text
stringlengths 7
318k
| id
stringlengths 14
166
| metadata
dict | __index_level_0__
int64 0
439
|
|---|---|---|---|
# Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
# William Peebles and Saining Xie
#
# Copyright (c) 2021 OpenAI
# MIT License
#
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import Dict, List, Optional, Tuple, Union
import torch
from ...models import AutoencoderKL, Transformer2DModel
from ...schedulers import KarrasDiffusionSchedulers
from ...utils.torch_utils import randn_tensor
from ..pipeline_utils import DiffusionPipeline, ImagePipelineOutput
class DiTPipeline(DiffusionPipeline):
r"""
Pipeline for image generation based on a Transformer backbone instead of a UNet.
This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
implemented for all pipelines (downloading, saving, running on a particular device, etc.).
Parameters:
transformer ([`Transformer2DModel`]):
A class conditioned `Transformer2DModel` to denoise the encoded image latents.
vae ([`AutoencoderKL`]):
Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations.
scheduler ([`DDIMScheduler`]):
A scheduler to be used in combination with `transformer` to denoise the encoded image latents.
"""
model_cpu_offload_seq = "transformer->vae"
def __init__(
self,
transformer: Transformer2DModel,
vae: AutoencoderKL,
scheduler: KarrasDiffusionSchedulers,
id2label: Optional[Dict[int, str]] = None,
):
super().__init__()
self.register_modules(transformer=transformer, vae=vae, scheduler=scheduler)
# create a imagenet -> id dictionary for easier use
self.labels = {}
if id2label is not None:
for key, value in id2label.items():
for label in value.split(","):
self.labels[label.lstrip().rstrip()] = int(key)
self.labels = dict(sorted(self.labels.items()))
def get_label_ids(self, label: Union[str, List[str]]) -> List[int]:
r"""
Map label strings from ImageNet to corresponding class ids.
Parameters:
label (`str` or `dict` of `str`):
Label strings to be mapped to class ids.
Returns:
`list` of `int`:
Class ids to be processed by pipeline.
"""
if not isinstance(label, list):
label = list(label)
for l in label:
if l not in self.labels:
raise ValueError(
f"{l} does not exist. Please make sure to select one of the following labels: \n {self.labels}."
)
return [self.labels[l] for l in label]
@torch.no_grad()
def __call__(
self,
class_labels: List[int],
guidance_scale: float = 4.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
num_inference_steps: int = 50,
output_type: Optional[str] = "pil",
return_dict: bool = True,
) -> Union[ImagePipelineOutput, Tuple]:
r"""
The call function to the pipeline for generation.
Args:
class_labels (List[int]):
List of ImageNet class labels for the images to be generated.
guidance_scale (`float`, *optional*, defaults to 4.0):
A higher guidance scale value encourages the model to generate images closely linked to the text
`prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
generator (`torch.Generator`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic.
num_inference_steps (`int`, *optional*, defaults to 250):
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generated image. Choose between `PIL.Image` or `np.array`.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`ImagePipelineOutput`] instead of a plain tuple.
Examples:
```py
>>> from diffusers import DiTPipeline, DPMSolverMultistepScheduler
>>> import torch
>>> pipe = DiTPipeline.from_pretrained("facebook/DiT-XL-2-256", torch_dtype=torch.float16)
>>> pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
>>> pipe = pipe.to("cuda")
>>> # pick words from Imagenet class labels
>>> pipe.labels # to print all available words
>>> # pick words that exist in ImageNet
>>> words = ["white shark", "umbrella"]
>>> class_ids = pipe.get_label_ids(words)
>>> generator = torch.manual_seed(33)
>>> output = pipe(class_labels=class_ids, num_inference_steps=25, generator=generator)
>>> image = output.images[0] # label 'white shark'
```
Returns:
[`~pipelines.ImagePipelineOutput`] or `tuple`:
If `return_dict` is `True`, [`~pipelines.ImagePipelineOutput`] is returned, otherwise a `tuple` is
returned where the first element is a list with the generated images
"""
batch_size = len(class_labels)
latent_size = self.transformer.config.sample_size
latent_channels = self.transformer.config.in_channels
latents = randn_tensor(
shape=(batch_size, latent_channels, latent_size, latent_size),
generator=generator,
device=self._execution_device,
dtype=self.transformer.dtype,
)
latent_model_input = torch.cat([latents] * 2) if guidance_scale > 1 else latents
class_labels = torch.tensor(class_labels, device=self._execution_device).reshape(-1)
class_null = torch.tensor([1000] * batch_size, device=self._execution_device)
class_labels_input = torch.cat([class_labels, class_null], 0) if guidance_scale > 1 else class_labels
# set step values
self.scheduler.set_timesteps(num_inference_steps)
for t in self.progress_bar(self.scheduler.timesteps):
if guidance_scale > 1:
half = latent_model_input[: len(latent_model_input) // 2]
latent_model_input = torch.cat([half, half], dim=0)
latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
timesteps = t
if not torch.is_tensor(timesteps):
# TODO: this requires sync between CPU and GPU. So try to pass timesteps as tensors if you can
# This would be a good case for the `match` statement (Python 3.10+)
is_mps = latent_model_input.device.type == "mps"
if isinstance(timesteps, float):
dtype = torch.float32 if is_mps else torch.float64
else:
dtype = torch.int32 if is_mps else torch.int64
timesteps = torch.tensor([timesteps], dtype=dtype, device=latent_model_input.device)
elif len(timesteps.shape) == 0:
timesteps = timesteps[None].to(latent_model_input.device)
# broadcast to batch dimension in a way that's compatible with ONNX/Core ML
timesteps = timesteps.expand(latent_model_input.shape[0])
# predict noise model_output
noise_pred = self.transformer(
latent_model_input, timestep=timesteps, class_labels=class_labels_input
).sample
# perform guidance
if guidance_scale > 1:
eps, rest = noise_pred[:, :latent_channels], noise_pred[:, latent_channels:]
cond_eps, uncond_eps = torch.split(eps, len(eps) // 2, dim=0)
half_eps = uncond_eps + guidance_scale * (cond_eps - uncond_eps)
eps = torch.cat([half_eps, half_eps], dim=0)
noise_pred = torch.cat([eps, rest], dim=1)
# learned sigma
if self.transformer.config.out_channels // 2 == latent_channels:
model_output, _ = torch.split(noise_pred, latent_channels, dim=1)
else:
model_output = noise_pred
# compute previous image: x_t -> x_t-1
latent_model_input = self.scheduler.step(model_output, t, latent_model_input).prev_sample
if guidance_scale > 1:
latents, _ = latent_model_input.chunk(2, dim=0)
else:
latents = latent_model_input
latents = 1 / self.vae.config.scaling_factor * latents
samples = self.vae.decode(latents).sample
samples = (samples / 2 + 0.5).clamp(0, 1)
# we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
samples = samples.cpu().permute(0, 2, 3, 1).float().numpy()
if output_type == "pil":
samples = self.numpy_to_pil(samples)
if not return_dict:
return (samples,)
return ImagePipelineOutput(images=samples)
| diffusers/src/diffusers/pipelines/dit/pipeline_dit.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/dit/pipeline_dit.py",
"repo_id": "diffusers",
"token_count": 4180
} | 118 |
from dataclasses import dataclass
from typing import TYPE_CHECKING, List, Optional, Union
import numpy as np
import PIL
from PIL import Image
from ...utils import (
DIFFUSERS_SLOW_IMPORT,
OptionalDependencyNotAvailable,
_LazyModule,
get_objects_from_module,
is_torch_available,
is_transformers_available,
)
_dummy_objects = {}
_import_structure = {}
try:
if not (is_transformers_available() and is_torch_available()):
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
from ...utils import dummy_torch_and_transformers_objects # noqa F403
_dummy_objects.update(get_objects_from_module(dummy_torch_and_transformers_objects))
else:
_import_structure["image_encoder"] = ["PaintByExampleImageEncoder"]
_import_structure["pipeline_paint_by_example"] = ["PaintByExamplePipeline"]
if TYPE_CHECKING or DIFFUSERS_SLOW_IMPORT:
try:
if not (is_transformers_available() and is_torch_available()):
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
from ...utils.dummy_torch_and_transformers_objects import *
else:
from .image_encoder import PaintByExampleImageEncoder
from .pipeline_paint_by_example import PaintByExamplePipeline
else:
import sys
sys.modules[__name__] = _LazyModule(
__name__,
globals()["__file__"],
_import_structure,
module_spec=__spec__,
)
for name, value in _dummy_objects.items():
setattr(sys.modules[__name__], name, value)
| diffusers/src/diffusers/pipelines/paint_by_example/__init__.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/paint_by_example/__init__.py",
"repo_id": "diffusers",
"token_count": 599
} | 119 |
# Copyright 2023 Open AI and The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import math
from dataclasses import dataclass
from typing import Dict, Optional, Tuple
import numpy as np
import torch
import torch.nn.functional as F
from torch import nn
from ...configuration_utils import ConfigMixin, register_to_config
from ...models import ModelMixin
from ...utils import BaseOutput
from .camera import create_pan_cameras
def sample_pmf(pmf: torch.Tensor, n_samples: int) -> torch.Tensor:
r"""
Sample from the given discrete probability distribution with replacement.
The i-th bin is assumed to have mass pmf[i].
Args:
pmf: [batch_size, *shape, n_samples, 1] where (pmf.sum(dim=-2) == 1).all()
n_samples: number of samples
Return:
indices sampled with replacement
"""
*shape, support_size, last_dim = pmf.shape
assert last_dim == 1
cdf = torch.cumsum(pmf.view(-1, support_size), dim=1)
inds = torch.searchsorted(cdf, torch.rand(cdf.shape[0], n_samples, device=cdf.device))
return inds.view(*shape, n_samples, 1).clamp(0, support_size - 1)
def posenc_nerf(x: torch.Tensor, min_deg: int = 0, max_deg: int = 15) -> torch.Tensor:
"""
Concatenate x and its positional encodings, following NeRF.
Reference: https://arxiv.org/pdf/2210.04628.pdf
"""
if min_deg == max_deg:
return x
scales = 2.0 ** torch.arange(min_deg, max_deg, dtype=x.dtype, device=x.device)
*shape, dim = x.shape
xb = (x.reshape(-1, 1, dim) * scales.view(1, -1, 1)).reshape(*shape, -1)
assert xb.shape[-1] == dim * (max_deg - min_deg)
emb = torch.cat([xb, xb + math.pi / 2.0], axis=-1).sin()
return torch.cat([x, emb], dim=-1)
def encode_position(position):
return posenc_nerf(position, min_deg=0, max_deg=15)
def encode_direction(position, direction=None):
if direction is None:
return torch.zeros_like(posenc_nerf(position, min_deg=0, max_deg=8))
else:
return posenc_nerf(direction, min_deg=0, max_deg=8)
def _sanitize_name(x: str) -> str:
return x.replace(".", "__")
def integrate_samples(volume_range, ts, density, channels):
r"""
Function integrating the model output.
Args:
volume_range: Specifies the integral range [t0, t1]
ts: timesteps
density: torch.Tensor [batch_size, *shape, n_samples, 1]
channels: torch.Tensor [batch_size, *shape, n_samples, n_channels]
returns:
channels: integrated rgb output weights: torch.Tensor [batch_size, *shape, n_samples, 1] (density
*transmittance)[i] weight for each rgb output at [..., i, :]. transmittance: transmittance of this volume
)
"""
# 1. Calculate the weights
_, _, dt = volume_range.partition(ts)
ddensity = density * dt
mass = torch.cumsum(ddensity, dim=-2)
transmittance = torch.exp(-mass[..., -1, :])
alphas = 1.0 - torch.exp(-ddensity)
Ts = torch.exp(torch.cat([torch.zeros_like(mass[..., :1, :]), -mass[..., :-1, :]], dim=-2))
# This is the probability of light hitting and reflecting off of
# something at depth [..., i, :].
weights = alphas * Ts
# 2. Integrate channels
channels = torch.sum(channels * weights, dim=-2)
return channels, weights, transmittance
def volume_query_points(volume, grid_size):
indices = torch.arange(grid_size**3, device=volume.bbox_min.device)
zs = indices % grid_size
ys = torch.div(indices, grid_size, rounding_mode="trunc") % grid_size
xs = torch.div(indices, grid_size**2, rounding_mode="trunc") % grid_size
combined = torch.stack([xs, ys, zs], dim=1)
return (combined.float() / (grid_size - 1)) * (volume.bbox_max - volume.bbox_min) + volume.bbox_min
def _convert_srgb_to_linear(u: torch.Tensor):
return torch.where(u <= 0.04045, u / 12.92, ((u + 0.055) / 1.055) ** 2.4)
def _create_flat_edge_indices(
flat_cube_indices: torch.Tensor,
grid_size: Tuple[int, int, int],
):
num_xs = (grid_size[0] - 1) * grid_size[1] * grid_size[2]
y_offset = num_xs
num_ys = grid_size[0] * (grid_size[1] - 1) * grid_size[2]
z_offset = num_xs + num_ys
return torch.stack(
[
# Edges spanning x-axis.
flat_cube_indices[:, 0] * grid_size[1] * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2],
flat_cube_indices[:, 0] * grid_size[1] * grid_size[2]
+ (flat_cube_indices[:, 1] + 1) * grid_size[2]
+ flat_cube_indices[:, 2],
flat_cube_indices[:, 0] * grid_size[1] * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2]
+ 1,
flat_cube_indices[:, 0] * grid_size[1] * grid_size[2]
+ (flat_cube_indices[:, 1] + 1) * grid_size[2]
+ flat_cube_indices[:, 2]
+ 1,
# Edges spanning y-axis.
(
y_offset
+ flat_cube_indices[:, 0] * (grid_size[1] - 1) * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2]
),
(
y_offset
+ (flat_cube_indices[:, 0] + 1) * (grid_size[1] - 1) * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2]
),
(
y_offset
+ flat_cube_indices[:, 0] * (grid_size[1] - 1) * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2]
+ 1
),
(
y_offset
+ (flat_cube_indices[:, 0] + 1) * (grid_size[1] - 1) * grid_size[2]
+ flat_cube_indices[:, 1] * grid_size[2]
+ flat_cube_indices[:, 2]
+ 1
),
# Edges spanning z-axis.
(
z_offset
+ flat_cube_indices[:, 0] * grid_size[1] * (grid_size[2] - 1)
+ flat_cube_indices[:, 1] * (grid_size[2] - 1)
+ flat_cube_indices[:, 2]
),
(
z_offset
+ (flat_cube_indices[:, 0] + 1) * grid_size[1] * (grid_size[2] - 1)
+ flat_cube_indices[:, 1] * (grid_size[2] - 1)
+ flat_cube_indices[:, 2]
),
(
z_offset
+ flat_cube_indices[:, 0] * grid_size[1] * (grid_size[2] - 1)
+ (flat_cube_indices[:, 1] + 1) * (grid_size[2] - 1)
+ flat_cube_indices[:, 2]
),
(
z_offset
+ (flat_cube_indices[:, 0] + 1) * grid_size[1] * (grid_size[2] - 1)
+ (flat_cube_indices[:, 1] + 1) * (grid_size[2] - 1)
+ flat_cube_indices[:, 2]
),
],
dim=-1,
)
class VoidNeRFModel(nn.Module):
"""
Implements the default empty space model where all queries are rendered as background.
"""
def __init__(self, background, channel_scale=255.0):
super().__init__()
background = nn.Parameter(torch.from_numpy(np.array(background)).to(dtype=torch.float32) / channel_scale)
self.register_buffer("background", background)
def forward(self, position):
background = self.background[None].to(position.device)
shape = position.shape[:-1]
ones = [1] * (len(shape) - 1)
n_channels = background.shape[-1]
background = torch.broadcast_to(background.view(background.shape[0], *ones, n_channels), [*shape, n_channels])
return background
@dataclass
class VolumeRange:
t0: torch.Tensor
t1: torch.Tensor
intersected: torch.Tensor
def __post_init__(self):
assert self.t0.shape == self.t1.shape == self.intersected.shape
def partition(self, ts):
"""
Partitions t0 and t1 into n_samples intervals.
Args:
ts: [batch_size, *shape, n_samples, 1]
Return:
lower: [batch_size, *shape, n_samples, 1] upper: [batch_size, *shape, n_samples, 1] delta: [batch_size,
*shape, n_samples, 1]
where
ts \\in [lower, upper] deltas = upper - lower
"""
mids = (ts[..., 1:, :] + ts[..., :-1, :]) * 0.5
lower = torch.cat([self.t0[..., None, :], mids], dim=-2)
upper = torch.cat([mids, self.t1[..., None, :]], dim=-2)
delta = upper - lower
assert lower.shape == upper.shape == delta.shape == ts.shape
return lower, upper, delta
class BoundingBoxVolume(nn.Module):
"""
Axis-aligned bounding box defined by the two opposite corners.
"""
def __init__(
self,
*,
bbox_min,
bbox_max,
min_dist: float = 0.0,
min_t_range: float = 1e-3,
):
"""
Args:
bbox_min: the left/bottommost corner of the bounding box
bbox_max: the other corner of the bounding box
min_dist: all rays should start at least this distance away from the origin.
"""
super().__init__()
self.min_dist = min_dist
self.min_t_range = min_t_range
self.bbox_min = torch.tensor(bbox_min)
self.bbox_max = torch.tensor(bbox_max)
self.bbox = torch.stack([self.bbox_min, self.bbox_max])
assert self.bbox.shape == (2, 3)
assert min_dist >= 0.0
assert min_t_range > 0.0
def intersect(
self,
origin: torch.Tensor,
direction: torch.Tensor,
t0_lower: Optional[torch.Tensor] = None,
epsilon=1e-6,
):
"""
Args:
origin: [batch_size, *shape, 3]
direction: [batch_size, *shape, 3]
t0_lower: Optional [batch_size, *shape, 1] lower bound of t0 when intersecting this volume.
params: Optional meta parameters in case Volume is parametric
epsilon: to stabilize calculations
Return:
A tuple of (t0, t1, intersected) where each has a shape [batch_size, *shape, 1]. If a ray intersects with
the volume, `o + td` is in the volume for all t in [t0, t1]. If the volume is bounded, t1 is guaranteed to
be on the boundary of the volume.
"""
batch_size, *shape, _ = origin.shape
ones = [1] * len(shape)
bbox = self.bbox.view(1, *ones, 2, 3).to(origin.device)
def _safe_divide(a, b, epsilon=1e-6):
return a / torch.where(b < 0, b - epsilon, b + epsilon)
ts = _safe_divide(bbox - origin[..., None, :], direction[..., None, :], epsilon=epsilon)
# Cases to think about:
#
# 1. t1 <= t0: the ray does not pass through the AABB.
# 2. t0 < t1 <= 0: the ray intersects but the BB is behind the origin.
# 3. t0 <= 0 <= t1: the ray starts from inside the BB
# 4. 0 <= t0 < t1: the ray is not inside and intersects with the BB twice.
#
# 1 and 4 are clearly handled from t0 < t1 below.
# Making t0 at least min_dist (>= 0) takes care of 2 and 3.
t0 = ts.min(dim=-2).values.max(dim=-1, keepdim=True).values.clamp(self.min_dist)
t1 = ts.max(dim=-2).values.min(dim=-1, keepdim=True).values
assert t0.shape == t1.shape == (batch_size, *shape, 1)
if t0_lower is not None:
assert t0.shape == t0_lower.shape
t0 = torch.maximum(t0, t0_lower)
intersected = t0 + self.min_t_range < t1
t0 = torch.where(intersected, t0, torch.zeros_like(t0))
t1 = torch.where(intersected, t1, torch.ones_like(t1))
return VolumeRange(t0=t0, t1=t1, intersected=intersected)
class StratifiedRaySampler(nn.Module):
"""
Instead of fixed intervals, a sample is drawn uniformly at random from each interval.
"""
def __init__(self, depth_mode: str = "linear"):
"""
:param depth_mode: linear samples ts linearly in depth. harmonic ensures
closer points are sampled more densely.
"""
self.depth_mode = depth_mode
assert self.depth_mode in ("linear", "geometric", "harmonic")
def sample(
self,
t0: torch.Tensor,
t1: torch.Tensor,
n_samples: int,
epsilon: float = 1e-3,
) -> torch.Tensor:
"""
Args:
t0: start time has shape [batch_size, *shape, 1]
t1: finish time has shape [batch_size, *shape, 1]
n_samples: number of ts to sample
Return:
sampled ts of shape [batch_size, *shape, n_samples, 1]
"""
ones = [1] * (len(t0.shape) - 1)
ts = torch.linspace(0, 1, n_samples).view(*ones, n_samples).to(t0.dtype).to(t0.device)
if self.depth_mode == "linear":
ts = t0 * (1.0 - ts) + t1 * ts
elif self.depth_mode == "geometric":
ts = (t0.clamp(epsilon).log() * (1.0 - ts) + t1.clamp(epsilon).log() * ts).exp()
elif self.depth_mode == "harmonic":
# The original NeRF recommends this interpolation scheme for
# spherical scenes, but there could be some weird edge cases when
# the observer crosses from the inner to outer volume.
ts = 1.0 / (1.0 / t0.clamp(epsilon) * (1.0 - ts) + 1.0 / t1.clamp(epsilon) * ts)
mids = 0.5 * (ts[..., 1:] + ts[..., :-1])
upper = torch.cat([mids, t1], dim=-1)
lower = torch.cat([t0, mids], dim=-1)
# yiyi notes: add a random seed here for testing, don't forget to remove
torch.manual_seed(0)
t_rand = torch.rand_like(ts)
ts = lower + (upper - lower) * t_rand
return ts.unsqueeze(-1)
class ImportanceRaySampler(nn.Module):
"""
Given the initial estimate of densities, this samples more from regions/bins expected to have objects.
"""
def __init__(
self,
volume_range: VolumeRange,
ts: torch.Tensor,
weights: torch.Tensor,
blur_pool: bool = False,
alpha: float = 1e-5,
):
"""
Args:
volume_range: the range in which a ray intersects the given volume.
ts: earlier samples from the coarse rendering step
weights: discretized version of density * transmittance
blur_pool: if true, use 2-tap max + 2-tap blur filter from mip-NeRF.
alpha: small value to add to weights.
