buf1, (16, 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, ), (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, 4), (4, 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)
| from torch.nn import Module
import torch
from torch import nn
import torch.autograd
from torch.nn.modules.module import Module
class GraphConvolution(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907.
"""
def __init__(self, state_dim, name='', out_state_dim=None):
super(GraphConvolution, self).__init__()
self.state_dim = state_dim
if out_state_dim is None:
self.out_state_dim = state_dim
else:
self.out_state_dim = out_state_dim
self.fc1 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.fc2 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.name = name
def forward(self, input, adj):
state_in = self.fc1(input)
forward_input = self.fc2(torch.bmm(adj, input))
return state_in + forward_input
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'state_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.nn import Module
from torch import nn
import torch.autograd
from torch.nn.modules.module import Module
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, 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_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + 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 = 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), (16, 4, 1))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (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 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(primals_4, primals_3, out=buf1)
del primals_4
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf2)
del primals_5
buf3 = reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_add_0[grid(64)](buf3, primals_2, buf2, primals_6,
64, XBLOCK=64, num_warps=1, num_stages=1)
del buf2
del primals_2
del primals_6
return buf3, reinterpret_tensor(primals_3, (16, 4), (4, 1), 0
), reinterpret_tensor(buf1, (16, 4), (4, 1), 0)
class GraphConvolutionNew(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907.
"""
def __init__(self, state_dim, name='', out_state_dim=None):
super(GraphConvolutionNew, self).__init__()
self.state_dim = state_dim
if out_state_dim is None:
self.out_state_dim = state_dim
else:
self.out_state_dim = out_state_dim
self.fc1 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.fc2 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.name = name
def forward(self, input_0, input_1):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_5 = self.fc2.weight
primals_6 = self.fc2.bias
primals_3 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
| sumanmichael/Palmira_pb | GraphConvolution | false | 4,397 | [
"MIT"
] | 0 | 8ca9f370ccd9bba694317be648ce5e4f4c55d0e7 | https://github.com/sumanmichael/Palmira_pb/tree/8ca9f370ccd9bba694317be648ce5e4f4c55d0e7 | from torch.nn import Module
import torch
from torch import nn
import torch.autograd
from torch.nn.modules.module import Module
class Model(Module):
"""
Simple GCN layer, similar to https://arxiv.org/abs/1609.02907.
"""
def __init__(self, state_dim, name='', out_state_dim=None):
super().__init__()
self.state_dim = state_dim
if out_state_dim is None:
self.out_state_dim = state_dim
else:
self.out_state_dim = out_state_dim
self.fc1 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.fc2 = nn.Linear(in_features=self.state_dim, out_features=self.
out_state_dim)
self.name = name
def forward(self, input, adj):
state_in = self.fc1(input)
forward_input = self.fc2(torch.bmm(adj, input))
return state_in + forward_input
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [4]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.