"""
self.volume_range = volume_range
self.ts = ts.clone().detach()
self.weights = weights.clone().detach()
self.blur_pool = blur_pool
self.alpha = alpha
@torch.no_grad()
def sample(self, t0: torch.Tensor, t1: torch.Tensor, n_samples: int) -> torch.Tensor:
"""
Args:
t0: start time has shape [batch_size, *shape, 1]
t1: finish time has shape [batch_size, *shape, 1]
n_samples: number of ts to sample
Return:
sampled ts of shape [batch_size, *shape, n_samples, 1]
"""
lower, upper, _ = self.volume_range.partition(self.ts)
batch_size, *shape, n_coarse_samples, _ = self.ts.shape
weights = self.weights
if self.blur_pool:
padded = torch.cat([weights[..., :1, :], weights, weights[..., -1:, :]], dim=-2)
maxes = torch.maximum(padded[..., :-1, :], padded[..., 1:, :])
weights = 0.5 * (maxes[..., :-1, :] + maxes[..., 1:, :])
weights = weights + self.alpha
pmf = weights / weights.sum(dim=-2, keepdim=True)
inds = sample_pmf(pmf, n_samples)
assert inds.shape == (batch_size, *shape, n_samples, 1)
assert (inds >= 0).all() and (inds < n_coarse_samples).all()
t_rand = torch.rand(inds.shape, device=inds.device)
lower_ = torch.gather(lower, -2, inds)
upper_ = torch.gather(upper, -2, inds)
ts = lower_ + (upper_ - lower_) * t_rand
ts = torch.sort(ts, dim=-2).values
return ts
@dataclass
class MeshDecoderOutput(BaseOutput):
"""
A 3D triangle mesh with optional data at the vertices and faces.
Args:
verts (`torch.Tensor` of shape `(N, 3)`):
array of vertext coordinates
faces (`torch.Tensor` of shape `(N, 3)`):
array of triangles, pointing to indices in verts.
vertext_channels (Dict):
vertext coordinates for each color channel
"""
verts: torch.Tensor
faces: torch.Tensor
vertex_channels: Dict[str, torch.Tensor]
class MeshDecoder(nn.Module):
"""
Construct meshes from Signed distance functions (SDFs) using marching cubes method
"""
def __init__(self):
super().__init__()
cases = torch.zeros(256, 5, 3, dtype=torch.long)
masks = torch.zeros(256, 5, dtype=torch.bool)
self.register_buffer("cases", cases)
self.register_buffer("masks", masks)
def forward(self, field: torch.Tensor, min_point: torch.Tensor, size: torch.Tensor):
"""
For a signed distance field, produce a mesh using marching cubes.
:param field: a 3D tensor of field values, where negative values correspond
to the outside of the shape. The dimensions correspond to the x, y, and z directions, respectively.
:param min_point: a tensor of shape [3] containing the point corresponding
to (0, 0, 0) in the field.
:param size: a tensor of shape [3] containing the per-axis distance from the
(0, 0, 0) field corner and the (-1, -1, -1) field corner.
"""
assert len(field.shape) == 3, "input must be a 3D scalar field"
dev = field.device
cases = self.cases.to(dev)
masks = self.masks.to(dev)
min_point = min_point.to(dev)
size = size.to(dev)
grid_size = field.shape
grid_size_tensor = torch.tensor(grid_size).to(size)
# Create bitmasks between 0 and 255 (inclusive) indicating the state
# of the eight corners of each cube.
bitmasks = (field > 0).to(torch.uint8)
bitmasks = bitmasks[:-1, :, :] | (bitmasks[1:, :, :] << 1)
bitmasks = bitmasks[:, :-1, :] | (bitmasks[:, 1:, :] << 2)
bitmasks = bitmasks[:, :, :-1] | (bitmasks[:, :, 1:] << 4)
# Compute corner coordinates across the entire grid.
corner_coords = torch.empty(*grid_size, 3, device=dev, dtype=field.dtype)
corner_coords[range(grid_size[0]), :, :, 0] = torch.arange(grid_size[0], device=dev, dtype=field.dtype)[
:, None, None
]
corner_coords[:, range(grid_size[1]), :, 1] = torch.arange(grid_size[1], device=dev, dtype=field.dtype)[
:, None
]
corner_coords[:, :, range(grid_size[2]), 2] = torch.arange(grid_size[2], device=dev, dtype=field.dtype)
# Compute all vertices across all edges in the grid, even though we will
# throw some out later. We have (X-1)*Y*Z + X*(Y-1)*Z + X*Y*(Z-1) vertices.
# These are all midpoints, and don't account for interpolation (which is
# done later based on the used edge midpoints).
edge_midpoints = torch.cat(
[
((corner_coords[:-1] + corner_coords[1:]) / 2).reshape(-1, 3),
((corner_coords[:, :-1] + corner_coords[:, 1:]) / 2).reshape(-1, 3),
((corner_coords[:, :, :-1] + corner_coords[:, :, 1:]) / 2).reshape(-1, 3),
],
dim=0,
)
# Create a flat array of [X, Y, Z] indices for each cube.
cube_indices = torch.zeros(
grid_size[0] - 1, grid_size[1] - 1, grid_size[2] - 1, 3, device=dev, dtype=torch.long
)
cube_indices[range(grid_size[0] - 1), :, :, 0] = torch.arange(grid_size[0] - 1, device=dev)[:, None, None]
cube_indices[:, range(grid_size[1] - 1), :, 1] = torch.arange(grid_size[1] - 1, device=dev)[:, None]
cube_indices[:, :, range(grid_size[2] - 1), 2] = torch.arange(grid_size[2] - 1, device=dev)
flat_cube_indices = cube_indices.reshape(-1, 3)
# Create a flat array mapping each cube to 12 global edge indices.
edge_indices = _create_flat_edge_indices(flat_cube_indices, grid_size)
# Apply the LUT to figure out the triangles.
flat_bitmasks = bitmasks.reshape(-1).long() # must cast to long for indexing to believe this not a mask
local_tris = cases[flat_bitmasks]
local_masks = masks[flat_bitmasks]
# Compute the global edge indices for the triangles.
global_tris = torch.gather(edge_indices, 1, local_tris.reshape(local_tris.shape[0], -1)).reshape(
local_tris.shape
)
# Select the used triangles for each cube.
selected_tris = global_tris.reshape(-1, 3)[local_masks.reshape(-1)]
# Now we have a bunch of indices into the full list of possible vertices,
# but we want to reduce this list to only the used vertices.
used_vertex_indices = torch.unique(selected_tris.view(-1))
used_edge_midpoints = edge_midpoints[used_vertex_indices]
old_index_to_new_index = torch.zeros(len(edge_midpoints), device=dev, dtype=torch.long)
old_index_to_new_index[used_vertex_indices] = torch.arange(
len(used_vertex_indices), device=dev, dtype=torch.long
)
# Rewrite the triangles to use the new indices
faces = torch.gather(old_index_to_new_index, 0, selected_tris.view(-1)).reshape(selected_tris.shape)
# Compute the actual interpolated coordinates corresponding to edge midpoints.
v1 = torch.floor(used_edge_midpoints).to(torch.long)
v2 = torch.ceil(used_edge_midpoints).to(torch.long)
s1 = field[v1[:, 0], v1[:, 1], v1[:, 2]]
s2 = field[v2[:, 0], v2[:, 1], v2[:, 2]]
p1 = (v1.float() / (grid_size_tensor - 1)) * size + min_point
p2 = (v2.float() / (grid_size_tensor - 1)) * size + min_point
# The signs of s1 and s2 should be different. We want to find
# t such that t*s2 + (1-t)*s1 = 0.
t = (s1 / (s1 - s2))[:, None]
verts = t * p2 + (1 - t) * p1
return MeshDecoderOutput(verts=verts, faces=faces, vertex_channels=None)
@dataclass
class MLPNeRFModelOutput(BaseOutput):
density: torch.Tensor
signed_distance: torch.Tensor
channels: torch.Tensor
ts: torch.Tensor
class MLPNeRSTFModel(ModelMixin, ConfigMixin):
@register_to_config
def __init__(
self,
d_hidden: int = 256,
n_output: int = 12,
n_hidden_layers: int = 6,
act_fn: str = "swish",
insert_direction_at: int = 4,
):
super().__init__()
# Instantiate the MLP
# Find out the dimension of encoded position and direction
dummy = torch.eye(1, 3)
d_posenc_pos = encode_position(position=dummy).shape[-1]
d_posenc_dir = encode_direction(position=dummy).shape[-1]
mlp_widths = [d_hidden] * n_hidden_layers
input_widths = [d_posenc_pos] + mlp_widths
output_widths = mlp_widths + [n_output]
if insert_direction_at is not None:
input_widths[insert_direction_at] += d_posenc_dir
self.mlp = nn.ModuleList([nn.Linear(d_in, d_out) for d_in, d_out in zip(input_widths, output_widths)])
if act_fn == "swish":
# self.activation = swish
# yiyi testing:
self.activation = lambda x: F.silu(x)
else:
raise ValueError(f"Unsupported activation function {act_fn}")
self.sdf_activation = torch.tanh
self.density_activation = torch.nn.functional.relu
self.channel_activation = torch.sigmoid
def map_indices_to_keys(self, output):
h_map = {
"sdf": (0, 1),
"density_coarse": (1, 2),
"density_fine": (2, 3),
"stf": (3, 6),
"nerf_coarse": (6, 9),
"nerf_fine": (9, 12),
}
mapped_output = {k: output[..., start:end] for k, (start, end) in h_map.items()}
return mapped_output
def forward(self, *, position, direction, ts, nerf_level="coarse", rendering_mode="nerf"):
h = encode_position(position)
h_preact = h
h_directionless = None
for i, layer in enumerate(self.mlp):
if i == self.config.insert_direction_at: # 4 in the config
h_directionless = h_preact
h_direction = encode_direction(position, direction=direction)
h = torch.cat([h, h_direction], dim=-1)
h = layer(h)
h_preact = h
if i < len(self.mlp) - 1:
h = self.activation(h)
h_final = h
if h_directionless is None:
h_directionless = h_preact
activation = self.map_indices_to_keys(h_final)
if nerf_level == "coarse":
h_density = activation["density_coarse"]
else:
h_density = activation["density_fine"]
if rendering_mode == "nerf":
if nerf_level == "coarse":
h_channels = activation["nerf_coarse"]
else:
h_channels = activation["nerf_fine"]
elif rendering_mode == "stf":
h_channels = activation["stf"]
density = self.density_activation(h_density)
signed_distance = self.sdf_activation(activation["sdf"])
channels = self.channel_activation(h_channels)
# yiyi notes: I think signed_distance is not used
return MLPNeRFModelOutput(density=density, signed_distance=signed_distance, channels=channels, ts=ts)
class ChannelsProj(nn.Module):
def __init__(
self,
*,
vectors: int,
channels: int,
d_latent: int,
):
super().__init__()
self.proj = nn.Linear(d_latent, vectors * channels)
self.norm = nn.LayerNorm(channels)
self.d_latent = d_latent
self.vectors = vectors
self.channels = channels
def forward(self, x: torch.Tensor) -> torch.Tensor:
x_bvd = x
w_vcd = self.proj.weight.view(self.vectors, self.channels, self.d_latent)
b_vc = self.proj.bias.view(1, self.vectors, self.channels)
h = torch.einsum("bvd,vcd->bvc", x_bvd, w_vcd)
h = self.norm(h)
h = h + b_vc
return h
class ShapEParamsProjModel(ModelMixin, ConfigMixin):
"""
project the latent representation of a 3D asset to obtain weights of a multi-layer perceptron (MLP).
For more details, see the original paper:
"""
@register_to_config
def __init__(
self,
*,
param_names: Tuple[str] = (
"nerstf.mlp.0.weight",
"nerstf.mlp.1.weight",
"nerstf.mlp.2.weight",
"nerstf.mlp.3.weight",
),
param_shapes: Tuple[Tuple[int]] = (
(256, 93),
(256, 256),
(256, 256),
(256, 256),
),
d_latent: int = 1024,
):
super().__init__()
# check inputs
if len(param_names) != len(param_shapes):
raise ValueError("Must provide same number of `param_names` as `param_shapes`")
self.projections = nn.ModuleDict({})
for k, (vectors, channels) in zip(param_names, param_shapes):
self.projections[_sanitize_name(k)] = ChannelsProj(
vectors=vectors,
channels=channels,
d_latent=d_latent,
)
def forward(self, x: torch.Tensor):
out = {}
start = 0
for k, shape in zip(self.config.param_names, self.config.param_shapes):
vectors, _ = shape
end = start + vectors
x_bvd = x[:, start:end]
out[k] = self.projections[_sanitize_name(k)](x_bvd).reshape(len(x), *shape)
start = end
return out
class ShapERenderer(ModelMixin, ConfigMixin):
@register_to_config
def __init__(
self,
*,
param_names: Tuple[str] = (
"nerstf.mlp.0.weight",
"nerstf.mlp.1.weight",
"nerstf.mlp.2.weight",
"nerstf.mlp.3.weight",
),
param_shapes: Tuple[Tuple[int]] = (
(256, 93),
(256, 256),
(256, 256),
(256, 256),
),
d_latent: int = 1024,
d_hidden: int = 256,
n_output: int = 12,
n_hidden_layers: int = 6,
act_fn: str = "swish",
insert_direction_at: int = 4,
background: Tuple[float] = (
255.0,
255.0,
255.0,
),
):
super().__init__()
self.params_proj = ShapEParamsProjModel(
param_names=param_names,
param_shapes=param_shapes,
d_latent=d_latent,
)
self.mlp = MLPNeRSTFModel(d_hidden, n_output, n_hidden_layers, act_fn, insert_direction_at)
self.void = VoidNeRFModel(background=background, channel_scale=255.0)
self.volume = BoundingBoxVolume(bbox_max=[1.0, 1.0, 1.0], bbox_min=[-1.0, -1.0, -1.0])
self.mesh_decoder = MeshDecoder()
@torch.no_grad()
def render_rays(self, rays, sampler, n_samples, prev_model_out=None, render_with_direction=False):
"""
Perform volumetric rendering over a partition of possible t's in the union of rendering volumes (written below
with some abuse of notations)
C(r) := sum(
transmittance(t[i]) * integrate(
lambda t: density(t) * channels(t) * transmittance(t), [t[i], t[i + 1]],
) for i in range(len(parts))
) + transmittance(t[-1]) * void_model(t[-1]).channels
where
1) transmittance(s) := exp(-integrate(density, [t[0], s])) calculates the probability of light passing through
the volume specified by [t[0], s]. (transmittance of 1 means light can pass freely) 2) density and channels are
obtained by evaluating the appropriate part.model at time t. 3) [t[i], t[i + 1]] is defined as the range of t
where the ray intersects (parts[i].volume \\ union(part.volume for part in parts[:i])) at the surface of the
shell (if bounded). If the ray does not intersect, the integral over this segment is evaluated as 0 and
transmittance(t[i + 1]) := transmittance(t[i]). 4) The last term is integration to infinity (e.g. [t[-1],
math.inf]) that is evaluated by the void_model (i.e. we consider this space to be empty).
args:
rays: [batch_size x ... x 2 x 3] origin and direction. sampler: disjoint volume integrals. n_samples:
number of ts to sample. prev_model_outputs: model outputs from the previous rendering step, including
:return: A tuple of
- `channels`
- A importance samplers for additional fine-grained rendering
- raw model output
"""
origin, direction = rays[..., 0, :], rays[..., 1, :]
# Integrate over [t[i], t[i + 1]]
# 1 Intersect the rays with the current volume and sample ts to integrate along.
vrange = self.volume.intersect(origin, direction, t0_lower=None)
ts = sampler.sample(vrange.t0, vrange.t1, n_samples)
ts = ts.to(rays.dtype)
if prev_model_out is not None:
# Append the previous ts now before fprop because previous
# rendering used a different model and we can't reuse the output.
ts = torch.sort(torch.cat([ts, prev_model_out.ts], dim=-2), dim=-2).values
batch_size, *_shape, _t0_dim = vrange.t0.shape
_, *ts_shape, _ts_dim = ts.shape
# 2. Get the points along the ray and query the model
directions = torch.broadcast_to(direction.unsqueeze(-2), [batch_size, *ts_shape, 3])
positions = origin.unsqueeze(-2) + ts * directions
directions = directions.to(self.mlp.dtype)
positions = positions.to(self.mlp.dtype)
optional_directions = directions if render_with_direction else None
model_out = self.mlp(
position=positions,
direction=optional_directions,
ts=ts,
nerf_level="coarse" if prev_model_out is None else "fine",
)
# 3. Integrate the model results
channels, weights, transmittance = integrate_samples(
vrange, model_out.ts, model_out.density, model_out.channels
)
# 4. Clean up results that do not intersect with the volume.
transmittance = torch.where(vrange.intersected, transmittance, torch.ones_like(transmittance))
channels = torch.where(vrange.intersected, channels, torch.zeros_like(channels))
# 5. integration to infinity (e.g. [t[-1], math.inf]) that is evaluated by the void_model (i.e. we consider this space to be empty).
channels = channels + transmittance * self.void(origin)
weighted_sampler = ImportanceRaySampler(vrange, ts=model_out.ts, weights=weights)
return channels, weighted_sampler, model_out
@torch.no_grad()
def decode_to_image(
self,
latents,
device,
size: int = 64,
ray_batch_size: int = 4096,
n_coarse_samples=64,
n_fine_samples=128,
):
# project the parameters from the generated latents
projected_params = self.params_proj(latents)
# update the mlp layers of the renderer
for name, param in self.mlp.state_dict().items():
if f"nerstf.{name}" in projected_params.keys():
param.copy_(projected_params[f"nerstf.{name}"].squeeze(0))
# create cameras object
camera = create_pan_cameras(size)
rays = camera.camera_rays
rays = rays.to(device)
n_batches = rays.shape[1] // ray_batch_size
coarse_sampler = StratifiedRaySampler()
images = []
for idx in range(n_batches):
rays_batch = rays[:, idx * ray_batch_size : (idx + 1) * ray_batch_size]
# render rays with coarse, stratified samples.
_, fine_sampler, coarse_model_out = self.render_rays(rays_batch, coarse_sampler, n_coarse_samples)
# Then, render with additional importance-weighted ray samples.
channels, _, _ = self.render_rays(
rays_batch, fine_sampler, n_fine_samples, prev_model_out=coarse_model_out
)
images.append(channels)
images = torch.cat(images, dim=1)
images = images.view(*camera.shape, camera.height, camera.width, -1).squeeze(0)
return images
@torch.no_grad()
def decode_to_mesh(
self,
latents,
device,
grid_size: int = 128,
query_batch_size: int = 4096,
texture_channels: Tuple = ("R", "G", "B"),
):
# 1. project the parameters from the generated latents
projected_params = self.params_proj(latents)
# 2. update the mlp layers of the renderer
for name, param in self.mlp.state_dict().items():
if f"nerstf.{name}" in projected_params.keys():
param.copy_(projected_params[f"nerstf.{name}"].squeeze(0))
# 3. decoding with STF rendering
# 3.1 query the SDF values at vertices along a regular 128**3 grid
query_points = volume_query_points(self.volume, grid_size)
query_positions = query_points[None].repeat(1, 1, 1).to(device=device, dtype=self.mlp.dtype)
fields = []
for idx in range(0, query_positions.shape[1], query_batch_size):
query_batch = query_positions[:, idx : idx + query_batch_size]
model_out = self.mlp(
position=query_batch, direction=None, ts=None, nerf_level="fine", rendering_mode="stf"
)
fields.append(model_out.signed_distance)
# predicted SDF values
fields = torch.cat(fields, dim=1)
fields = fields.float()
assert (
len(fields.shape) == 3 and fields.shape[-1] == 1
), f"expected [meta_batch x inner_batch] SDF results, but got {fields.shape}"
fields = fields.reshape(1, *([grid_size] * 3))
# create grid 128 x 128 x 128
# - force a negative border around the SDFs to close off all the models.
full_grid = torch.zeros(
1,
grid_size + 2,
grid_size + 2,
grid_size + 2,
device=fields.device,
dtype=fields.dtype,
)
full_grid.fill_(-1.0)
full_grid[:, 1:-1, 1:-1, 1:-1] = fields
fields = full_grid
# apply a differentiable implementation of Marching Cubes to construct meshs
raw_meshes = []
mesh_mask = []
for field in fields:
raw_mesh = self.mesh_decoder(field, self.volume.bbox_min, self.volume.bbox_max - self.volume.bbox_min)
mesh_mask.append(True)
raw_meshes.append(raw_mesh)
mesh_mask = torch.tensor(mesh_mask, device=fields.device)
max_vertices = max(len(m.verts) for m in raw_meshes)
# 3.2. query the texture color head at each vertex of the resulting mesh.
texture_query_positions = torch.stack(
[m.verts[torch.arange(0, max_vertices) % len(m.verts)] for m in raw_meshes],
dim=0,
)
texture_query_positions = texture_query_positions.to(device=device, dtype=self.mlp.dtype)
textures = []
for idx in range(0, texture_query_positions.shape[1], query_batch_size):
query_batch = texture_query_positions[:, idx : idx + query_batch_size]
texture_model_out = self.mlp(
position=query_batch, direction=None, ts=None, nerf_level="fine", rendering_mode="stf"
)
textures.append(texture_model_out.channels)
# predict texture color
textures = torch.cat(textures, dim=1)
textures = _convert_srgb_to_linear(textures)
textures = textures.float()
# 3.3 augument the mesh with texture data
assert len(textures.shape) == 3 and textures.shape[-1] == len(
texture_channels
), f"expected [meta_batch x inner_batch x texture_channels] field results, but got {textures.shape}"
for m, texture in zip(raw_meshes, textures):
texture = texture[: len(m.verts)]
m.vertex_channels = dict(zip(texture_channels, texture.unbind(-1)))
return raw_meshes[0]
| diffusers/src/diffusers/pipelines/shap_e/renderer.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/shap_e/renderer.py",
"repo_id": "diffusers",
"token_count": 18167
} | 120 |
from dataclasses import dataclass
from typing import List, Union
import numpy as np
import PIL.Image
from ...utils import BaseOutput, is_flax_available
@dataclass
class StableDiffusionXLPipelineOutput(BaseOutput):
"""
Output class for Stable Diffusion pipelines.
Args:
images (`List[PIL.Image.Image]` or `np.ndarray`)
List of denoised PIL images of length `batch_size` or numpy array of shape `(batch_size, height, width,
num_channels)`. PIL images or numpy array present the denoised images of the diffusion pipeline.
"""
images: Union[List[PIL.Image.Image], np.ndarray]
if is_flax_available():
import flax
@flax.struct.dataclass
class FlaxStableDiffusionXLPipelineOutput(BaseOutput):
"""
Output class for Flax Stable Diffusion XL pipelines.
Args:
images (`np.ndarray`)
Array of shape `(batch_size, height, width, num_channels)` with images from the diffusion pipeline.
"""
images: np.ndarray
| diffusers/src/diffusers/pipelines/stable_diffusion_xl/pipeline_output.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/stable_diffusion_xl/pipeline_output.py",
"repo_id": "diffusers",
"token_count": 401
} | 121 |
import copy
import inspect
from dataclasses import dataclass
from typing import Any, Callable, Dict, List, Optional, Tuple, Union
import numpy as np
import PIL
import torch
import torch.nn.functional as F
from torch.nn.functional import grid_sample
from transformers import (
CLIPImageProcessor,
CLIPTextModel,
CLIPTextModelWithProjection,
CLIPTokenizer,
CLIPVisionModelWithProjection,
)
from ...image_processor import VaeImageProcessor
from ...loaders import StableDiffusionXLLoraLoaderMixin, TextualInversionLoaderMixin
from ...models import AutoencoderKL, UNet2DConditionModel
from ...models.attention_processor import (
AttnProcessor2_0,
FusedAttnProcessor2_0,
LoRAAttnProcessor2_0,
LoRAXFormersAttnProcessor,
XFormersAttnProcessor,
)
from ...models.lora import adjust_lora_scale_text_encoder
from ...schedulers import KarrasDiffusionSchedulers
from ...utils import (
USE_PEFT_BACKEND,
BaseOutput,
is_invisible_watermark_available,
logging,
scale_lora_layers,
unscale_lora_layers,
)
from ...utils.torch_utils import randn_tensor
from ..pipeline_utils import DiffusionPipeline
if is_invisible_watermark_available():
from ..stable_diffusion_xl.watermark import StableDiffusionXLWatermarker
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.rearrange_0
def rearrange_0(tensor, f):
F, C, H, W = tensor.size()
tensor = torch.permute(torch.reshape(tensor, (F // f, f, C, H, W)), (0, 2, 1, 3, 4))
return tensor
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.rearrange_1
def rearrange_1(tensor):
B, C, F, H, W = tensor.size()
return torch.reshape(torch.permute(tensor, (0, 2, 1, 3, 4)), (B * F, C, H, W))
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.rearrange_3
def rearrange_3(tensor, f):
F, D, C = tensor.size()
return torch.reshape(tensor, (F // f, f, D, C))
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.rearrange_4
def rearrange_4(tensor):
B, F, D, C = tensor.size()
return torch.reshape(tensor, (B * F, D, C))
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.CrossFrameAttnProcessor
class CrossFrameAttnProcessor:
"""
Cross frame attention processor. Each frame attends the first frame.
Args:
batch_size: The number that represents actual batch size, other than the frames.
For example, calling unet with a single prompt and num_images_per_prompt=1, batch_size should be equal to
2, due to classifier-free guidance.
"""
def __init__(self, batch_size=2):
self.batch_size = batch_size
def __call__(self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None):
batch_size, sequence_length, _ = hidden_states.shape
attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
query = attn.to_q(hidden_states)
is_cross_attention = encoder_hidden_states is not None
if encoder_hidden_states is None:
encoder_hidden_states = hidden_states
elif attn.norm_cross:
encoder_hidden_states = attn.norm_encoder_hidden_states(encoder_hidden_states)
key = attn.to_k(encoder_hidden_states)
value = attn.to_v(encoder_hidden_states)
# Cross Frame Attention
if not is_cross_attention:
video_length = key.size()[0] // self.batch_size
first_frame_index = [0] * video_length
# rearrange keys to have batch and frames in the 1st and 2nd dims respectively
key = rearrange_3(key, video_length)
key = key[:, first_frame_index]
# rearrange values to have batch and frames in the 1st and 2nd dims respectively
value = rearrange_3(value, video_length)
value = value[:, first_frame_index]
# rearrange back to original shape
key = rearrange_4(key)
value = rearrange_4(value)
query = attn.head_to_batch_dim(query)
key = attn.head_to_batch_dim(key)
value = attn.head_to_batch_dim(value)
attention_probs = attn.get_attention_scores(query, key, attention_mask)
hidden_states = torch.bmm(attention_probs, value)
hidden_states = attn.batch_to_head_dim(hidden_states)
# linear proj
hidden_states = attn.to_out[0](hidden_states)
# dropout
hidden_states = attn.to_out[1](hidden_states)
return hidden_states
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.CrossFrameAttnProcessor2_0
class CrossFrameAttnProcessor2_0:
"""
Cross frame attention processor with scaled_dot_product attention of Pytorch 2.0.
Args:
batch_size: The number that represents actual batch size, other than the frames.
For example, calling unet with a single prompt and num_images_per_prompt=1, batch_size should be equal to
2, due to classifier-free guidance.
"""
def __init__(self, batch_size=2):
if not hasattr(F, "scaled_dot_product_attention"):
raise ImportError("AttnProcessor2_0 requires PyTorch 2.0, to use it, please upgrade PyTorch to 2.0.")
self.batch_size = batch_size
def __call__(self, attn, hidden_states, encoder_hidden_states=None, attention_mask=None):
batch_size, sequence_length, _ = (
hidden_states.shape if encoder_hidden_states is None else encoder_hidden_states.shape
)
inner_dim = hidden_states.shape[-1]
if attention_mask is not None:
attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
# scaled_dot_product_attention expects attention_mask shape to be
# (batch, heads, source_length, target_length)
attention_mask = attention_mask.view(batch_size, attn.heads, -1, attention_mask.shape[-1])
query = attn.to_q(hidden_states)
is_cross_attention = encoder_hidden_states is not None
if encoder_hidden_states is None:
encoder_hidden_states = hidden_states
elif attn.norm_cross:
encoder_hidden_states = attn.norm_encoder_hidden_states(encoder_hidden_states)
key = attn.to_k(encoder_hidden_states)
value = attn.to_v(encoder_hidden_states)
# Cross Frame Attention
if not is_cross_attention:
video_length = max(1, key.size()[0] // self.batch_size)
first_frame_index = [0] * video_length
# rearrange keys to have batch and frames in the 1st and 2nd dims respectively
key = rearrange_3(key, video_length)
key = key[:, first_frame_index]
# rearrange values to have batch and frames in the 1st and 2nd dims respectively
value = rearrange_3(value, video_length)
value = value[:, first_frame_index]
# rearrange back to original shape
key = rearrange_4(key)
value = rearrange_4(value)
head_dim = inner_dim // attn.heads
query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
# the output of sdp = (batch, num_heads, seq_len, head_dim)
# TODO: add support for attn.scale when we move to Torch 2.1
hidden_states = F.scaled_dot_product_attention(
query, key, value, attn_mask=attention_mask, dropout_p=0.0, is_causal=False
)
hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)
hidden_states = hidden_states.to(query.dtype)
# linear proj
hidden_states = attn.to_out[0](hidden_states)
# dropout
hidden_states = attn.to_out[1](hidden_states)
return hidden_states
@dataclass
class TextToVideoSDXLPipelineOutput(BaseOutput):
"""
Output class for zero-shot text-to-video pipeline.
Args:
images (`List[PIL.Image.Image]` or `np.ndarray`)
List of denoised PIL images of length `batch_size` or numpy array of shape `(batch_size, height, width,
num_channels)`. PIL images or numpy array present the denoised images of the diffusion pipeline.
"""
images: Union[List[PIL.Image.Image], np.ndarray]
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.coords_grid
def coords_grid(batch, ht, wd, device):
# Adapted from https://github.com/princeton-vl/RAFT/blob/master/core/utils/utils.py
coords = torch.meshgrid(torch.arange(ht, device=device), torch.arange(wd, device=device))
coords = torch.stack(coords[::-1], dim=0).float()
return coords[None].repeat(batch, 1, 1, 1)
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.warp_single_latent
def warp_single_latent(latent, reference_flow):
"""
Warp latent of a single frame with given flow
Args:
latent: latent code of a single frame
reference_flow: flow which to warp the latent with
Returns:
warped: warped latent
"""
_, _, H, W = reference_flow.size()
_, _, h, w = latent.size()
coords0 = coords_grid(1, H, W, device=latent.device).to(latent.dtype)
coords_t0 = coords0 + reference_flow
coords_t0[:, 0] /= W
coords_t0[:, 1] /= H
coords_t0 = coords_t0 * 2.0 - 1.0
coords_t0 = F.interpolate(coords_t0, size=(h, w), mode="bilinear")
coords_t0 = torch.permute(coords_t0, (0, 2, 3, 1))
warped = grid_sample(latent, coords_t0, mode="nearest", padding_mode="reflection")
return warped
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.create_motion_field
def create_motion_field(motion_field_strength_x, motion_field_strength_y, frame_ids, device, dtype):
"""
Create translation motion field
Args:
motion_field_strength_x: motion strength along x-axis
motion_field_strength_y: motion strength along y-axis
frame_ids: indexes of the frames the latents of which are being processed.
This is needed when we perform chunk-by-chunk inference
device: device
dtype: dtype
Returns:
"""
seq_length = len(frame_ids)
reference_flow = torch.zeros((seq_length, 2, 512, 512), device=device, dtype=dtype)
for fr_idx in range(seq_length):
reference_flow[fr_idx, 0, :, :] = motion_field_strength_x * (frame_ids[fr_idx])
reference_flow[fr_idx, 1, :, :] = motion_field_strength_y * (frame_ids[fr_idx])
return reference_flow
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.create_motion_field_and_warp_latents
def create_motion_field_and_warp_latents(motion_field_strength_x, motion_field_strength_y, frame_ids, latents):
"""
Creates translation motion and warps the latents accordingly
Args:
motion_field_strength_x: motion strength along x-axis
motion_field_strength_y: motion strength along y-axis
frame_ids: indexes of the frames the latents of which are being processed.
This is needed when we perform chunk-by-chunk inference
latents: latent codes of frames
Returns:
warped_latents: warped latents
"""
motion_field = create_motion_field(
motion_field_strength_x=motion_field_strength_x,
motion_field_strength_y=motion_field_strength_y,
frame_ids=frame_ids,
device=latents.device,
dtype=latents.dtype,
)
warped_latents = latents.clone().detach()
for i in range(len(warped_latents)):
warped_latents[i] = warp_single_latent(latents[i][None], motion_field[i][None])
return warped_latents
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.rescale_noise_cfg
def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
"""
Rescale `noise_cfg` according to `guidance_rescale`. Based on findings of [Common Diffusion Noise Schedules and
Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf). See Section 3.4
"""
std_text = noise_pred_text.std(dim=list(range(1, noise_pred_text.ndim)), keepdim=True)
std_cfg = noise_cfg.std(dim=list(range(1, noise_cfg.ndim)), keepdim=True)
# rescale the results from guidance (fixes overexposure)
noise_pred_rescaled = noise_cfg * (std_text / std_cfg)
# mix with the original results from guidance by factor guidance_rescale to avoid "plain looking" images
noise_cfg = guidance_rescale * noise_pred_rescaled + (1 - guidance_rescale) * noise_cfg
return noise_cfg
class TextToVideoZeroSDXLPipeline(
DiffusionPipeline,
StableDiffusionXLLoraLoaderMixin,
TextualInversionLoaderMixin,
):
r"""
Pipeline for zero-shot text-to-video generation using Stable Diffusion XL.
This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
implemented for all pipelines (downloading, saving, running on a particular device, etc.).
Args:
vae ([`AutoencoderKL`]):
Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations.
text_encoder ([`CLIPTextModel`]):
Frozen text-encoder. Stable Diffusion XL uses the text portion of
[CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModel), specifically
the [clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14) variant.
text_encoder_2 ([` CLIPTextModelWithProjection`]):
Second frozen text-encoder. Stable Diffusion XL uses the text and pool portion of
[CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModelWithProjection),
specifically the
[laion/CLIP-ViT-bigG-14-laion2B-39B-b160k](https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k)
variant.
tokenizer (`CLIPTokenizer`):
Tokenizer of class
[CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
tokenizer_2 (`CLIPTokenizer`):
Second Tokenizer of class
[CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
unet ([`UNet2DConditionModel`]): Conditional U-Net architecture to denoise the encoded image latents.
scheduler ([`SchedulerMixin`]):
A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
[`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`].
"""
model_cpu_offload_seq = "text_encoder->text_encoder_2->unet->vae"
_optional_components = [
"tokenizer",
"tokenizer_2",
"text_encoder",
"text_encoder_2",
"image_encoder",
"feature_extractor",
]
def __init__(
self,
vae: AutoencoderKL,
text_encoder: CLIPTextModel,
text_encoder_2: CLIPTextModelWithProjection,
tokenizer: CLIPTokenizer,
tokenizer_2: CLIPTokenizer,
unet: UNet2DConditionModel,
scheduler: KarrasDiffusionSchedulers,
image_encoder: CLIPVisionModelWithProjection = None,
feature_extractor: CLIPImageProcessor = None,
force_zeros_for_empty_prompt: bool = True,
add_watermarker: Optional[bool] = None,
):
super().__init__()
self.register_modules(
vae=vae,
text_encoder=text_encoder,
text_encoder_2=text_encoder_2,
tokenizer=tokenizer,
tokenizer_2=tokenizer_2,
unet=unet,
scheduler=scheduler,
image_encoder=image_encoder,
feature_extractor=feature_extractor,
)
self.register_to_config(force_zeros_for_empty_prompt=force_zeros_for_empty_prompt)
self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1)
self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor)
self.default_sample_size = self.unet.config.sample_size
add_watermarker = add_watermarker if add_watermarker is not None else is_invisible_watermark_available()
if add_watermarker:
self.watermark = StableDiffusionXLWatermarker()
else:
self.watermark = None
processor = (
CrossFrameAttnProcessor2_0(batch_size=2)
if hasattr(F, "scaled_dot_product_attention")
else CrossFrameAttnProcessor(batch_size=2)
)
self.unet.set_attn_processor(processor)
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
def prepare_extra_step_kwargs(self, generator, eta):
# prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
# eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
# eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
# and should be between [0, 1]
accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
extra_step_kwargs = {}
if accepts_eta:
extra_step_kwargs["eta"] = eta
# check if the scheduler accepts generator
accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
if accepts_generator:
extra_step_kwargs["generator"] = generator
return extra_step_kwargs
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.enable_vae_slicing
def enable_vae_slicing(self):
r"""
Enable sliced VAE decoding. When this option is enabled, the VAE will split the input tensor in slices to
compute decoding in several steps. This is useful to save some memory and allow larger batch sizes.
"""
self.vae.enable_slicing()
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.disable_vae_slicing
def disable_vae_slicing(self):
r"""
Disable sliced VAE decoding. If `enable_vae_slicing` was previously enabled, this method will go back to
computing decoding in one step.
"""
self.vae.disable_slicing()
# Copied from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl.StableDiffusionXLPipeline.upcast_vae
def upcast_vae(self):
dtype = self.vae.dtype
self.vae.to(dtype=torch.float32)
use_torch_2_0_or_xformers = isinstance(
self.vae.decoder.mid_block.attentions[0].processor,
(
AttnProcessor2_0,
XFormersAttnProcessor,
LoRAXFormersAttnProcessor,
LoRAAttnProcessor2_0,
FusedAttnProcessor2_0,
),
)
# if xformers or torch_2_0 is used attention block does not need
# to be in float32 which can save lots of memory
if use_torch_2_0_or_xformers:
self.vae.post_quant_conv.to(dtype)
self.vae.decoder.conv_in.to(dtype)
self.vae.decoder.mid_block.to(dtype)
# Copied from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl.StableDiffusionXLPipeline._get_add_time_ids
def _get_add_time_ids(
self, original_size, crops_coords_top_left, target_size, dtype, text_encoder_projection_dim=None
):
add_time_ids = list(original_size + crops_coords_top_left + target_size)
passed_add_embed_dim = (
self.unet.config.addition_time_embed_dim * len(add_time_ids) + text_encoder_projection_dim
)
expected_add_embed_dim = self.unet.add_embedding.linear_1.in_features
if expected_add_embed_dim != passed_add_embed_dim:
raise ValueError(
f"Model expects an added time embedding vector of length {expected_add_embed_dim}, but a vector of {passed_add_embed_dim} was created. The model has an incorrect config. Please check `unet.config.time_embedding_type` and `text_encoder_2.config.projection_dim`."
)
add_time_ids = torch.tensor([add_time_ids], dtype=dtype)
return add_time_ids
# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_latents
def prepare_latents(self, batch_size, num_channels_latents, height, width, dtype, device, generator, latents=None):
shape = (batch_size, num_channels_latents, height // self.vae_scale_factor, width // self.vae_scale_factor)
if isinstance(generator, list) and len(generator) != batch_size:
raise ValueError(
f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
f" size of {batch_size}. Make sure the batch size matches the length of the generators."
)
if latents is None:
latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
else:
latents = latents.to(device)
# scale the initial noise by the standard deviation required by the scheduler
latents = latents * self.scheduler.init_noise_sigma
return latents
# Copied from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl.StableDiffusionXLPipeline.check_inputs
def check_inputs(
self,
prompt,
prompt_2,
height,
width,
callback_steps,
negative_prompt=None,
negative_prompt_2=None,
prompt_embeds=None,
negative_prompt_embeds=None,
pooled_prompt_embeds=None,
negative_pooled_prompt_embeds=None,
callback_on_step_end_tensor_inputs=None,
):
if height % 8 != 0 or width % 8 != 0:
raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
raise ValueError(
f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
f" {type(callback_steps)}."
)
if callback_on_step_end_tensor_inputs is not None and not all(
k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
):
raise ValueError(
f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
)
if prompt is not None and prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
" only forward one of the two."
)
elif prompt_2 is not None and prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `prompt_2`: {prompt_2} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
" only forward one of the two."
)
elif prompt is None and prompt_embeds is None:
raise ValueError(
"Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
)
elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
elif prompt_2 is not None and (not isinstance(prompt_2, str) and not isinstance(prompt_2, list)):
raise ValueError(f"`prompt_2` has to be of type `str` or `list` but is {type(prompt_2)}")
if negative_prompt is not None and negative_prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
)
elif negative_prompt_2 is not None and negative_prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `negative_prompt_2`: {negative_prompt_2} and `negative_prompt_embeds`:"
f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
)
if prompt_embeds is not None and negative_prompt_embeds is not None:
if prompt_embeds.shape != negative_prompt_embeds.shape:
raise ValueError(
"`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
f" {negative_prompt_embeds.shape}."
)
if prompt_embeds is not None and pooled_prompt_embeds is None:
raise ValueError(
"If `prompt_embeds` are provided, `pooled_prompt_embeds` also have to be passed. Make sure to generate `pooled_prompt_embeds` from the same text encoder that was used to generate `prompt_embeds`."
)
if negative_prompt_embeds is not None and negative_pooled_prompt_embeds is None:
raise ValueError(
"If `negative_prompt_embeds` are provided, `negative_pooled_prompt_embeds` also have to be passed. Make sure to generate `negative_pooled_prompt_embeds` from the same text encoder that was used to generate `negative_prompt_embeds`."
)
# Copied from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl.StableDiffusionXLPipeline.encode_prompt
def encode_prompt(
self,
prompt: str,
prompt_2: Optional[str] = None,
device: Optional[torch.device] = None,
num_images_per_prompt: int = 1,
do_classifier_free_guidance: bool = True,
negative_prompt: Optional[str] = None,
negative_prompt_2: Optional[str] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
lora_scale: Optional[float] = None,
clip_skip: Optional[int] = None,
):
r"""
Encodes the prompt into text encoder hidden states.
Args:
prompt (`str` or `List[str]`, *optional*):
prompt to be encoded
prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders
device: (`torch.device`):
torch device
num_images_per_prompt (`int`):
number of images that should be generated per prompt
do_classifier_free_guidance (`bool`):
whether to use classifier free guidance or not
negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`).
negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
lora_scale (`float`, *optional*):
A lora scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
clip_skip (`int`, *optional*):
Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
the output of the pre-final layer will be used for computing the prompt embeddings.
"""
device = device or self._execution_device
# set lora scale so that monkey patched LoRA
# function of text encoder can correctly access it
if lora_scale is not None and isinstance(self, StableDiffusionXLLoraLoaderMixin):
self._lora_scale = lora_scale
# dynamically adjust the LoRA scale
if self.text_encoder is not None:
if not USE_PEFT_BACKEND:
adjust_lora_scale_text_encoder(self.text_encoder, lora_scale)
else:
scale_lora_layers(self.text_encoder, lora_scale)
if self.text_encoder_2 is not None:
if not USE_PEFT_BACKEND:
adjust_lora_scale_text_encoder(self.text_encoder_2, lora_scale)
else:
scale_lora_layers(self.text_encoder_2, lora_scale)
prompt = [prompt] if isinstance(prompt, str) else prompt
if prompt is not None:
batch_size = len(prompt)
else:
batch_size = prompt_embeds.shape[0]
# Define tokenizers and text encoders
tokenizers = [self.tokenizer, self.tokenizer_2] if self.tokenizer is not None else [self.tokenizer_2]
text_encoders = (
[self.text_encoder, self.text_encoder_2] if self.text_encoder is not None else [self.text_encoder_2]
)
if prompt_embeds is None:
prompt_2 = prompt_2 or prompt
prompt_2 = [prompt_2] if isinstance(prompt_2, str) else prompt_2
# textual inversion: procecss multi-vector tokens if necessary
prompt_embeds_list = []
prompts = [prompt, prompt_2]
for prompt, tokenizer, text_encoder in zip(prompts, tokenizers, text_encoders):
if isinstance(self, TextualInversionLoaderMixin):
prompt = self.maybe_convert_prompt(prompt, tokenizer)
text_inputs = tokenizer(
prompt,
padding="max_length",
max_length=tokenizer.model_max_length,
truncation=True,
return_tensors="pt",
)
text_input_ids = text_inputs.input_ids
untruncated_ids = tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
text_input_ids, untruncated_ids
):
removed_text = tokenizer.batch_decode(untruncated_ids[:, tokenizer.model_max_length - 1 : -1])
logger.warning(
"The following part of your input was truncated because CLIP can only handle sequences up to"
f" {tokenizer.model_max_length} tokens: {removed_text}"
)
prompt_embeds = text_encoder(text_input_ids.to(device), output_hidden_states=True)
# We are only ALWAYS interested in the pooled output of the final text encoder
pooled_prompt_embeds = prompt_embeds[0]
if clip_skip is None:
prompt_embeds = prompt_embeds.hidden_states[-2]
else:
# "2" because SDXL always indexes from the penultimate layer.
prompt_embeds = prompt_embeds.hidden_states[-(clip_skip + 2)]
prompt_embeds_list.append(prompt_embeds)
prompt_embeds = torch.concat(prompt_embeds_list, dim=-1)
# get unconditional embeddings for classifier free guidance
zero_out_negative_prompt = negative_prompt is None and self.config.force_zeros_for_empty_prompt
if do_classifier_free_guidance and negative_prompt_embeds is None and zero_out_negative_prompt:
negative_prompt_embeds = torch.zeros_like(prompt_embeds)
negative_pooled_prompt_embeds = torch.zeros_like(pooled_prompt_embeds)
elif do_classifier_free_guidance and negative_prompt_embeds is None:
negative_prompt = negative_prompt or ""
negative_prompt_2 = negative_prompt_2 or negative_prompt
# normalize str to list
negative_prompt = batch_size * [negative_prompt] if isinstance(negative_prompt, str) else negative_prompt
negative_prompt_2 = (
batch_size * [negative_prompt_2] if isinstance(negative_prompt_2, str) else negative_prompt_2
)
uncond_tokens: List[str]
if prompt is not None and type(prompt) is not type(negative_prompt):
raise TypeError(
f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
f" {type(prompt)}."
)
elif batch_size != len(negative_prompt):
raise ValueError(
f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
" the batch size of `prompt`."
)
else:
uncond_tokens = [negative_prompt, negative_prompt_2]
negative_prompt_embeds_list = []
for negative_prompt, tokenizer, text_encoder in zip(uncond_tokens, tokenizers, text_encoders):
if isinstance(self, TextualInversionLoaderMixin):
negative_prompt = self.maybe_convert_prompt(negative_prompt, tokenizer)
max_length = prompt_embeds.shape[1]
uncond_input = tokenizer(
negative_prompt,
padding="max_length",
max_length=max_length,
truncation=True,
return_tensors="pt",
)
negative_prompt_embeds = text_encoder(
uncond_input.input_ids.to(device),
output_hidden_states=True,
)
# We are only ALWAYS interested in the pooled output of the final text encoder
negative_pooled_prompt_embeds = negative_prompt_embeds[0]
negative_prompt_embeds = negative_prompt_embeds.hidden_states[-2]
negative_prompt_embeds_list.append(negative_prompt_embeds)
negative_prompt_embeds = torch.concat(negative_prompt_embeds_list, dim=-1)
if self.text_encoder_2 is not None:
prompt_embeds = prompt_embeds.to(dtype=self.text_encoder_2.dtype, device=device)
else:
prompt_embeds = prompt_embeds.to(dtype=self.unet.dtype, device=device)
bs_embed, seq_len, _ = prompt_embeds.shape
# duplicate text embeddings for each generation per prompt, using mps friendly method
prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt, 1)
prompt_embeds = prompt_embeds.view(bs_embed * num_images_per_prompt, seq_len, -1)
if do_classifier_free_guidance:
# duplicate unconditional embeddings for each generation per prompt, using mps friendly method
seq_len = negative_prompt_embeds.shape[1]
if self.text_encoder_2 is not None:
negative_prompt_embeds = negative_prompt_embeds.to(dtype=self.text_encoder_2.dtype, device=device)
else:
negative_prompt_embeds = negative_prompt_embeds.to(dtype=self.unet.dtype, device=device)
negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt, 1)
negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
pooled_prompt_embeds = pooled_prompt_embeds.repeat(1, num_images_per_prompt).view(
bs_embed * num_images_per_prompt, -1
)
if do_classifier_free_guidance:
negative_pooled_prompt_embeds = negative_pooled_prompt_embeds.repeat(1, num_images_per_prompt).view(
bs_embed * num_images_per_prompt, -1
)
if self.text_encoder is not None:
if isinstance(self, StableDiffusionXLLoraLoaderMixin) and USE_PEFT_BACKEND:
# Retrieve the original scale by scaling back the LoRA layers
unscale_lora_layers(self.text_encoder, lora_scale)
if self.text_encoder_2 is not None:
if isinstance(self, StableDiffusionXLLoraLoaderMixin) and USE_PEFT_BACKEND:
# Retrieve the original scale by scaling back the LoRA layers
unscale_lora_layers(self.text_encoder_2, lora_scale)
return prompt_embeds, negative_prompt_embeds, pooled_prompt_embeds, negative_pooled_prompt_embeds
# Copied from diffusers.pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.TextToVideoZeroPipeline.forward_loop
def forward_loop(self, x_t0, t0, t1, generator):
"""
Perform DDPM forward process from time t0 to t1. This is the same as adding noise with corresponding variance.
Args:
x_t0:
Latent code at time t0.
t0:
Timestep at t0.
t1:
Timestamp at t1.
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
generation deterministic.
Returns:
x_t1:
Forward process applied to x_t0 from time t0 to t1.
"""
eps = randn_tensor(x_t0.size(), generator=generator, dtype=x_t0.dtype, device=x_t0.device)
alpha_vec = torch.prod(self.scheduler.alphas[t0:t1])
x_t1 = torch.sqrt(alpha_vec) * x_t0 + torch.sqrt(1 - alpha_vec) * eps
return x_t1
def backward_loop(
self,
latents,
timesteps,
prompt_embeds,
guidance_scale,
callback,
callback_steps,
num_warmup_steps,
extra_step_kwargs,
add_text_embeds,
add_time_ids,
cross_attention_kwargs=None,
guidance_rescale: float = 0.0,
):
"""
Perform backward process given list of time steps
Args:
latents:
Latents at time timesteps[0].
timesteps:
Time steps along which to perform backward process.
prompt_embeds:
Pre-generated text embeddings.
guidance_scale:
A higher guidance scale value encourages the model to generate images closely linked to the text
`prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
callback (`Callable`, *optional*):
A function that calls every `callback_steps` steps during inference. The function is called with the
following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function is called. If not specified, the callback is called at
every step.
extra_step_kwargs:
Extra_step_kwargs.
cross_attention_kwargs:
A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
[`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
num_warmup_steps:
number of warmup steps.
Returns:
latents: latents of backward process output at time timesteps[-1]
"""
do_classifier_free_guidance = guidance_scale > 1.0
num_steps = (len(timesteps) - num_warmup_steps) // self.scheduler.order
with self.progress_bar(total=num_steps) as progress_bar:
for i, t in enumerate(timesteps):
# expand the latents if we are doing classifier free guidance
latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
# predict the noise residual
added_cond_kwargs = {"text_embeds": add_text_embeds, "time_ids": add_time_ids}
noise_pred = self.unet(
latent_model_input,
t,
encoder_hidden_states=prompt_embeds,
cross_attention_kwargs=cross_attention_kwargs,
added_cond_kwargs=added_cond_kwargs,
return_dict=False,
)[0]
# perform guidance
if do_classifier_free_guidance:
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
if do_classifier_free_guidance and guidance_rescale > 0.0:
# Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=guidance_rescale)
# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs).prev_sample
# call the callback, if provided
if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
progress_bar.update()
if callback is not None and i % callback_steps == 0:
callback(i, t, latents)
return latents.clone().detach()
@torch.no_grad()
def __call__(
self,
prompt: Union[str, List[str]],
prompt_2: Optional[Union[str, List[str]]] = None,
video_length: Optional[int] = 8,
height: Optional[int] = None,
width: Optional[int] = None,
num_inference_steps: int = 50,
denoising_end: Optional[float] = None,
guidance_scale: float = 7.5,
negative_prompt: Optional[Union[str, List[str]]] = None,
negative_prompt_2: Optional[Union[str, List[str]]] = None,
num_videos_per_prompt: Optional[int] = 1,
eta: float = 0.0,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
frame_ids: Optional[List[int]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None,
latents: Optional[torch.FloatTensor] = None,
motion_field_strength_x: float = 12,
motion_field_strength_y: float = 12,
output_type: Optional[str] = "tensor",
return_dict: bool = True,
callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
callback_steps: int = 1,
cross_attention_kwargs: Optional[Dict[str, Any]] = None,
guidance_rescale: float = 0.0,
original_size: Optional[Tuple[int, int]] = None,
crops_coords_top_left: Tuple[int, int] = (0, 0),
target_size: Optional[Tuple[int, int]] = None,
t0: int = 44,
t1: int = 47,
):
"""
Function invoked when calling the pipeline for generation.
Args:
prompt (`str` or `List[str]`, *optional*):
The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
instead.
prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
used in both text-encoders
video_length (`int`, *optional*, defaults to 8):
The number of generated video frames.
height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
The height in pixels of the generated image.
width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
The width in pixels of the generated image.
num_inference_steps (`int`, *optional*, defaults to 50):
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.
denoising_end (`float`, *optional*):
When specified, determines the fraction (between 0.0 and 1.0) of the total denoising process to be
completed before it is intentionally prematurely terminated. As a result, the returned sample will
still retain a substantial amount of noise as determined by the discrete timesteps selected by the
scheduler. The denoising_end parameter should ideally be utilized when this pipeline forms a part of a
"Mixture of Denoisers" multi-pipeline setup, as elaborated in [**Refining the Image
Output**](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion/stable_diffusion_xl#refining-the-image-output)
guidance_scale (`float`, *optional*, defaults to 7.5):
Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
`guidance_scale` is defined as `w` of equation 2. of [Imagen
Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
usually at the expense of lower image quality.
negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation. If not defined, one has to pass
`negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
less than `1`).
negative_prompt_2 (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and
`text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders
num_videos_per_prompt (`int`, *optional*, defaults to 1):
The number of videos to generate per prompt.
eta (`float`, *optional*, defaults to 0.0):
Corresponds to parameter eta (η) in the DDIM paper: https://arxiv.org/abs/2010.02502. Only applies to
[`schedulers.DDIMScheduler`], will be ignored for others.
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic.
frame_ids (`List[int]`, *optional*):
Indexes of the frames that are being generated. This is used when generating longer videos
chunk-by-chunk.
prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
If not provided, pooled text embeddings will be generated from `prompt` input argument.
negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt`
input argument.
latents (`torch.FloatTensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`.
motion_field_strength_x (`float`, *optional*, defaults to 12):
Strength of motion in generated video along x-axis. See the [paper](https://arxiv.org/abs/2303.13439),
Sect. 3.3.1.
motion_field_strength_y (`float`, *optional*, defaults to 12):
Strength of motion in generated video along y-axis. See the [paper](https://arxiv.org/abs/2303.13439),
Sect. 3.3.1.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between
[PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.stable_diffusion_xl.StableDiffusionXLPipelineOutput`] instead
of a plain tuple.
callback (`Callable`, *optional*):
A function that will be called every `callback_steps` steps during inference. The function will be
called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
callback_steps (`int`, *optional*, defaults to 1):
The frequency at which the `callback` function will be called. If not specified, the callback will be
called at every step.
cross_attention_kwargs (`dict`, *optional*):
A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
`self.processor` in
[diffusers.cross_attention](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/cross_attention.py).
guidance_rescale (`float`, *optional*, defaults to 0.7):
Guidance rescale factor proposed by [Common Diffusion Noise Schedules and Sample Steps are
Flawed](https://arxiv.org/pdf/2305.08891.pdf) `guidance_scale` is defined as `φ` in equation 16. of
[Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf).
Guidance rescale factor should fix overexposure when using zero terminal SNR.
original_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)):
If `original_size` is not the same as `target_size` the image will appear to be down- or upsampled.
`original_size` defaults to `(width, height)` if not specified. Part of SDXL's micro-conditioning as
explained in section 2.2 of
[https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952).
crops_coords_top_left (`Tuple[int]`, *optional*, defaults to (0, 0)):
`crops_coords_top_left` can be used to generate an image that appears to be "cropped" from the position
`crops_coords_top_left` downwards. Favorable, well-centered images are usually achieved by setting
`crops_coords_top_left` to (0, 0). Part of SDXL's micro-conditioning as explained in section 2.2 of
[https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952).
target_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)):
For most cases, `target_size` should be set to the desired height and width of the generated image. If
not specified it will default to `(width, height)`. Part of SDXL's micro-conditioning as explained in
section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952).
t0 (`int`, *optional*, defaults to 44):
Timestep t0. Should be in the range [0, num_inference_steps - 1]. See the
[paper](https://arxiv.org/abs/2303.13439), Sect. 3.3.1.
t1 (`int`, *optional*, defaults to 47):
Timestep t0. Should be in the range [t0 + 1, num_inference_steps - 1]. See the
[paper](https://arxiv.org/abs/2303.13439), Sect. 3.3.1.
Returns:
[`~pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.TextToVideoSDXLPipelineOutput`] or
`tuple`: [`~pipelines.text_to_video_synthesis.pipeline_text_to_video_zero.TextToVideoSDXLPipelineOutput`]
if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is a list with the
generated images.
"""
assert video_length > 0
if frame_ids is None:
frame_ids = list(range(video_length))
assert len(frame_ids) == video_length
assert num_videos_per_prompt == 1
if isinstance(prompt, str):
prompt = [prompt]
if isinstance(negative_prompt, str):
negative_prompt = [negative_prompt]
# 0. Default height and width to unet
height = height or self.default_sample_size * self.vae_scale_factor
width = width or self.default_sample_size * self.vae_scale_factor
original_size = original_size or (height, width)
target_size = target_size or (height, width)
# 1. Check inputs. Raise error if not correct
self.check_inputs(
prompt,
prompt_2,
height,
width,
callback_steps,
negative_prompt,
negative_prompt_2,
prompt_embeds,
negative_prompt_embeds,
pooled_prompt_embeds,
negative_pooled_prompt_embeds,
)
# 2. Define call parameters
batch_size = (
1 if isinstance(prompt, str) else len(prompt) if isinstance(prompt, list) else prompt_embeds.shape[0]
)
device = self._execution_device
# here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
# of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
# corresponds to doing no classifier free guidance.
do_classifier_free_guidance = guidance_scale > 1.0
# 3. Encode input prompt
text_encoder_lora_scale = (
cross_attention_kwargs.get("scale", None) if cross_attention_kwargs is not None else None
)
(
prompt_embeds,
negative_prompt_embeds,
pooled_prompt_embeds,
negative_pooled_prompt_embeds,
) = self.encode_prompt(
prompt=prompt,
prompt_2=prompt_2,
device=device,
num_images_per_prompt=num_videos_per_prompt,
do_classifier_free_guidance=do_classifier_free_guidance,
negative_prompt=negative_prompt,
negative_prompt_2=negative_prompt_2,
prompt_embeds=prompt_embeds,
negative_prompt_embeds=negative_prompt_embeds,
pooled_prompt_embeds=pooled_prompt_embeds,
negative_pooled_prompt_embeds=negative_pooled_prompt_embeds,
lora_scale=text_encoder_lora_scale,
)
# 4. Prepare timesteps
self.scheduler.set_timesteps(num_inference_steps, device=device)
timesteps = self.scheduler.timesteps
# 5. Prepare latent variables
num_channels_latents = self.unet.config.in_channels
latents = self.prepare_latents(
batch_size * num_videos_per_prompt,
num_channels_latents,
height,
width,
prompt_embeds.dtype,
device,
generator,
latents,
)
# 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
# 7. Prepare added time ids & embeddings
add_text_embeds = pooled_prompt_embeds
if self.text_encoder_2 is None:
text_encoder_projection_dim = int(pooled_prompt_embeds.shape[-1])
else:
text_encoder_projection_dim = self.text_encoder_2.config.projection_dim
add_time_ids = self._get_add_time_ids(
original_size,
crops_coords_top_left,
target_size,
dtype=prompt_embeds.dtype,
text_encoder_projection_dim=text_encoder_projection_dim,
)
if do_classifier_free_guidance:
prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds], dim=0)
add_text_embeds = torch.cat([negative_pooled_prompt_embeds, add_text_embeds], dim=0)
add_time_ids = torch.cat([add_time_ids, add_time_ids], dim=0)
prompt_embeds = prompt_embeds.to(device)
add_text_embeds = add_text_embeds.to(device)
add_time_ids = add_time_ids.to(device).repeat(batch_size * num_videos_per_prompt, 1)
num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
# Perform the first backward process up to time T_1
x_1_t1 = self.backward_loop(
timesteps=timesteps[: -t1 - 1],
prompt_embeds=prompt_embeds,
latents=latents,
guidance_scale=guidance_scale,
callback=callback,
callback_steps=callback_steps,
extra_step_kwargs=extra_step_kwargs,
num_warmup_steps=num_warmup_steps,
add_text_embeds=add_text_embeds,
add_time_ids=add_time_ids,
)
scheduler_copy = copy.deepcopy(self.scheduler)
# Perform the second backward process up to time T_0
x_1_t0 = self.backward_loop(
timesteps=timesteps[-t1 - 1 : -t0 - 1],
prompt_embeds=prompt_embeds,
latents=x_1_t1,
guidance_scale=guidance_scale,
callback=callback,
callback_steps=callback_steps,
extra_step_kwargs=extra_step_kwargs,
num_warmup_steps=0,
add_text_embeds=add_text_embeds,
add_time_ids=add_time_ids,
)
# Propagate first frame latents at time T_0 to remaining frames
x_2k_t0 = x_1_t0.repeat(video_length - 1, 1, 1, 1)
# Add motion in latents at time T_0
x_2k_t0 = create_motion_field_and_warp_latents(
motion_field_strength_x=motion_field_strength_x,
motion_field_strength_y=motion_field_strength_y,
latents=x_2k_t0,
frame_ids=frame_ids[1:],
)
# Perform forward process up to time T_1
x_2k_t1 = self.forward_loop(
x_t0=x_2k_t0,
t0=timesteps[-t0 - 1].to(torch.long),
t1=timesteps[-t1 - 1].to(torch.long),
generator=generator,
)
# Perform backward process from time T_1 to 0
latents = torch.cat([x_1_t1, x_2k_t1])
self.scheduler = scheduler_copy
timesteps = timesteps[-t1 - 1 :]
b, l, d = prompt_embeds.size()
prompt_embeds = prompt_embeds[:, None].repeat(1, video_length, 1, 1).reshape(b * video_length, l, d)
b, k = add_text_embeds.size()
add_text_embeds = add_text_embeds[:, None].repeat(1, video_length, 1).reshape(b * video_length, k)
b, k = add_time_ids.size()
add_time_ids = add_time_ids[:, None].repeat(1, video_length, 1).reshape(b * video_length, k)
# 7.1 Apply denoising_end
if denoising_end is not None and isinstance(denoising_end, float) and denoising_end > 0 and denoising_end < 1:
discrete_timestep_cutoff = int(
round(
self.scheduler.config.num_train_timesteps
- (denoising_end * self.scheduler.config.num_train_timesteps)
)
)
num_inference_steps = len(list(filter(lambda ts: ts >= discrete_timestep_cutoff, timesteps)))
timesteps = timesteps[:num_inference_steps]
x_1k_0 = self.backward_loop(
timesteps=timesteps,
prompt_embeds=prompt_embeds,
latents=latents,
guidance_scale=guidance_scale,
callback=callback,
callback_steps=callback_steps,
extra_step_kwargs=extra_step_kwargs,
num_warmup_steps=0,
add_text_embeds=add_text_embeds,
add_time_ids=add_time_ids,
)
latents = x_1k_0
if not output_type == "latent":
# make sure the VAE is in float32 mode, as it overflows in float16
needs_upcasting = self.vae.dtype == torch.float16 and self.vae.config.force_upcast
if needs_upcasting:
self.upcast_vae()
latents = latents.to(next(iter(self.vae.post_quant_conv.parameters())).dtype)
image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
# cast back to fp16 if needed
if needs_upcasting:
self.vae.to(dtype=torch.float16)
else:
image = latents
return TextToVideoSDXLPipelineOutput(images=image)
# apply watermark if available
if self.watermark is not None:
image = self.watermark.apply_watermark(image)
image = self.image_processor.postprocess(image, output_type=output_type)
# Offload last model to CPU manually for max memory savings
if hasattr(self, "final_offload_hook") and self.final_offload_hook is not None:
self.final_offload_hook.offload()
if not return_dict:
return (image,)
return TextToVideoSDXLPipelineOutput(images=image)
| diffusers/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_zero_sdxl.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/text_to_video_synthesis/pipeline_text_to_video_zero_sdxl.py",
"repo_id": "diffusers",
"token_count": 28769
} | 122 |
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from dataclasses import dataclass
from math import ceil
from typing import Callable, Dict, List, Optional, Union
import numpy as np
import torch
from transformers import CLIPTextModel, CLIPTokenizer
from ...loaders import LoraLoaderMixin
from ...schedulers import DDPMWuerstchenScheduler
from ...utils import BaseOutput, deprecate, logging, replace_example_docstring
from ...utils.torch_utils import randn_tensor
from ..pipeline_utils import DiffusionPipeline
from .modeling_wuerstchen_prior import WuerstchenPrior
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
DEFAULT_STAGE_C_TIMESTEPS = list(np.linspace(1.0, 2 / 3, 20)) + list(np.linspace(2 / 3, 0.0, 11))[1:]
EXAMPLE_DOC_STRING = """
Examples:
```py
>>> import torch
>>> from diffusers import WuerstchenPriorPipeline
>>> prior_pipe = WuerstchenPriorPipeline.from_pretrained(
... "warp-ai/wuerstchen-prior", torch_dtype=torch.float16
... ).to("cuda")
>>> prompt = "an image of a shiba inu, donning a spacesuit and helmet"
>>> prior_output = pipe(prompt)
```
"""
@dataclass
class WuerstchenPriorPipelineOutput(BaseOutput):
"""
Output class for WuerstchenPriorPipeline.
Args:
image_embeddings (`torch.FloatTensor` or `np.ndarray`)
Prior image embeddings for text prompt
"""
image_embeddings: Union[torch.FloatTensor, np.ndarray]
class WuerstchenPriorPipeline(DiffusionPipeline, LoraLoaderMixin):
"""
Pipeline for generating image prior for Wuerstchen.
This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the
library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.)
The pipeline also inherits the following loading methods:
- [`~loaders.LoraLoaderMixin.load_lora_weights`] for loading LoRA weights
- [`~loaders.LoraLoaderMixin.save_lora_weights`] for saving LoRA weights
Args:
prior ([`Prior`]):
The canonical unCLIP prior to approximate the image embedding from the text embedding.
text_encoder ([`CLIPTextModelWithProjection`]):
Frozen text-encoder.
tokenizer (`CLIPTokenizer`):
Tokenizer of class
[CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer).
scheduler ([`DDPMWuerstchenScheduler`]):
A scheduler to be used in combination with `prior` to generate image embedding.
latent_mean ('float', *optional*, defaults to 42.0):
Mean value for latent diffusers.
latent_std ('float', *optional*, defaults to 1.0):
Standard value for latent diffusers.
resolution_multiple ('float', *optional*, defaults to 42.67):
Default resolution for multiple images generated.
"""
unet_name = "prior"
text_encoder_name = "text_encoder"
model_cpu_offload_seq = "text_encoder->prior"
_callback_tensor_inputs = ["latents", "text_encoder_hidden_states", "negative_prompt_embeds"]
def __init__(
self,
tokenizer: CLIPTokenizer,
text_encoder: CLIPTextModel,
prior: WuerstchenPrior,
scheduler: DDPMWuerstchenScheduler,
latent_mean: float = 42.0,
latent_std: float = 1.0,
resolution_multiple: float = 42.67,
) -> None:
super().__init__()
self.register_modules(
tokenizer=tokenizer,
text_encoder=text_encoder,
prior=prior,
scheduler=scheduler,
)
self.register_to_config(
latent_mean=latent_mean, latent_std=latent_std, resolution_multiple=resolution_multiple
)
# Copied from diffusers.pipelines.unclip.pipeline_unclip.UnCLIPPipeline.prepare_latents
def prepare_latents(self, shape, dtype, device, generator, latents, scheduler):
if latents is None:
latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
else:
if latents.shape != shape:
raise ValueError(f"Unexpected latents shape, got {latents.shape}, expected {shape}")
latents = latents.to(device)
latents = latents * scheduler.init_noise_sigma
return latents
def encode_prompt(
self,
device,
num_images_per_prompt,
do_classifier_free_guidance,
prompt=None,
negative_prompt=None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
):
if prompt is not None and isinstance(prompt, str):
batch_size = 1
elif prompt is not None and isinstance(prompt, list):
batch_size = len(prompt)
else:
batch_size = prompt_embeds.shape[0]
if prompt_embeds is None:
# get prompt text embeddings
text_inputs = self.tokenizer(
prompt,
padding="max_length",
max_length=self.tokenizer.model_max_length,
truncation=True,
return_tensors="pt",
)
text_input_ids = text_inputs.input_ids
attention_mask = text_inputs.attention_mask
untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
text_input_ids, untruncated_ids
):
removed_text = self.tokenizer.batch_decode(
untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
)
logger.warning(
"The following part of your input was truncated because CLIP can only handle sequences up to"
f" {self.tokenizer.model_max_length} tokens: {removed_text}"
)
text_input_ids = text_input_ids[:, : self.tokenizer.model_max_length]
attention_mask = attention_mask[:, : self.tokenizer.model_max_length]
text_encoder_output = self.text_encoder(
text_input_ids.to(device), attention_mask=attention_mask.to(device)
)
prompt_embeds = text_encoder_output.last_hidden_state
prompt_embeds = prompt_embeds.to(dtype=self.text_encoder.dtype, device=device)
prompt_embeds = prompt_embeds.repeat_interleave(num_images_per_prompt, dim=0)
if negative_prompt_embeds is None and do_classifier_free_guidance:
uncond_tokens: List[str]
if negative_prompt is None:
uncond_tokens = [""] * batch_size
elif type(prompt) is not type(negative_prompt):
raise TypeError(
f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
f" {type(prompt)}."
)
elif isinstance(negative_prompt, str):
uncond_tokens = [negative_prompt]
elif batch_size != len(negative_prompt):
raise ValueError(
f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
" the batch size of `prompt`."
)
else:
uncond_tokens = negative_prompt
uncond_input = self.tokenizer(
uncond_tokens,
padding="max_length",
max_length=self.tokenizer.model_max_length,
truncation=True,
return_tensors="pt",
)
negative_prompt_embeds_text_encoder_output = self.text_encoder(
uncond_input.input_ids.to(device), attention_mask=uncond_input.attention_mask.to(device)
)
negative_prompt_embeds = negative_prompt_embeds_text_encoder_output.last_hidden_state
if do_classifier_free_guidance:
# duplicate unconditional embeddings for each generation per prompt, using mps friendly method
seq_len = negative_prompt_embeds.shape[1]
negative_prompt_embeds = negative_prompt_embeds.to(dtype=self.text_encoder.dtype, device=device)
negative_prompt_embeds = negative_prompt_embeds.repeat(1, num_images_per_prompt, 1)
negative_prompt_embeds = negative_prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
# done duplicates
return prompt_embeds, negative_prompt_embeds
def check_inputs(
self,
prompt,
negative_prompt,
num_inference_steps,
do_classifier_free_guidance,
prompt_embeds=None,
negative_prompt_embeds=None,
):
if prompt is not None and prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
" only forward one of the two."
)
elif prompt is None and prompt_embeds is None:
raise ValueError(
"Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
)
elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
if negative_prompt is not None and negative_prompt_embeds is not None:
raise ValueError(
f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
)
if prompt_embeds is not None and negative_prompt_embeds is not None:
if prompt_embeds.shape != negative_prompt_embeds.shape:
raise ValueError(
"`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
f" {negative_prompt_embeds.shape}."
)
if not isinstance(num_inference_steps, int):
raise TypeError(
f"'num_inference_steps' must be of type 'int', but got {type(num_inference_steps)}\
In Case you want to provide explicit timesteps, please use the 'timesteps' argument."
)
@property
def guidance_scale(self):
return self._guidance_scale
@property
def do_classifier_free_guidance(self):
return self._guidance_scale > 1
@property
def num_timesteps(self):
return self._num_timesteps
@torch.no_grad()
@replace_example_docstring(EXAMPLE_DOC_STRING)
def __call__(
self,
prompt: Optional[Union[str, List[str]]] = None,
height: int = 1024,
width: int = 1024,
num_inference_steps: int = 60,
timesteps: List[float] = None,
guidance_scale: float = 8.0,
negative_prompt: Optional[Union[str, List[str]]] = None,
prompt_embeds: Optional[torch.FloatTensor] = None,
negative_prompt_embeds: Optional[torch.FloatTensor] = None,
num_images_per_prompt: Optional[int] = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.FloatTensor] = None,
output_type: Optional[str] = "pt",
return_dict: bool = True,
callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
callback_on_step_end_tensor_inputs: List[str] = ["latents"],
**kwargs,
):
"""
Function invoked when calling the pipeline for generation.
Args:
prompt (`str` or `List[str]`):
The prompt or prompts to guide the image generation.
height (`int`, *optional*, defaults to 1024):
The height in pixels of the generated image.
width (`int`, *optional*, defaults to 1024):
The width in pixels of the generated image.
num_inference_steps (`int`, *optional*, defaults to 60):
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.
timesteps (`List[int]`, *optional*):
Custom timesteps to use for the denoising process. If not defined, equal spaced `num_inference_steps`
timesteps are used. Must be in descending order.
guidance_scale (`float`, *optional*, defaults to 8.0):
Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
`decoder_guidance_scale` is defined as `w` of equation 2. of [Imagen
Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting
`decoder_guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely
linked to the text `prompt`, usually at the expense of lower image quality.
negative_prompt (`str` or `List[str]`, *optional*):
The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
if `decoder_guidance_scale` is less than `1`).
prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
provided, text embeddings will be generated from `prompt` input argument.
negative_prompt_embeds (`torch.FloatTensor`, *optional*):
Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
argument.
num_images_per_prompt (`int`, *optional*, defaults to 1):
The number of images to generate per prompt.
generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
to make generation deterministic.
latents (`torch.FloatTensor`, *optional*):
Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
tensor will ge generated by sampling using the supplied random `generator`.
output_type (`str`, *optional*, defaults to `"pil"`):
The output format of the generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"`
(`np.array`) or `"pt"` (`torch.Tensor`).
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
callback_on_step_end (`Callable`, *optional*):
A function that calls at the end of each denoising steps during the inference. The function is called
with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int,
callback_kwargs: Dict)`. `callback_kwargs` will include a list of all tensors as specified by
`callback_on_step_end_tensor_inputs`.
callback_on_step_end_tensor_inputs (`List`, *optional*):
The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list
will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the
`._callback_tensor_inputs` attribute of your pipeline class.
Examples:
Returns:
[`~pipelines.WuerstchenPriorPipelineOutput`] or `tuple` [`~pipelines.WuerstchenPriorPipelineOutput`] if
`return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is a list with the
generated image embeddings.
"""
callback = kwargs.pop("callback", None)
callback_steps = kwargs.pop("callback_steps", None)
if callback is not None:
deprecate(
"callback",
"1.0.0",
"Passing `callback` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`",
)
if callback_steps is not None:
deprecate(
"callback_steps",
"1.0.0",
"Passing `callback_steps` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`",
)
if callback_on_step_end_tensor_inputs is not None and not all(
k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
):
raise ValueError(
f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
)
# 0. Define commonly used variables
device = self._execution_device
self._guidance_scale = guidance_scale
if prompt is not None and isinstance(prompt, str):
batch_size = 1
elif prompt is not None and isinstance(prompt, list):
batch_size = len(prompt)
else:
batch_size = prompt_embeds.shape[0]
# 1. Check inputs. Raise error if not correct
if prompt is not None and not isinstance(prompt, list):
if isinstance(prompt, str):
prompt = [prompt]
else:
raise TypeError(f"'prompt' must be of type 'list' or 'str', but got {type(prompt)}.")
if self.do_classifier_free_guidance:
if negative_prompt is not None and not isinstance(negative_prompt, list):
if isinstance(negative_prompt, str):
negative_prompt = [negative_prompt]
else:
raise TypeError(
f"'negative_prompt' must be of type 'list' or 'str', but got {type(negative_prompt)}."
)
self.check_inputs(
prompt,
negative_prompt,
num_inference_steps,
self.do_classifier_free_guidance,
prompt_embeds=prompt_embeds,
negative_prompt_embeds=negative_prompt_embeds,
)
# 2. Encode caption
prompt_embeds, negative_prompt_embeds = self.encode_prompt(
prompt=prompt,
device=device,
num_images_per_prompt=num_images_per_prompt,
do_classifier_free_guidance=self.do_classifier_free_guidance,
negative_prompt=negative_prompt,
prompt_embeds=prompt_embeds,
negative_prompt_embeds=negative_prompt_embeds,
)
# For classifier free guidance, we need to do two forward passes.
# Here we concatenate the unconditional and text embeddings into a single batch
# to avoid doing two forward passes
text_encoder_hidden_states = (
torch.cat([prompt_embeds, negative_prompt_embeds]) if negative_prompt_embeds is not None else prompt_embeds
)
# 3. Determine latent shape of image embeddings
dtype = text_encoder_hidden_states.dtype
latent_height = ceil(height / self.config.resolution_multiple)
latent_width = ceil(width / self.config.resolution_multiple)
num_channels = self.prior.config.c_in
effnet_features_shape = (num_images_per_prompt * batch_size, num_channels, latent_height, latent_width)
# 4. Prepare and set timesteps
if timesteps is not None:
self.scheduler.set_timesteps(timesteps=timesteps, device=device)
timesteps = self.scheduler.timesteps
num_inference_steps = len(timesteps)
else:
self.scheduler.set_timesteps(num_inference_steps, device=device)
timesteps = self.scheduler.timesteps
# 5. Prepare latents
latents = self.prepare_latents(effnet_features_shape, dtype, device, generator, latents, self.scheduler)
# 6. Run denoising loop
self._num_timesteps = len(timesteps[:-1])
for i, t in enumerate(self.progress_bar(timesteps[:-1])):
ratio = t.expand(latents.size(0)).to(dtype)
# 7. Denoise image embeddings
predicted_image_embedding = self.prior(
torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents,
r=torch.cat([ratio] * 2) if self.do_classifier_free_guidance else ratio,
c=text_encoder_hidden_states,
)
# 8. Check for classifier free guidance and apply it
if self.do_classifier_free_guidance:
predicted_image_embedding_text, predicted_image_embedding_uncond = predicted_image_embedding.chunk(2)
predicted_image_embedding = torch.lerp(
predicted_image_embedding_uncond, predicted_image_embedding_text, self.guidance_scale
)
# 9. Renoise latents to next timestep
latents = self.scheduler.step(
model_output=predicted_image_embedding,
timestep=ratio,
sample=latents,
generator=generator,
).prev_sample
if callback_on_step_end is not None:
callback_kwargs = {}
for k in callback_on_step_end_tensor_inputs:
callback_kwargs[k] = locals()[k]
callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)
latents = callback_outputs.pop("latents", latents)
text_encoder_hidden_states = callback_outputs.pop(
"text_encoder_hidden_states", text_encoder_hidden_states
)
negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)
if callback is not None and i % callback_steps == 0:
step_idx = i // getattr(self.scheduler, "order", 1)
callback(step_idx, t, latents)
# 10. Denormalize the latents
latents = latents * self.config.latent_mean - self.config.latent_std
# Offload all models
self.maybe_free_model_hooks()
if output_type == "np":
latents = latents.cpu().numpy()
if not return_dict:
return (latents,)
return WuerstchenPriorPipelineOutput(latents)
| diffusers/src/diffusers/pipelines/wuerstchen/pipeline_wuerstchen_prior.py/0 | {
"file_path": "diffusers/src/diffusers/pipelines/wuerstchen/pipeline_wuerstchen_prior.py",
"repo_id": "diffusers",
"token_count": 10628
} | 123 |
# Copyright 2023 Stanford University Team and The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# DISCLAIMER: This code is strongly influenced by https://github.com/pesser/pytorch_diffusion
# and https://github.com/hojonathanho/diffusion
import math
from dataclasses import dataclass
from typing import List, Optional, Tuple, Union
import numpy as np
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, logging
from ..utils.torch_utils import randn_tensor
from .scheduling_utils import SchedulerMixin
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
@dataclass
class LCMSchedulerOutput(BaseOutput):
"""
Output class for the scheduler's `step` function output.
Args:
prev_sample (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)` for images):
Computed sample `(x_{t-1})` of previous timestep. `prev_sample` should be used as next model input in the
denoising loop.
pred_original_sample (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)` for images):
The predicted denoised sample `(x_{0})` based on the model output from the current timestep.
`pred_original_sample` can be used to preview progress or for guidance.
"""
prev_sample: torch.FloatTensor
denoised: Optional[torch.FloatTensor] = None
# Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
def betas_for_alpha_bar(
num_diffusion_timesteps,
max_beta=0.999,
alpha_transform_type="cosine",
):
"""
Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
(1-beta) over time from t = [0,1].
Contains a function alpha_bar that takes an argument t and transforms it to the cumulative product of (1-beta) up
to that part of the diffusion process.
Args:
num_diffusion_timesteps (`int`): the number of betas to produce.
max_beta (`float`): the maximum beta to use; use values lower than 1 to
prevent singularities.
alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
Choose from `cosine` or `exp`
Returns:
betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
"""
if alpha_transform_type == "cosine":
def alpha_bar_fn(t):
return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
elif alpha_transform_type == "exp":
def alpha_bar_fn(t):
return math.exp(t * -12.0)
else:
raise ValueError(f"Unsupported alpha_tranform_type: {alpha_transform_type}")
betas = []
for i in range(num_diffusion_timesteps):
t1 = i / num_diffusion_timesteps
t2 = (i + 1) / num_diffusion_timesteps
betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
return torch.tensor(betas, dtype=torch.float32)
# Copied from diffusers.schedulers.scheduling_ddim.rescale_zero_terminal_snr
def rescale_zero_terminal_snr(betas: torch.FloatTensor) -> torch.FloatTensor:
"""
Rescales betas to have zero terminal SNR Based on https://arxiv.org/pdf/2305.08891.pdf (Algorithm 1)
Args:
betas (`torch.FloatTensor`):
the betas that the scheduler is being initialized with.
Returns:
`torch.FloatTensor`: rescaled betas with zero terminal SNR
"""
# Convert betas to alphas_bar_sqrt
alphas = 1.0 - betas
alphas_cumprod = torch.cumprod(alphas, dim=0)
alphas_bar_sqrt = alphas_cumprod.sqrt()
# Store old values.
alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone()
alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone()
# Shift so the last timestep is zero.
alphas_bar_sqrt -= alphas_bar_sqrt_T
# Scale so the first timestep is back to the old value.
alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T)
# Convert alphas_bar_sqrt to betas
alphas_bar = alphas_bar_sqrt**2 # Revert sqrt
alphas = alphas_bar[1:] / alphas_bar[:-1] # Revert cumprod
alphas = torch.cat([alphas_bar[0:1], alphas])
betas = 1 - alphas
return betas
class LCMScheduler(SchedulerMixin, ConfigMixin):
"""
`LCMScheduler` extends the denoising procedure introduced in denoising diffusion probabilistic models (DDPMs) with
non-Markovian guidance.
This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. [`~ConfigMixin`] takes care of storing all config
attributes that are passed in the scheduler's `__init__` function, such as `num_train_timesteps`. They can be
accessed via `scheduler.config.num_train_timesteps`. [`SchedulerMixin`] provides general loading and saving
functionality via the [`SchedulerMixin.save_pretrained`] and [`~SchedulerMixin.from_pretrained`] functions.
Args:
num_train_timesteps (`int`, defaults to 1000):
The number of diffusion steps to train the model.
beta_start (`float`, defaults to 0.0001):
The starting `beta` value of inference.
beta_end (`float`, defaults to 0.02):
The final `beta` value.
beta_schedule (`str`, defaults to `"linear"`):
The beta schedule, a mapping from a beta range to a sequence of betas for stepping the model. Choose from
`linear`, `scaled_linear`, or `squaredcos_cap_v2`.
trained_betas (`np.ndarray`, *optional*):
Pass an array of betas directly to the constructor to bypass `beta_start` and `beta_end`.
original_inference_steps (`int`, *optional*, defaults to 50):
The default number of inference steps used to generate a linearly-spaced timestep schedule, from which we
will ultimately take `num_inference_steps` evenly spaced timesteps to form the final timestep schedule.
clip_sample (`bool`, defaults to `True`):
Clip the predicted sample for numerical stability.
clip_sample_range (`float`, defaults to 1.0):
The maximum magnitude for sample clipping. Valid only when `clip_sample=True`.
set_alpha_to_one (`bool`, defaults to `True`):
Each diffusion step uses the alphas product value at that step and at the previous one. For the final step
there is no previous alpha. When this option is `True` the previous alpha product is fixed to `1`,
otherwise it uses the alpha value at step 0.
steps_offset (`int`, defaults to 0):
An offset added to the inference steps. You can use a combination of `offset=1` and
`set_alpha_to_one=False` to make the last step use step 0 for the previous alpha product like in Stable
Diffusion.
prediction_type (`str`, defaults to `epsilon`, *optional*):
Prediction type of the scheduler function; can be `epsilon` (predicts the noise of the diffusion process),
`sample` (directly predicts the noisy sample`) or `v_prediction` (see section 2.4 of [Imagen
Video](https://imagen.research.google/video/paper.pdf) paper).
thresholding (`bool`, defaults to `False`):
Whether to use the "dynamic thresholding" method. This is unsuitable for latent-space diffusion models such
as Stable Diffusion.
dynamic_thresholding_ratio (`float`, defaults to 0.995):
The ratio for the dynamic thresholding method. Valid only when `thresholding=True`.
sample_max_value (`float`, defaults to 1.0):
The threshold value for dynamic thresholding. Valid only when `thresholding=True`.
timestep_spacing (`str`, defaults to `"leading"`):
The way the timesteps should be scaled. Refer to Table 2 of the [Common Diffusion Noise Schedules and
Sample Steps are Flawed](https://huggingface.co/papers/2305.08891) for more information.
timestep_scaling (`float`, defaults to 10.0):
The factor the timesteps will be multiplied by when calculating the consistency model boundary conditions
`c_skip` and `c_out`. Increasing this will decrease the approximation error (although the approximation
error at the default of `10.0` is already pretty small).
rescale_betas_zero_snr (`bool`, defaults to `False`):
Whether to rescale the betas to have zero terminal SNR. This enables the model to generate very bright and
dark samples instead of limiting it to samples with medium brightness. Loosely related to
[`--offset_noise`](https://github.com/huggingface/diffusers/blob/74fd735eb073eb1d774b1ab4154a0876eb82f055/examples/dreambooth/train_dreambooth.py#L506).
"""
order = 1
@register_to_config
def __init__(
self,
num_train_timesteps: int = 1000,
beta_start: float = 0.00085,
beta_end: float = 0.012,
beta_schedule: str = "scaled_linear",
trained_betas: Optional[Union[np.ndarray, List[float]]] = None,
original_inference_steps: int = 50,
clip_sample: bool = False,
clip_sample_range: float = 1.0,
set_alpha_to_one: bool = True,
steps_offset: int = 0,
prediction_type: str = "epsilon",
thresholding: bool = False,
dynamic_thresholding_ratio: float = 0.995,
sample_max_value: float = 1.0,
timestep_spacing: str = "leading",
timestep_scaling: float = 10.0,
rescale_betas_zero_snr: bool = False,
):
if trained_betas is not None:
self.betas = torch.tensor(trained_betas, dtype=torch.float32)
elif beta_schedule == "linear":
self.betas = torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32)
elif beta_schedule == "scaled_linear":
# this schedule is very specific to the latent diffusion model.
self.betas = torch.linspace(beta_start**0.5, beta_end**0.5, num_train_timesteps, dtype=torch.float32) ** 2
elif beta_schedule == "squaredcos_cap_v2":
# Glide cosine schedule
self.betas = betas_for_alpha_bar(num_train_timesteps)
else:
raise NotImplementedError(f"{beta_schedule} does is not implemented for {self.__class__}")
# Rescale for zero SNR
if rescale_betas_zero_snr:
self.betas = rescale_zero_terminal_snr(self.betas)
self.alphas = 1.0 - self.betas
self.alphas_cumprod = torch.cumprod(self.alphas, dim=0)
# At every step in ddim, we are looking into the previous alphas_cumprod
# For the final step, there is no previous alphas_cumprod because we are already at 0
# `set_alpha_to_one` decides whether we set this parameter simply to one or
# whether we use the final alpha of the "non-previous" one.
self.final_alpha_cumprod = torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]
# standard deviation of the initial noise distribution
self.init_noise_sigma = 1.0
# setable values
self.num_inference_steps = None
self.timesteps = torch.from_numpy(np.arange(0, num_train_timesteps)[::-1].copy().astype(np.int64))
self.custom_timesteps = False
self._step_index = None
self._begin_index = None
# Copied from diffusers.schedulers.scheduling_euler_discrete.EulerDiscreteScheduler.index_for_timestep
def index_for_timestep(self, timestep, schedule_timesteps=None):
if schedule_timesteps is None:
schedule_timesteps = self.timesteps
indices = (schedule_timesteps == timestep).nonzero()
# The sigma index that is taken for the **very** first `step`
# is always the second index (or the last index if there is only 1)
# This way we can ensure we don't accidentally skip a sigma in
# case we start in the middle of the denoising schedule (e.g. for image-to-image)
pos = 1 if len(indices) > 1 else 0
return indices[pos].item()
# Copied from diffusers.schedulers.scheduling_euler_discrete.EulerDiscreteScheduler._init_step_index
def _init_step_index(self, timestep):
if self.begin_index is None:
if isinstance(timestep, torch.Tensor):
timestep = timestep.to(self.timesteps.device)
self._step_index = self.index_for_timestep(timestep)
else:
self._step_index = self._begin_index
@property
def step_index(self):
return self._step_index
@property
def begin_index(self):
"""
The index for the first timestep. It should be set from pipeline with `set_begin_index` method.
"""
return self._begin_index
# Copied from diffusers.schedulers.scheduling_dpmsolver_multistep.DPMSolverMultistepScheduler.set_begin_index
def set_begin_index(self, begin_index: int = 0):
"""
Sets the begin index for the scheduler. This function should be run from pipeline before the inference.
Args:
begin_index (`int`):
The begin index for the scheduler.
"""
self._begin_index = begin_index
def scale_model_input(self, sample: torch.FloatTensor, timestep: Optional[int] = None) -> torch.FloatTensor:
"""
Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
current timestep.
Args:
sample (`torch.FloatTensor`):
The input sample.
timestep (`int`, *optional*):
The current timestep in the diffusion chain.
Returns:
`torch.FloatTensor`:
A scaled input sample.
"""
return sample
# Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler._threshold_sample
def _threshold_sample(self, sample: torch.FloatTensor) -> torch.FloatTensor:
"""
"Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the
prediction of x_0 at timestep t), and if s > 1, then we threshold xt0 to the range [-s, s] and then divide by
s. Dynamic thresholding pushes saturated pixels (those near -1 and 1) inwards, thereby actively preventing
pixels from saturation at each step. We find that dynamic thresholding results in significantly better
photorealism as well as better image-text alignment, especially when using very large guidance weights."
https://arxiv.org/abs/2205.11487
"""
dtype = sample.dtype
batch_size, channels, *remaining_dims = sample.shape
if dtype not in (torch.float32, torch.float64):
sample = sample.float() # upcast for quantile calculation, and clamp not implemented for cpu half
# Flatten sample for doing quantile calculation along each image
sample = sample.reshape(batch_size, channels * np.prod(remaining_dims))
abs_sample = sample.abs() # "a certain percentile absolute pixel value"
s = torch.quantile(abs_sample, self.config.dynamic_thresholding_ratio, dim=1)
s = torch.clamp(
s, min=1, max=self.config.sample_max_value
) # When clamped to min=1, equivalent to standard clipping to [-1, 1]
s = s.unsqueeze(1) # (batch_size, 1) because clamp will broadcast along dim=0
sample = torch.clamp(sample, -s, s) / s # "we threshold xt0 to the range [-s, s] and then divide by s"
sample = sample.reshape(batch_size, channels, *remaining_dims)
sample = sample.to(dtype)
return sample
def set_timesteps(
self,
num_inference_steps: Optional[int] = None,
device: Union[str, torch.device] = None,
original_inference_steps: Optional[int] = None,
timesteps: Optional[List[int]] = None,
strength: int = 1.0,
):
"""
Sets the discrete timesteps used for the diffusion chain (to be run before inference).
Args:
num_inference_steps (`int`, *optional*):
The number of diffusion steps used when generating samples with a pre-trained model. If used,
`timesteps` must be `None`.
device (`str` or `torch.device`, *optional*):
The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
original_inference_steps (`int`, *optional*):
The original number of inference steps, which will be used to generate a linearly-spaced timestep
schedule (which is different from the standard `diffusers` implementation). We will then take
`num_inference_steps` timesteps from this schedule, evenly spaced in terms of indices, and use that as
our final timestep schedule. If not set, this will default to the `original_inference_steps` attribute.
timesteps (`List[int]`, *optional*):
Custom timesteps used to support arbitrary spacing between timesteps. If `None`, then the default
timestep spacing strategy of equal spacing between timesteps on the training/distillation timestep
schedule is used. If `timesteps` is passed, `num_inference_steps` must be `None`.
"""
# 0. Check inputs
if num_inference_steps is None and timesteps is None:
raise ValueError("Must pass exactly one of `num_inference_steps` or `custom_timesteps`.")
if num_inference_steps is not None and timesteps is not None:
raise ValueError("Can only pass one of `num_inference_steps` or `custom_timesteps`.")
# 1. Calculate the LCM original training/distillation timestep schedule.
original_steps = (
original_inference_steps if original_inference_steps is not None else self.config.original_inference_steps
)
if original_steps > self.config.num_train_timesteps:
raise ValueError(
f"`original_steps`: {original_steps} cannot be larger than `self.config.train_timesteps`:"
f" {self.config.num_train_timesteps} as the unet model trained with this scheduler can only handle"
f" maximal {self.config.num_train_timesteps} timesteps."
)
# LCM Timesteps Setting
# The skipping step parameter k from the paper.
k = self.config.num_train_timesteps // original_steps
# LCM Training/Distillation Steps Schedule
# Currently, only a linearly-spaced schedule is supported (same as in the LCM distillation scripts).
lcm_origin_timesteps = np.asarray(list(range(1, int(original_steps * strength) + 1))) * k - 1
# 2. Calculate the LCM inference timestep schedule.
if timesteps is not None:
# 2.1 Handle custom timestep schedules.
train_timesteps = set(lcm_origin_timesteps)
non_train_timesteps = []
for i in range(1, len(timesteps)):
if timesteps[i] >= timesteps[i - 1]:
raise ValueError("`custom_timesteps` must be in descending order.")
if timesteps[i] not in train_timesteps:
non_train_timesteps.append(timesteps[i])
if timesteps[0] >= self.config.num_train_timesteps:
raise ValueError(
f"`timesteps` must start before `self.config.train_timesteps`:"
f" {self.config.num_train_timesteps}."
)
# Raise warning if timestep schedule does not start with self.config.num_train_timesteps - 1
if strength == 1.0 and timesteps[0] != self.config.num_train_timesteps - 1:
logger.warning(
f"The first timestep on the custom timestep schedule is {timesteps[0]}, not"
f" `self.config.num_train_timesteps - 1`: {self.config.num_train_timesteps - 1}. You may get"
f" unexpected results when using this timestep schedule."
)
# Raise warning if custom timestep schedule contains timesteps not on original timestep schedule
if non_train_timesteps:
logger.warning(
f"The custom timestep schedule contains the following timesteps which are not on the original"
f" training/distillation timestep schedule: {non_train_timesteps}. You may get unexpected results"
f" when using this timestep schedule."
)
# Raise warning if custom timestep schedule is longer than original_steps
if len(timesteps) > original_steps:
logger.warning(
f"The number of timesteps in the custom timestep schedule is {len(timesteps)}, which exceeds the"
f" the length of the timestep schedule used for training: {original_steps}. You may get some"
f" unexpected results when using this timestep schedule."
)
timesteps = np.array(timesteps, dtype=np.int64)
self.num_inference_steps = len(timesteps)
self.custom_timesteps = True
# Apply strength (e.g. for img2img pipelines) (see StableDiffusionImg2ImgPipeline.get_timesteps)
init_timestep = min(int(self.num_inference_steps * strength), self.num_inference_steps)
t_start = max(self.num_inference_steps - init_timestep, 0)
timesteps = timesteps[t_start * self.order :]
# TODO: also reset self.num_inference_steps?
else:
# 2.2 Create the "standard" LCM inference timestep schedule.
if num_inference_steps > self.config.num_train_timesteps:
raise ValueError(
f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.config.train_timesteps`:"
f" {self.config.num_train_timesteps} as the unet model trained with this scheduler can only handle"
f" maximal {self.config.num_train_timesteps} timesteps."
)
skipping_step = len(lcm_origin_timesteps) // num_inference_steps
if skipping_step < 1:
raise ValueError(
f"The combination of `original_steps x strength`: {original_steps} x {strength} is smaller than `num_inference_steps`: {num_inference_steps}. Make sure to either reduce `num_inference_steps` to a value smaller than {int(original_steps * strength)} or increase `strength` to a value higher than {float(num_inference_steps / original_steps)}."
)
self.num_inference_steps = num_inference_steps
if num_inference_steps > original_steps:
raise ValueError(
f"`num_inference_steps`: {num_inference_steps} cannot be larger than `original_inference_steps`:"
f" {original_steps} because the final timestep schedule will be a subset of the"
f" `original_inference_steps`-sized initial timestep schedule."
)
# LCM Inference Steps Schedule
lcm_origin_timesteps = lcm_origin_timesteps[::-1].copy()
# Select (approximately) evenly spaced indices from lcm_origin_timesteps.
inference_indices = np.linspace(0, len(lcm_origin_timesteps), num=num_inference_steps, endpoint=False)
inference_indices = np.floor(inference_indices).astype(np.int64)
timesteps = lcm_origin_timesteps[inference_indices]
self.timesteps = torch.from_numpy(timesteps).to(device=device, dtype=torch.long)
self._step_index = None
self._begin_index = None
def get_scalings_for_boundary_condition_discrete(self, timestep):
self.sigma_data = 0.5 # Default: 0.5
scaled_timestep = timestep * self.config.timestep_scaling
c_skip = self.sigma_data**2 / (scaled_timestep**2 + self.sigma_data**2)
c_out = scaled_timestep / (scaled_timestep**2 + self.sigma_data**2) ** 0.5
return c_skip, c_out
def step(
self,
model_output: torch.FloatTensor,
timestep: int,
sample: torch.FloatTensor,
generator: Optional[torch.Generator] = None,
return_dict: bool = True,
) -> Union[LCMSchedulerOutput, Tuple]:
"""
Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion
process from the learned model outputs (most often the predicted noise).
Args:
model_output (`torch.FloatTensor`):
The direct output from learned diffusion model.
timestep (`float`):
The current discrete timestep in the diffusion chain.
sample (`torch.FloatTensor`):
A current instance of a sample created by the diffusion process.
generator (`torch.Generator`, *optional*):
A random number generator.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~schedulers.scheduling_lcm.LCMSchedulerOutput`] or `tuple`.
Returns:
[`~schedulers.scheduling_utils.LCMSchedulerOutput`] or `tuple`:
If return_dict is `True`, [`~schedulers.scheduling_lcm.LCMSchedulerOutput`] is returned, otherwise a
tuple is returned where the first element is the sample tensor.
"""
if self.num_inference_steps is None:
raise ValueError(
"Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
)
if self.step_index is None:
self._init_step_index(timestep)
# 1. get previous step value
prev_step_index = self.step_index + 1
if prev_step_index < len(self.timesteps):
prev_timestep = self.timesteps[prev_step_index]
else:
prev_timestep = timestep
# 2. compute alphas, betas
alpha_prod_t = self.alphas_cumprod[timestep]
alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
beta_prod_t = 1 - alpha_prod_t
beta_prod_t_prev = 1 - alpha_prod_t_prev
# 3. Get scalings for boundary conditions
c_skip, c_out = self.get_scalings_for_boundary_condition_discrete(timestep)
# 4. Compute the predicted original sample x_0 based on the model parameterization
if self.config.prediction_type == "epsilon": # noise-prediction
predicted_original_sample = (sample - beta_prod_t.sqrt() * model_output) / alpha_prod_t.sqrt()
elif self.config.prediction_type == "sample": # x-prediction
predicted_original_sample = model_output
elif self.config.prediction_type == "v_prediction": # v-prediction
predicted_original_sample = alpha_prod_t.sqrt() * sample - beta_prod_t.sqrt() * model_output
else:
raise ValueError(
f"prediction_type given as {self.config.prediction_type} must be one of `epsilon`, `sample` or"
" `v_prediction` for `LCMScheduler`."
)
# 5. Clip or threshold "predicted x_0"
if self.config.thresholding:
predicted_original_sample = self._threshold_sample(predicted_original_sample)
elif self.config.clip_sample:
predicted_original_sample = predicted_original_sample.clamp(
-self.config.clip_sample_range, self.config.clip_sample_range
)
# 6. Denoise model output using boundary conditions
denoised = c_out * predicted_original_sample + c_skip * sample
# 7. Sample and inject noise z ~ N(0, I) for MultiStep Inference
# Noise is not used on the final timestep of the timestep schedule.
# This also means that noise is not used for one-step sampling.
if self.step_index != self.num_inference_steps - 1:
noise = randn_tensor(
model_output.shape, generator=generator, device=model_output.device, dtype=denoised.dtype
)
prev_sample = alpha_prod_t_prev.sqrt() * denoised + beta_prod_t_prev.sqrt() * noise
else:
prev_sample = denoised
# upon completion increase step index by one
self._step_index += 1
if not return_dict:
return (prev_sample, denoised)
return LCMSchedulerOutput(prev_sample=prev_sample, denoised=denoised)
# Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.add_noise
def add_noise(
self,
original_samples: torch.FloatTensor,
noise: torch.FloatTensor,
timesteps: torch.IntTensor,
) -> torch.FloatTensor:
# Make sure alphas_cumprod and timestep have same device and dtype as original_samples
# Move the self.alphas_cumprod to device to avoid redundant CPU to GPU data movement
# for the subsequent add_noise calls
self.alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device)
alphas_cumprod = self.alphas_cumprod.to(dtype=original_samples.dtype)
timesteps = timesteps.to(original_samples.device)
sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
sqrt_alpha_prod = sqrt_alpha_prod.flatten()
while len(sqrt_alpha_prod.shape) < len(original_samples.shape):
sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
while len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape):
sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
noisy_samples = sqrt_alpha_prod * original_samples + sqrt_one_minus_alpha_prod * noise
return noisy_samples
# Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.get_velocity
def get_velocity(
self, sample: torch.FloatTensor, noise: torch.FloatTensor, timesteps: torch.IntTensor
) -> torch.FloatTensor:
# Make sure alphas_cumprod and timestep have same device and dtype as sample
self.alphas_cumprod = self.alphas_cumprod.to(device=sample.device)
alphas_cumprod = self.alphas_cumprod.to(dtype=sample.dtype)
timesteps = timesteps.to(sample.device)
sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
sqrt_alpha_prod = sqrt_alpha_prod.flatten()
while len(sqrt_alpha_prod.shape) < len(sample.shape):
sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
while len(sqrt_one_minus_alpha_prod.shape) < len(sample.shape):
sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
velocity = sqrt_alpha_prod * noise - sqrt_one_minus_alpha_prod * sample
return velocity
def __len__(self):
return self.config.num_train_timesteps
# Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.previous_timestep
def previous_timestep(self, timestep):
if self.custom_timesteps:
index = (self.timesteps == timestep).nonzero(as_tuple=True)[0][0]
if index == self.timesteps.shape[0] - 1:
prev_t = torch.tensor(-1)
else:
prev_t = self.timesteps[index + 1]
else:
num_inference_steps = (
self.num_inference_steps if self.num_inference_steps else self.config.num_train_timesteps
)
prev_t = timestep - self.config.num_train_timesteps // num_inference_steps
return prev_t
| diffusers/src/diffusers/schedulers/scheduling_lcm.py/0 | {
"file_path": "diffusers/src/diffusers/schedulers/scheduling_lcm.py",
"repo_id": "diffusers",
"token_count": 13496
} | 124 |
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Accelerate utilities: Utilities related to accelerate
"""
from packaging import version
from .import_utils import is_accelerate_available
if is_accelerate_available():
import accelerate
def apply_forward_hook(method):
"""
Decorator that applies a registered CpuOffload hook to an arbitrary function rather than `forward`. This is useful
for cases where a PyTorch module provides functions other than `forward` that should trigger a move to the
appropriate acceleration device. This is the case for `encode` and `decode` in [`AutoencoderKL`].
This decorator looks inside the internal `_hf_hook` property to find a registered offload hook.
:param method: The method to decorate. This method should be a method of a PyTorch module.
"""
if not is_accelerate_available():
return method
accelerate_version = version.parse(accelerate.__version__).base_version
if version.parse(accelerate_version) < version.parse("0.17.0"):
return method
def wrapper(self, *args, **kwargs):
if hasattr(self, "_hf_hook") and hasattr(self._hf_hook, "pre_forward"):
self._hf_hook.pre_forward(self)
return method(self, *args, **kwargs)
return wrapper
| diffusers/src/diffusers/utils/accelerate_utils.py/0 | {
"file_path": "diffusers/src/diffusers/utils/accelerate_utils.py",
"repo_id": "diffusers",
"token_count": 559
} | 125 |
# coding=utf-8
# Copyright 2023 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Utilities to dynamically load objects from the Hub."""
import importlib
import inspect
import json
import os
import re
import shutil
import sys
from pathlib import Path
from typing import Dict, Optional, Union
from urllib import request
from huggingface_hub import cached_download, hf_hub_download, model_info
from huggingface_hub.utils import validate_hf_hub_args
from packaging import version
from .. import __version__
from . import DIFFUSERS_DYNAMIC_MODULE_NAME, HF_MODULES_CACHE, logging
COMMUNITY_PIPELINES_URL = (
"https://raw.githubusercontent.com/huggingface/diffusers/{revision}/examples/community/{pipeline}.py"
)
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
def get_diffusers_versions():
url = "https://pypi.org/pypi/diffusers/json"
releases = json.loads(request.urlopen(url).read())["releases"].keys()
return sorted(releases, key=lambda x: version.Version(x))
def init_hf_modules():
"""
Creates the cache directory for modules with an init, and adds it to the Python path.
"""
# This function has already been executed if HF_MODULES_CACHE already is in the Python path.
if HF_MODULES_CACHE in sys.path:
return
sys.path.append(HF_MODULES_CACHE)
os.makedirs(HF_MODULES_CACHE, exist_ok=True)
init_path = Path(HF_MODULES_CACHE) / "__init__.py"
if not init_path.exists():
init_path.touch()
def create_dynamic_module(name: Union[str, os.PathLike]):
"""
Creates a dynamic module in the cache directory for modules.
"""
init_hf_modules()
dynamic_module_path = Path(HF_MODULES_CACHE) / name
# If the parent module does not exist yet, recursively create it.
if not dynamic_module_path.parent.exists():
create_dynamic_module(dynamic_module_path.parent)
os.makedirs(dynamic_module_path, exist_ok=True)
init_path = dynamic_module_path / "__init__.py"
if not init_path.exists():
init_path.touch()
def get_relative_imports(module_file):
"""
Get the list of modules that are relatively imported in a module file.
Args:
module_file (`str` or `os.PathLike`): The module file to inspect.
"""
with open(module_file, "r", encoding="utf-8") as f:
content = f.read()
# Imports of the form `import .xxx`
relative_imports = re.findall(r"^\s*import\s+\.(\S+)\s*$", content, flags=re.MULTILINE)
# Imports of the form `from .xxx import yyy`
relative_imports += re.findall(r"^\s*from\s+\.(\S+)\s+import", content, flags=re.MULTILINE)
# Unique-ify
return list(set(relative_imports))
def get_relative_import_files(module_file):
"""
Get the list of all files that are needed for a given module. Note that this function recurses through the relative
imports (if a imports b and b imports c, it will return module files for b and c).
Args:
module_file (`str` or `os.PathLike`): The module file to inspect.
"""
no_change = False
files_to_check = [module_file]
all_relative_imports = []
# Let's recurse through all relative imports
while not no_change:
new_imports = []
for f in files_to_check:
new_imports.extend(get_relative_imports(f))
module_path = Path(module_file).parent
new_import_files = [str(module_path / m) for m in new_imports]
new_import_files = [f for f in new_import_files if f not in all_relative_imports]
files_to_check = [f"{f}.py" for f in new_import_files]
no_change = len(new_import_files) == 0
all_relative_imports.extend(files_to_check)
return all_relative_imports
def check_imports(filename):
"""
Check if the current Python environment contains all the libraries that are imported in a file.
"""
with open(filename, "r", encoding="utf-8") as f:
content = f.read()
# Imports of the form `import xxx`
imports = re.findall(r"^\s*import\s+(\S+)\s*$", content, flags=re.MULTILINE)
# Imports of the form `from xxx import yyy`
imports += re.findall(r"^\s*from\s+(\S+)\s+import", content, flags=re.MULTILINE)
# Only keep the top-level module
imports = [imp.split(".")[0] for imp in imports if not imp.startswith(".")]
# Unique-ify and test we got them all
imports = list(set(imports))
missing_packages = []
for imp in imports:
try:
importlib.import_module(imp)
except ImportError:
missing_packages.append(imp)
if len(missing_packages) > 0:
raise ImportError(
"This modeling file requires the following packages that were not found in your environment: "
f"{', '.join(missing_packages)}. Run `pip install {' '.join(missing_packages)}`"
)
return get_relative_imports(filename)
def get_class_in_module(class_name, module_path):
"""
Import a module on the cache directory for modules and extract a class from it.
"""
module_path = module_path.replace(os.path.sep, ".")
module = importlib.import_module(module_path)
if class_name is None:
return find_pipeline_class(module)
return getattr(module, class_name)
def find_pipeline_class(loaded_module):
"""
Retrieve pipeline class that inherits from `DiffusionPipeline`. Note that there has to be exactly one class
inheriting from `DiffusionPipeline`.
"""
from ..pipelines import DiffusionPipeline
cls_members = dict(inspect.getmembers(loaded_module, inspect.isclass))
pipeline_class = None
for cls_name, cls in cls_members.items():
if (
cls_name != DiffusionPipeline.__name__
and issubclass(cls, DiffusionPipeline)
and cls.__module__.split(".")[0] != "diffusers"
):
if pipeline_class is not None:
raise ValueError(
f"Multiple classes that inherit from {DiffusionPipeline.__name__} have been found:"
f" {pipeline_class.__name__}, and {cls_name}. Please make sure to define only one in"
f" {loaded_module}."
)
pipeline_class = cls
return pipeline_class
@validate_hf_hub_args
def get_cached_module_file(
pretrained_model_name_or_path: Union[str, os.PathLike],
module_file: str,
cache_dir: Optional[Union[str, os.PathLike]] = None,
force_download: bool = False,
resume_download: bool = False,
proxies: Optional[Dict[str, str]] = None,
token: Optional[Union[bool, str]] = None,
revision: Optional[str] = None,
local_files_only: bool = False,
):
"""
Prepares Downloads a module from a local folder or a distant repo and returns its path inside the cached
Transformers module.
Args:
pretrained_model_name_or_path (`str` or `os.PathLike`):
This can be either:
- a string, the *model id* of a pretrained model configuration hosted inside a model repo on
huggingface.co. Valid model ids can be located at the root-level, like `bert-base-uncased`, or namespaced
under a user or organization name, like `dbmdz/bert-base-german-cased`.
- a path to a *directory* containing a configuration file saved using the
[`~PreTrainedTokenizer.save_pretrained`] method, e.g., `./my_model_directory/`.
module_file (`str`):
The name of the module file containing the class to look for.
cache_dir (`str` or `os.PathLike`, *optional*):
Path to a directory in which a downloaded pretrained model configuration should be cached if the standard
cache should not be used.
force_download (`bool`, *optional*, defaults to `False`):
Whether or not to force to (re-)download the configuration files and override the cached versions if they
exist.
resume_download (`bool`, *optional*, defaults to `False`):
Whether or not to delete incompletely received file. Attempts to resume the download if such a file exists.
proxies (`Dict[str, str]`, *optional*):
A dictionary of proxy servers to use by protocol or endpoint, e.g., `{'http': 'foo.bar:3128',
'http://hostname': 'foo.bar:4012'}.` The proxies are used on each request.
token (`str` or *bool*, *optional*):
The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated
when running `transformers-cli login` (stored in `~/.huggingface`).
revision (`str`, *optional*, defaults to `"main"`):
The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a
git-based system for storing models and other artifacts on huggingface.co, so `revision` can be any
identifier allowed by git.
local_files_only (`bool`, *optional*, defaults to `False`):
If `True`, will only try to load the tokenizer configuration from local files.
<Tip>
You may pass a token in `token` if you are not logged in (`huggingface-cli login`) and want to use private
or [gated models](https://huggingface.co/docs/hub/models-gated#gated-models).
</Tip>
Returns:
`str`: The path to the module inside the cache.
"""
# Download and cache module_file from the repo `pretrained_model_name_or_path` of grab it if it's a local file.
pretrained_model_name_or_path = str(pretrained_model_name_or_path)
module_file_or_url = os.path.join(pretrained_model_name_or_path, module_file)
if os.path.isfile(module_file_or_url):
resolved_module_file = module_file_or_url
submodule = "local"
elif pretrained_model_name_or_path.count("/") == 0:
available_versions = get_diffusers_versions()
# cut ".dev0"
latest_version = "v" + ".".join(__version__.split(".")[:3])
# retrieve github version that matches
if revision is None:
revision = latest_version if latest_version[1:] in available_versions else "main"
logger.info(f"Defaulting to latest_version: {revision}.")
elif revision in available_versions:
revision = f"v{revision}"
elif revision == "main":
revision = revision
else:
raise ValueError(
f"`custom_revision`: {revision} does not exist. Please make sure to choose one of"
f" {', '.join(available_versions + ['main'])}."
)
# community pipeline on GitHub
github_url = COMMUNITY_PIPELINES_URL.format(revision=revision, pipeline=pretrained_model_name_or_path)
try:
resolved_module_file = cached_download(
github_url,
cache_dir=cache_dir,
force_download=force_download,
proxies=proxies,
resume_download=resume_download,
local_files_only=local_files_only,
token=False,
)
submodule = "git"
module_file = pretrained_model_name_or_path + ".py"
except EnvironmentError:
logger.error(f"Could not locate the {module_file} inside {pretrained_model_name_or_path}.")
raise
else:
try:
# Load from URL or cache if already cached
resolved_module_file = hf_hub_download(
pretrained_model_name_or_path,
module_file,
cache_dir=cache_dir,
force_download=force_download,
proxies=proxies,
resume_download=resume_download,
local_files_only=local_files_only,
token=token,
)
submodule = os.path.join("local", "--".join(pretrained_model_name_or_path.split("/")))
except EnvironmentError:
logger.error(f"Could not locate the {module_file} inside {pretrained_model_name_or_path}.")
raise
# Check we have all the requirements in our environment
modules_needed = check_imports(resolved_module_file)
# Now we move the module inside our cached dynamic modules.
full_submodule = DIFFUSERS_DYNAMIC_MODULE_NAME + os.path.sep + submodule
create_dynamic_module(full_submodule)
submodule_path = Path(HF_MODULES_CACHE) / full_submodule
if submodule == "local" or submodule == "git":
# We always copy local files (we could hash the file to see if there was a change, and give them the name of
# that hash, to only copy when there is a modification but it seems overkill for now).
# The only reason we do the copy is to avoid putting too many folders in sys.path.
shutil.copy(resolved_module_file, submodule_path / module_file)
for module_needed in modules_needed:
module_needed = f"{module_needed}.py"
shutil.copy(os.path.join(pretrained_model_name_or_path, module_needed), submodule_path / module_needed)
else:
# Get the commit hash
# TODO: we will get this info in the etag soon, so retrieve it from there and not here.
commit_hash = model_info(pretrained_model_name_or_path, revision=revision, token=token).sha
# The module file will end up being placed in a subfolder with the git hash of the repo. This way we get the
# benefit of versioning.
submodule_path = submodule_path / commit_hash
full_submodule = full_submodule + os.path.sep + commit_hash
create_dynamic_module(full_submodule)
if not (submodule_path / module_file).exists():
shutil.copy(resolved_module_file, submodule_path / module_file)
# Make sure we also have every file with relative
for module_needed in modules_needed:
if not (submodule_path / module_needed).exists():
get_cached_module_file(
pretrained_model_name_or_path,
f"{module_needed}.py",
cache_dir=cache_dir,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
token=token,
revision=revision,
local_files_only=local_files_only,
)
return os.path.join(full_submodule, module_file)
@validate_hf_hub_args
def get_class_from_dynamic_module(
pretrained_model_name_or_path: Union[str, os.PathLike],
module_file: str,
class_name: Optional[str] = None,
cache_dir: Optional[Union[str, os.PathLike]] = None,
force_download: bool = False,
resume_download: bool = False,
proxies: Optional[Dict[str, str]] = None,
token: Optional[Union[bool, str]] = None,
revision: Optional[str] = None,
local_files_only: bool = False,
**kwargs,
):
"""
Extracts a class from a module file, present in the local folder or repository of a model.
<Tip warning={true}>
Calling this function will execute the code in the module file found locally or downloaded from the Hub. It should
therefore only be called on trusted repos.
</Tip>
Args:
pretrained_model_name_or_path (`str` or `os.PathLike`):
This can be either:
- a string, the *model id* of a pretrained model configuration hosted inside a model repo on
huggingface.co. Valid model ids can be located at the root-level, like `bert-base-uncased`, or namespaced
under a user or organization name, like `dbmdz/bert-base-german-cased`.
- a path to a *directory* containing a configuration file saved using the
[`~PreTrainedTokenizer.save_pretrained`] method, e.g., `./my_model_directory/`.
module_file (`str`):
The name of the module file containing the class to look for.
class_name (`str`):
The name of the class to import in the module.
cache_dir (`str` or `os.PathLike`, *optional*):
Path to a directory in which a downloaded pretrained model configuration should be cached if the standard
cache should not be used.
force_download (`bool`, *optional*, defaults to `False`):
Whether or not to force to (re-)download the configuration files and override the cached versions if they
exist.
resume_download (`bool`, *optional*, defaults to `False`):
Whether or not to delete incompletely received file. Attempts to resume the download if such a file exists.
proxies (`Dict[str, str]`, *optional*):
A dictionary of proxy servers to use by protocol or endpoint, e.g., `{'http': 'foo.bar:3128',
'http://hostname': 'foo.bar:4012'}.` The proxies are used on each request.
token (`str` or `bool`, *optional*):
The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated
when running `transformers-cli login` (stored in `~/.huggingface`).
revision (`str`, *optional*, defaults to `"main"`):
The specific model version to use. It can be a branch name, a tag name, or a commit id, since we use a
git-based system for storing models and other artifacts on huggingface.co, so `revision` can be any
identifier allowed by git.
local_files_only (`bool`, *optional*, defaults to `False`):
If `True`, will only try to load the tokenizer configuration from local files.
<Tip>
You may pass a token in `token` if you are not logged in (`huggingface-cli login`) and want to use private
or [gated models](https://huggingface.co/docs/hub/models-gated#gated-models).
</Tip>
Returns:
`type`: The class, dynamically imported from the module.
Examples:
```python
# Download module `modeling.py` from huggingface.co and cache then extract the class `MyBertModel` from this
# module.
cls = get_class_from_dynamic_module("sgugger/my-bert-model", "modeling.py", "MyBertModel")
```"""
# And lastly we get the class inside our newly created module
final_module = get_cached_module_file(
pretrained_model_name_or_path,
module_file,
cache_dir=cache_dir,
force_download=force_download,
resume_download=resume_download,
proxies=proxies,
token=token,
revision=revision,
local_files_only=local_files_only,
)
return get_class_in_module(class_name, final_module.replace(".py", ""))
| diffusers/src/diffusers/utils/dynamic_modules_utils.py/0 | {
"file_path": "diffusers/src/diffusers/utils/dynamic_modules_utils.py",
"repo_id": "diffusers",
"token_count": 7560
} | 126 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import inspect
import unittest
import torch
from parameterized import parameterized
from diffusers import PriorTransformer
from diffusers.utils.testing_utils import (
backend_empty_cache,
enable_full_determinism,
floats_tensor,
slow,
torch_all_close,
torch_device,
)
from ..test_modeling_common import ModelTesterMixin
enable_full_determinism()
class PriorTransformerTests(ModelTesterMixin, unittest.TestCase):
model_class = PriorTransformer
main_input_name = "hidden_states"
@property
def dummy_input(self):
batch_size = 4
embedding_dim = 8
num_embeddings = 7
hidden_states = floats_tensor((batch_size, embedding_dim)).to(torch_device)
proj_embedding = floats_tensor((batch_size, embedding_dim)).to(torch_device)
encoder_hidden_states = floats_tensor((batch_size, num_embeddings, embedding_dim)).to(torch_device)
return {
"hidden_states": hidden_states,
"timestep": 2,
"proj_embedding": proj_embedding,
"encoder_hidden_states": encoder_hidden_states,
}
def get_dummy_seed_input(self, seed=0):
torch.manual_seed(seed)
batch_size = 4
embedding_dim = 8
num_embeddings = 7
hidden_states = torch.randn((batch_size, embedding_dim)).to(torch_device)
proj_embedding = torch.randn((batch_size, embedding_dim)).to(torch_device)
encoder_hidden_states = torch.randn((batch_size, num_embeddings, embedding_dim)).to(torch_device)
return {
"hidden_states": hidden_states,
"timestep": 2,
"proj_embedding": proj_embedding,
"encoder_hidden_states": encoder_hidden_states,
}
@property
def input_shape(self):
return (4, 8)
@property
def output_shape(self):
return (4, 8)
def prepare_init_args_and_inputs_for_common(self):
init_dict = {
"num_attention_heads": 2,
"attention_head_dim": 4,
"num_layers": 2,
"embedding_dim": 8,
"num_embeddings": 7,
"additional_embeddings": 4,
}
inputs_dict = self.dummy_input
return init_dict, inputs_dict
def test_from_pretrained_hub(self):
model, loading_info = PriorTransformer.from_pretrained(
"hf-internal-testing/prior-dummy", output_loading_info=True
)
self.assertIsNotNone(model)
self.assertEqual(len(loading_info["missing_keys"]), 0)
model.to(torch_device)
hidden_states = model(**self.dummy_input)[0]
assert hidden_states is not None, "Make sure output is not None"
def test_forward_signature(self):
init_dict, _ = self.prepare_init_args_and_inputs_for_common()
model = self.model_class(**init_dict)
signature = inspect.signature(model.forward)
# signature.parameters is an OrderedDict => so arg_names order is deterministic
arg_names = [*signature.parameters.keys()]
expected_arg_names = ["hidden_states", "timestep"]
self.assertListEqual(arg_names[:2], expected_arg_names)
def test_output_pretrained(self):
model = PriorTransformer.from_pretrained("hf-internal-testing/prior-dummy")
model = model.to(torch_device)
if hasattr(model, "set_default_attn_processor"):
model.set_default_attn_processor()
input = self.get_dummy_seed_input()
with torch.no_grad():
output = model(**input)[0]
output_slice = output[0, :5].flatten().cpu()
print(output_slice)
# Since the VAE Gaussian prior's generator is seeded on the appropriate device,
# the expected output slices are not the same for CPU and GPU.
expected_output_slice = torch.tensor([-1.3436, -0.2870, 0.7538, 0.4368, -0.0239])
self.assertTrue(torch_all_close(output_slice, expected_output_slice, rtol=1e-2))
@slow
class PriorTransformerIntegrationTests(unittest.TestCase):
def get_dummy_seed_input(self, batch_size=1, embedding_dim=768, num_embeddings=77, seed=0):
torch.manual_seed(seed)
batch_size = batch_size
embedding_dim = embedding_dim
num_embeddings = num_embeddings
hidden_states = torch.randn((batch_size, embedding_dim)).to(torch_device)
proj_embedding = torch.randn((batch_size, embedding_dim)).to(torch_device)
encoder_hidden_states = torch.randn((batch_size, num_embeddings, embedding_dim)).to(torch_device)
return {
"hidden_states": hidden_states,
"timestep": 2,
"proj_embedding": proj_embedding,
"encoder_hidden_states": encoder_hidden_states,
}
def tearDown(self):
# clean up the VRAM after each test
super().tearDown()
gc.collect()
backend_empty_cache(torch_device)
@parameterized.expand(
[
# fmt: off
[13, [-0.5861, 0.1283, -0.0931, 0.0882, 0.4476, 0.1329, -0.0498, 0.0640]],
[37, [-0.4913, 0.0110, -0.0483, 0.0541, 0.4954, -0.0170, 0.0354, 0.1651]],
# fmt: on
]
)
def test_kandinsky_prior(self, seed, expected_slice):
model = PriorTransformer.from_pretrained("kandinsky-community/kandinsky-2-1-prior", subfolder="prior")
model.to(torch_device)
input = self.get_dummy_seed_input(seed=seed)
with torch.no_grad():
sample = model(**input)[0]
assert list(sample.shape) == [1, 768]
output_slice = sample[0, :8].flatten().cpu()
print(output_slice)
expected_output_slice = torch.tensor(expected_slice)
assert torch_all_close(output_slice, expected_output_slice, atol=1e-3)
| diffusers/tests/models/transformers/test_models_prior.py/0 | {
"file_path": "diffusers/tests/models/transformers/test_models_prior.py",
"repo_id": "diffusers",
"token_count": 2767
} | 127 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import unittest
import numpy as np
import torch
from transformers import (
ClapAudioConfig,
ClapConfig,
ClapFeatureExtractor,
ClapModel,
ClapTextConfig,
GPT2Config,
GPT2Model,
RobertaTokenizer,
SpeechT5HifiGan,
SpeechT5HifiGanConfig,
T5Config,
T5EncoderModel,
T5Tokenizer,
)
from diffusers import (
AudioLDM2Pipeline,
AudioLDM2ProjectionModel,
AudioLDM2UNet2DConditionModel,
AutoencoderKL,
DDIMScheduler,
LMSDiscreteScheduler,
PNDMScheduler,
)
from diffusers.utils.testing_utils import enable_full_determinism, nightly, torch_device
from ..pipeline_params import TEXT_TO_AUDIO_BATCH_PARAMS, TEXT_TO_AUDIO_PARAMS
from ..test_pipelines_common import PipelineTesterMixin
enable_full_determinism()
class AudioLDM2PipelineFastTests(PipelineTesterMixin, unittest.TestCase):
pipeline_class = AudioLDM2Pipeline
params = TEXT_TO_AUDIO_PARAMS
batch_params = TEXT_TO_AUDIO_BATCH_PARAMS
required_optional_params = frozenset(
[
"num_inference_steps",
"num_waveforms_per_prompt",
"generator",
"latents",
"output_type",
"return_dict",
"callback",
"callback_steps",
]
)
def get_dummy_components(self):
torch.manual_seed(0)
unet = AudioLDM2UNet2DConditionModel(
block_out_channels=(32, 64),
layers_per_block=2,
sample_size=32,
in_channels=4,
out_channels=4,
down_block_types=("DownBlock2D", "CrossAttnDownBlock2D"),
up_block_types=("CrossAttnUpBlock2D", "UpBlock2D"),
cross_attention_dim=([None, 16, 32], [None, 16, 32]),
)
scheduler = DDIMScheduler(
beta_start=0.00085,
beta_end=0.012,
beta_schedule="scaled_linear",
clip_sample=False,
set_alpha_to_one=False,
)
torch.manual_seed(0)
vae = AutoencoderKL(
block_out_channels=[32, 64],
in_channels=1,
out_channels=1,
down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"],
up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"],
latent_channels=4,
)
torch.manual_seed(0)
text_branch_config = ClapTextConfig(
bos_token_id=0,
eos_token_id=2,
hidden_size=16,
intermediate_size=37,
layer_norm_eps=1e-05,
num_attention_heads=2,
num_hidden_layers=2,
pad_token_id=1,
vocab_size=1000,
projection_dim=16,
)
audio_branch_config = ClapAudioConfig(
spec_size=64,
window_size=4,
num_mel_bins=64,
intermediate_size=37,
layer_norm_eps=1e-05,
depths=[2, 2],
num_attention_heads=[2, 2],
num_hidden_layers=2,
hidden_size=192,
projection_dim=16,
patch_size=2,
patch_stride=2,
patch_embed_input_channels=4,
)
text_encoder_config = ClapConfig.from_text_audio_configs(
text_config=text_branch_config, audio_config=audio_branch_config, projection_dim=16
)
text_encoder = ClapModel(text_encoder_config)
tokenizer = RobertaTokenizer.from_pretrained("hf-internal-testing/tiny-random-roberta", model_max_length=77)
feature_extractor = ClapFeatureExtractor.from_pretrained(
"hf-internal-testing/tiny-random-ClapModel", hop_length=7900
)
torch.manual_seed(0)
text_encoder_2_config = T5Config(
vocab_size=32100,
d_model=32,
d_ff=37,
d_kv=8,
num_heads=2,
num_layers=2,
)
text_encoder_2 = T5EncoderModel(text_encoder_2_config)
tokenizer_2 = T5Tokenizer.from_pretrained("hf-internal-testing/tiny-random-T5Model", model_max_length=77)
torch.manual_seed(0)
language_model_config = GPT2Config(
n_embd=16,
n_head=2,
n_layer=2,
vocab_size=1000,
n_ctx=99,
n_positions=99,
)
language_model = GPT2Model(language_model_config)
language_model.config.max_new_tokens = 8
torch.manual_seed(0)
projection_model = AudioLDM2ProjectionModel(text_encoder_dim=16, text_encoder_1_dim=32, langauge_model_dim=16)
vocoder_config = SpeechT5HifiGanConfig(
model_in_dim=8,
sampling_rate=16000,
upsample_initial_channel=16,
upsample_rates=[2, 2],
upsample_kernel_sizes=[4, 4],
resblock_kernel_sizes=[3, 7],
resblock_dilation_sizes=[[1, 3, 5], [1, 3, 5]],
normalize_before=False,
)
vocoder = SpeechT5HifiGan(vocoder_config)
components = {
"unet": unet,
"scheduler": scheduler,
"vae": vae,
"text_encoder": text_encoder,
"text_encoder_2": text_encoder_2,
"tokenizer": tokenizer,
"tokenizer_2": tokenizer_2,
"feature_extractor": feature_extractor,
"language_model": language_model,
"projection_model": projection_model,
"vocoder": vocoder,
}
return components
def get_dummy_inputs(self, device, seed=0):
if str(device).startswith("mps"):
generator = torch.manual_seed(seed)
else:
generator = torch.Generator(device=device).manual_seed(seed)
inputs = {
"prompt": "A hammer hitting a wooden surface",
"generator": generator,
"num_inference_steps": 2,
"guidance_scale": 6.0,
}
return inputs
def test_audioldm2_ddim(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
output = audioldm_pipe(**inputs)
audio = output.audios[0]
assert audio.ndim == 1
assert len(audio) == 256
audio_slice = audio[:10]
expected_slice = np.array(
[0.0025, 0.0018, 0.0018, -0.0023, -0.0026, -0.0020, -0.0026, -0.0021, -0.0027, -0.0020]
)
assert np.abs(audio_slice - expected_slice).max() < 1e-4
def test_audioldm2_prompt_embeds(self):
components = self.get_dummy_components()
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
inputs["prompt"] = 3 * [inputs["prompt"]]
# forward
output = audioldm_pipe(**inputs)
audio_1 = output.audios[0]
inputs = self.get_dummy_inputs(torch_device)
prompt = 3 * [inputs.pop("prompt")]
text_inputs = audioldm_pipe.tokenizer(
prompt,
padding="max_length",
max_length=audioldm_pipe.tokenizer.model_max_length,
truncation=True,
return_tensors="pt",
)
text_inputs = text_inputs["input_ids"].to(torch_device)
clap_prompt_embeds = audioldm_pipe.text_encoder.get_text_features(text_inputs)
clap_prompt_embeds = clap_prompt_embeds[:, None, :]
text_inputs = audioldm_pipe.tokenizer_2(
prompt,
padding="max_length",
max_length=True,
truncation=True,
return_tensors="pt",
)
text_inputs = text_inputs["input_ids"].to(torch_device)
t5_prompt_embeds = audioldm_pipe.text_encoder_2(
text_inputs,
)
t5_prompt_embeds = t5_prompt_embeds[0]
projection_embeds = audioldm_pipe.projection_model(clap_prompt_embeds, t5_prompt_embeds)[0]
generated_prompt_embeds = audioldm_pipe.generate_language_model(projection_embeds, max_new_tokens=8)
inputs["prompt_embeds"] = t5_prompt_embeds
inputs["generated_prompt_embeds"] = generated_prompt_embeds
# forward
output = audioldm_pipe(**inputs)
audio_2 = output.audios[0]
assert np.abs(audio_1 - audio_2).max() < 1e-2
def test_audioldm2_negative_prompt_embeds(self):
components = self.get_dummy_components()
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
negative_prompt = 3 * ["this is a negative prompt"]
inputs["negative_prompt"] = negative_prompt
inputs["prompt"] = 3 * [inputs["prompt"]]
# forward
output = audioldm_pipe(**inputs)
audio_1 = output.audios[0]
inputs = self.get_dummy_inputs(torch_device)
prompt = 3 * [inputs.pop("prompt")]
embeds = []
generated_embeds = []
for p in [prompt, negative_prompt]:
text_inputs = audioldm_pipe.tokenizer(
p,
padding="max_length",
max_length=audioldm_pipe.tokenizer.model_max_length,
truncation=True,
return_tensors="pt",
)
text_inputs = text_inputs["input_ids"].to(torch_device)
clap_prompt_embeds = audioldm_pipe.text_encoder.get_text_features(text_inputs)
clap_prompt_embeds = clap_prompt_embeds[:, None, :]
text_inputs = audioldm_pipe.tokenizer_2(
prompt,
padding="max_length",
max_length=True if len(embeds) == 0 else embeds[0].shape[1],
truncation=True,
return_tensors="pt",
)
text_inputs = text_inputs["input_ids"].to(torch_device)
t5_prompt_embeds = audioldm_pipe.text_encoder_2(
text_inputs,
)
t5_prompt_embeds = t5_prompt_embeds[0]
projection_embeds = audioldm_pipe.projection_model(clap_prompt_embeds, t5_prompt_embeds)[0]
generated_prompt_embeds = audioldm_pipe.generate_language_model(projection_embeds, max_new_tokens=8)
embeds.append(t5_prompt_embeds)
generated_embeds.append(generated_prompt_embeds)
inputs["prompt_embeds"], inputs["negative_prompt_embeds"] = embeds
inputs["generated_prompt_embeds"], inputs["negative_generated_prompt_embeds"] = generated_embeds
# forward
output = audioldm_pipe(**inputs)
audio_2 = output.audios[0]
assert np.abs(audio_1 - audio_2).max() < 1e-2
def test_audioldm2_negative_prompt(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
components["scheduler"] = PNDMScheduler(skip_prk_steps=True)
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
negative_prompt = "egg cracking"
output = audioldm_pipe(**inputs, negative_prompt=negative_prompt)
audio = output.audios[0]
assert audio.ndim == 1
assert len(audio) == 256
audio_slice = audio[:10]
expected_slice = np.array(
[0.0025, 0.0018, 0.0018, -0.0023, -0.0026, -0.0020, -0.0026, -0.0021, -0.0027, -0.0020]
)
assert np.abs(audio_slice - expected_slice).max() < 1e-4
def test_audioldm2_num_waveforms_per_prompt(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
components["scheduler"] = PNDMScheduler(skip_prk_steps=True)
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(device)
audioldm_pipe.set_progress_bar_config(disable=None)
prompt = "A hammer hitting a wooden surface"
# test num_waveforms_per_prompt=1 (default)
audios = audioldm_pipe(prompt, num_inference_steps=2).audios
assert audios.shape == (1, 256)
# test num_waveforms_per_prompt=1 (default) for batch of prompts
batch_size = 2
audios = audioldm_pipe([prompt] * batch_size, num_inference_steps=2).audios
assert audios.shape == (batch_size, 256)
# test num_waveforms_per_prompt for single prompt
num_waveforms_per_prompt = 2
audios = audioldm_pipe(prompt, num_inference_steps=2, num_waveforms_per_prompt=num_waveforms_per_prompt).audios
assert audios.shape == (num_waveforms_per_prompt, 256)
# test num_waveforms_per_prompt for batch of prompts
batch_size = 2
audios = audioldm_pipe(
[prompt] * batch_size, num_inference_steps=2, num_waveforms_per_prompt=num_waveforms_per_prompt
).audios
assert audios.shape == (batch_size * num_waveforms_per_prompt, 256)
def test_audioldm2_audio_length_in_s(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
vocoder_sampling_rate = audioldm_pipe.vocoder.config.sampling_rate
inputs = self.get_dummy_inputs(device)
output = audioldm_pipe(audio_length_in_s=0.016, **inputs)
audio = output.audios[0]
assert audio.ndim == 1
assert len(audio) / vocoder_sampling_rate == 0.016
output = audioldm_pipe(audio_length_in_s=0.032, **inputs)
audio = output.audios[0]
assert audio.ndim == 1
assert len(audio) / vocoder_sampling_rate == 0.032
def test_audioldm2_vocoder_model_in_dim(self):
components = self.get_dummy_components()
audioldm_pipe = AudioLDM2Pipeline(**components)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
prompt = ["hey"]
output = audioldm_pipe(prompt, num_inference_steps=1)
audio_shape = output.audios.shape
assert audio_shape == (1, 256)
config = audioldm_pipe.vocoder.config
config.model_in_dim *= 2
audioldm_pipe.vocoder = SpeechT5HifiGan(config).to(torch_device)
output = audioldm_pipe(prompt, num_inference_steps=1)
audio_shape = output.audios.shape
# waveform shape is unchanged, we just have 2x the number of mel channels in the spectrogram
assert audio_shape == (1, 256)
def test_attention_slicing_forward_pass(self):
self._test_attention_slicing_forward_pass(test_mean_pixel_difference=False)
@unittest.skip("Raises a not implemented error in AudioLDM2")
def test_xformers_attention_forwardGenerator_pass(self):
pass
def test_dict_tuple_outputs_equivalent(self):
# increase tolerance from 1e-4 -> 2e-4 to account for large composite model
super().test_dict_tuple_outputs_equivalent(expected_max_difference=2e-4)
def test_inference_batch_single_identical(self):
# increase tolerance from 1e-4 -> 2e-4 to account for large composite model
self._test_inference_batch_single_identical(expected_max_diff=2e-4)
def test_save_load_local(self):
# increase tolerance from 1e-4 -> 2e-4 to account for large composite model
super().test_save_load_local(expected_max_difference=2e-4)
def test_save_load_optional_components(self):
# increase tolerance from 1e-4 -> 2e-4 to account for large composite model
super().test_save_load_optional_components(expected_max_difference=2e-4)
def test_to_dtype(self):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.set_progress_bar_config(disable=None)
# The method component.dtype returns the dtype of the first parameter registered in the model, not the
# dtype of the entire model. In the case of CLAP, the first parameter is a float64 constant (logit scale)
model_dtypes = {key: component.dtype for key, component in components.items() if hasattr(component, "dtype")}
# Without the logit scale parameters, everything is float32
model_dtypes.pop("text_encoder")
self.assertTrue(all(dtype == torch.float32 for dtype in model_dtypes.values()))
# the CLAP sub-models are float32
model_dtypes["clap_text_branch"] = components["text_encoder"].text_model.dtype
self.assertTrue(all(dtype == torch.float32 for dtype in model_dtypes.values()))
# Once we send to fp16, all params are in half-precision, including the logit scale
pipe.to(torch_dtype=torch.float16)
model_dtypes = {key: component.dtype for key, component in components.items() if hasattr(component, "dtype")}
self.assertTrue(all(dtype == torch.float16 for dtype in model_dtypes.values()))
def test_sequential_cpu_offload_forward_pass(self):
pass
@nightly
class AudioLDM2PipelineSlowTests(unittest.TestCase):
def tearDown(self):
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def get_inputs(self, device, generator_device="cpu", dtype=torch.float32, seed=0):
generator = torch.Generator(device=generator_device).manual_seed(seed)
latents = np.random.RandomState(seed).standard_normal((1, 8, 128, 16))
latents = torch.from_numpy(latents).to(device=device, dtype=dtype)
inputs = {
"prompt": "A hammer hitting a wooden surface",
"latents": latents,
"generator": generator,
"num_inference_steps": 3,
"guidance_scale": 2.5,
}
return inputs
def test_audioldm2(self):
audioldm_pipe = AudioLDM2Pipeline.from_pretrained("cvssp/audioldm2")
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_inputs(torch_device)
inputs["num_inference_steps"] = 25
audio = audioldm_pipe(**inputs).audios[0]
assert audio.ndim == 1
assert len(audio) == 81952
# check the portion of the generated audio with the largest dynamic range (reduces flakiness)
audio_slice = audio[17275:17285]
expected_slice = np.array([0.0791, 0.0666, 0.1158, 0.1227, 0.1171, -0.2880, -0.1940, -0.0283, -0.0126, 0.1127])
max_diff = np.abs(expected_slice - audio_slice).max()
assert max_diff < 1e-3
def test_audioldm2_lms(self):
audioldm_pipe = AudioLDM2Pipeline.from_pretrained("cvssp/audioldm2")
audioldm_pipe.scheduler = LMSDiscreteScheduler.from_config(audioldm_pipe.scheduler.config)
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_inputs(torch_device)
audio = audioldm_pipe(**inputs).audios[0]
assert audio.ndim == 1
assert len(audio) == 81952
# check the portion of the generated audio with the largest dynamic range (reduces flakiness)
audio_slice = audio[31390:31400]
expected_slice = np.array(
[-0.1318, -0.0577, 0.0446, -0.0573, 0.0659, 0.1074, -0.2600, 0.0080, -0.2190, -0.4301]
)
max_diff = np.abs(expected_slice - audio_slice).max()
assert max_diff < 1e-3
def test_audioldm2_large(self):
audioldm_pipe = AudioLDM2Pipeline.from_pretrained("cvssp/audioldm2-large")
audioldm_pipe = audioldm_pipe.to(torch_device)
audioldm_pipe.set_progress_bar_config(disable=None)
inputs = self.get_inputs(torch_device)
audio = audioldm_pipe(**inputs).audios[0]
assert audio.ndim == 1
assert len(audio) == 81952
# check the portion of the generated audio with the largest dynamic range (reduces flakiness)
audio_slice = audio[8825:8835]
expected_slice = np.array(
[-0.1829, -0.1461, 0.0759, -0.1493, -0.1396, 0.5783, 0.3001, -0.3038, -0.0639, -0.2244]
)
max_diff = np.abs(expected_slice - audio_slice).max()
assert max_diff < 1e-3
| diffusers/tests/pipelines/audioldm2/test_audioldm2.py/0 | {
"file_path": "diffusers/tests/pipelines/audioldm2/test_audioldm2.py",
"repo_id": "diffusers",
"token_count": 10092
} | 128 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import AutoTokenizer, T5EncoderModel
from diffusers import (
AutoPipelineForImage2Image,
AutoPipelineForText2Image,
Kandinsky3Pipeline,
Kandinsky3UNet,
VQModel,
)
from diffusers.image_processor import VaeImageProcessor
from diffusers.schedulers.scheduling_ddpm import DDPMScheduler
from diffusers.utils.testing_utils import (
enable_full_determinism,
load_image,
require_torch_gpu,
slow,
)
from ..pipeline_params import (
TEXT_TO_IMAGE_BATCH_PARAMS,
TEXT_TO_IMAGE_CALLBACK_CFG_PARAMS,
TEXT_TO_IMAGE_IMAGE_PARAMS,
TEXT_TO_IMAGE_PARAMS,
)
from ..test_pipelines_common import PipelineTesterMixin
enable_full_determinism()
class Kandinsky3PipelineFastTests(PipelineTesterMixin, unittest.TestCase):
pipeline_class = Kandinsky3Pipeline
params = TEXT_TO_IMAGE_PARAMS - {"cross_attention_kwargs"}
batch_params = TEXT_TO_IMAGE_BATCH_PARAMS
image_params = TEXT_TO_IMAGE_IMAGE_PARAMS
image_latents_params = TEXT_TO_IMAGE_IMAGE_PARAMS
callback_cfg_params = TEXT_TO_IMAGE_CALLBACK_CFG_PARAMS
test_xformers_attention = False
@property
def dummy_movq_kwargs(self):
return {
"block_out_channels": [32, 64],
"down_block_types": ["DownEncoderBlock2D", "AttnDownEncoderBlock2D"],
"in_channels": 3,
"latent_channels": 4,
"layers_per_block": 1,
"norm_num_groups": 8,
"norm_type": "spatial",
"num_vq_embeddings": 12,
"out_channels": 3,
"up_block_types": [
"AttnUpDecoderBlock2D",
"UpDecoderBlock2D",
],
"vq_embed_dim": 4,
}
@property
def dummy_movq(self):
torch.manual_seed(0)
model = VQModel(**self.dummy_movq_kwargs)
return model
def get_dummy_components(self, time_cond_proj_dim=None):
torch.manual_seed(0)
unet = Kandinsky3UNet(
in_channels=4,
time_embedding_dim=4,
groups=2,
attention_head_dim=4,
layers_per_block=3,
block_out_channels=(32, 64),
cross_attention_dim=4,
encoder_hid_dim=32,
)
scheduler = DDPMScheduler(
beta_start=0.00085,
beta_end=0.012,
steps_offset=1,
beta_schedule="squaredcos_cap_v2",
clip_sample=True,
thresholding=False,
)
torch.manual_seed(0)
movq = self.dummy_movq
torch.manual_seed(0)
text_encoder = T5EncoderModel.from_pretrained("hf-internal-testing/tiny-random-t5")
torch.manual_seed(0)
tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-t5")
components = {
"unet": unet,
"scheduler": scheduler,
"movq": movq,
"text_encoder": text_encoder,
"tokenizer": tokenizer,
}
return components
def get_dummy_inputs(self, device, seed=0):
if str(device).startswith("mps"):
generator = torch.manual_seed(seed)
else:
generator = torch.Generator(device=device).manual_seed(seed)
inputs = {
"prompt": "A painting of a squirrel eating a burger",
"generator": generator,
"num_inference_steps": 2,
"guidance_scale": 6.0,
"output_type": "np",
"width": 16,
"height": 16,
}
return inputs
def test_kandinsky3(self):
device = "cpu"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe = pipe.to(device)
pipe.set_progress_bar_config(disable=None)
output = pipe(**self.get_dummy_inputs(device))
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 16, 16, 3)
expected_slice = np.array([0.3768, 0.4373, 0.4865, 0.4890, 0.4299, 0.5122, 0.4921, 0.4924, 0.5599])
assert (
np.abs(image_slice.flatten() - expected_slice).max() < 1e-2
), f" expected_slice {expected_slice}, but got {image_slice.flatten()}"
def test_float16_inference(self):
super().test_float16_inference(expected_max_diff=1e-1)
def test_inference_batch_single_identical(self):
super().test_inference_batch_single_identical(expected_max_diff=1e-2)
@slow
@require_torch_gpu
class Kandinsky3PipelineIntegrationTests(unittest.TestCase):
def tearDown(self):
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def test_kandinskyV3(self):
pipe = AutoPipelineForText2Image.from_pretrained(
"kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16
)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=None)
prompt = "A photograph of the inside of a subway train. There are raccoons sitting on the seats. One of them is reading a newspaper. The window shows the city in the background."
generator = torch.Generator(device="cpu").manual_seed(0)
image = pipe(prompt, num_inference_steps=25, generator=generator).images[0]
assert image.size == (1024, 1024)
expected_image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/t2i.png"
)
image_processor = VaeImageProcessor()
image_np = image_processor.pil_to_numpy(image)
expected_image_np = image_processor.pil_to_numpy(expected_image)
self.assertTrue(np.allclose(image_np, expected_image_np, atol=5e-2))
def test_kandinskyV3_img2img(self):
pipe = AutoPipelineForImage2Image.from_pretrained(
"kandinsky-community/kandinsky-3", variant="fp16", torch_dtype=torch.float16
)
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=None)
generator = torch.Generator(device="cpu").manual_seed(0)
image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/t2i.png"
)
w, h = 512, 512
image = image.resize((w, h), resample=Image.BICUBIC, reducing_gap=1)
prompt = "A painting of the inside of a subway train with tiny raccoons."
image = pipe(prompt, image=image, strength=0.75, num_inference_steps=25, generator=generator).images[0]
assert image.size == (512, 512)
expected_image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky3/i2i.png"
)
image_processor = VaeImageProcessor()
image_np = image_processor.pil_to_numpy(image)
expected_image_np = image_processor.pil_to_numpy(expected_image)
self.assertTrue(np.allclose(image_np, expected_image_np, atol=5e-2))
| diffusers/tests/pipelines/kandinsky3/test_kandinsky3.py/0 | {
"file_path": "diffusers/tests/pipelines/kandinsky3/test_kandinsky3.py",
"repo_id": "diffusers",
"token_count": 3517
} | 129 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import tempfile
import unittest
import numpy as np
import torch
from transformers import AutoTokenizer, T5EncoderModel
from diffusers import (
AutoencoderKL,
DDIMScheduler,
PixArtAlphaPipeline,
Transformer2DModel,
)
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu, slow, torch_device
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineTesterMixin, to_np
enable_full_determinism()
class PixArtAlphaPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
pipeline_class = PixArtAlphaPipeline
params = TEXT_TO_IMAGE_PARAMS - {"cross_attention_kwargs"}
batch_params = TEXT_TO_IMAGE_BATCH_PARAMS
image_params = TEXT_TO_IMAGE_IMAGE_PARAMS
image_latents_params = TEXT_TO_IMAGE_IMAGE_PARAMS
required_optional_params = PipelineTesterMixin.required_optional_params
def get_dummy_components(self):
torch.manual_seed(0)
transformer = Transformer2DModel(
sample_size=8,
num_layers=2,
patch_size=2,
attention_head_dim=8,
num_attention_heads=3,
caption_channels=32,
in_channels=4,
cross_attention_dim=24,
out_channels=8,
attention_bias=True,
activation_fn="gelu-approximate",
num_embeds_ada_norm=1000,
norm_type="ada_norm_single",
norm_elementwise_affine=False,
norm_eps=1e-6,
)
torch.manual_seed(0)
vae = AutoencoderKL()
scheduler = DDIMScheduler()
text_encoder = T5EncoderModel.from_pretrained("hf-internal-testing/tiny-random-t5")
tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-t5")
components = {
"transformer": transformer.eval(),
"vae": vae.eval(),
"scheduler": scheduler,
"text_encoder": text_encoder,
"tokenizer": tokenizer,
}
return components
def get_dummy_inputs(self, device, seed=0):
if str(device).startswith("mps"):
generator = torch.manual_seed(seed)
else:
generator = torch.Generator(device=device).manual_seed(seed)
inputs = {
"prompt": "A painting of a squirrel eating a burger",
"generator": generator,
"num_inference_steps": 2,
"guidance_scale": 5.0,
"use_resolution_binning": False,
"output_type": "np",
}
return inputs
def test_sequential_cpu_offload_forward_pass(self):
# TODO(PVP, Sayak) need to fix later
return
def test_save_load_optional_components(self):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
prompt = inputs["prompt"]
generator = inputs["generator"]
num_inference_steps = inputs["num_inference_steps"]
output_type = inputs["output_type"]
(
prompt_embeds,
prompt_attention_mask,
negative_prompt_embeds,
negative_prompt_attention_mask,
) = pipe.encode_prompt(prompt)
# inputs with prompt converted to embeddings
inputs = {
"prompt_embeds": prompt_embeds,
"prompt_attention_mask": prompt_attention_mask,
"negative_prompt": None,
"negative_prompt_embeds": negative_prompt_embeds,
"negative_prompt_attention_mask": negative_prompt_attention_mask,
"generator": generator,
"num_inference_steps": num_inference_steps,
"output_type": output_type,
"use_resolution_binning": False,
}
# set all optional components to None
for optional_component in pipe._optional_components:
setattr(pipe, optional_component, None)
output = pipe(**inputs)[0]
with tempfile.TemporaryDirectory() as tmpdir:
pipe.save_pretrained(tmpdir)
pipe_loaded = self.pipeline_class.from_pretrained(tmpdir)
pipe_loaded.to(torch_device)
pipe_loaded.set_progress_bar_config(disable=None)
for optional_component in pipe._optional_components:
self.assertTrue(
getattr(pipe_loaded, optional_component) is None,
f"`{optional_component}` did not stay set to None after loading.",
)
inputs = self.get_dummy_inputs(torch_device)
generator = inputs["generator"]
num_inference_steps = inputs["num_inference_steps"]
output_type = inputs["output_type"]
# inputs with prompt converted to embeddings
inputs = {
"prompt_embeds": prompt_embeds,
"prompt_attention_mask": prompt_attention_mask,
"negative_prompt": None,
"negative_prompt_embeds": negative_prompt_embeds,
"negative_prompt_attention_mask": negative_prompt_attention_mask,
"generator": generator,
"num_inference_steps": num_inference_steps,
"output_type": output_type,
"use_resolution_binning": False,
}
output_loaded = pipe_loaded(**inputs)[0]
max_diff = np.abs(to_np(output) - to_np(output_loaded)).max()
self.assertLess(max_diff, 1e-4)
def test_inference(self):
device = "cpu"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
self.assertEqual(image.shape, (1, 8, 8, 3))
expected_slice = np.array([0.6319, 0.3526, 0.3806, 0.6327, 0.4639, 0.483, 0.2583, 0.5331, 0.4852])
max_diff = np.abs(image_slice.flatten() - expected_slice).max()
self.assertLessEqual(max_diff, 1e-3)
def test_inference_non_square_images(self):
device = "cpu"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = pipe(**inputs, height=32, width=48).images
image_slice = image[0, -3:, -3:, -1]
self.assertEqual(image.shape, (1, 32, 48, 3))
expected_slice = np.array([0.6493, 0.537, 0.4081, 0.4762, 0.3695, 0.4711, 0.3026, 0.5218, 0.5263])
max_diff = np.abs(image_slice.flatten() - expected_slice).max()
self.assertLessEqual(max_diff, 1e-3)
def test_inference_with_embeddings_and_multiple_images(self):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
prompt = inputs["prompt"]
generator = inputs["generator"]
num_inference_steps = inputs["num_inference_steps"]
output_type = inputs["output_type"]
prompt_embeds, prompt_attn_mask, negative_prompt_embeds, neg_prompt_attn_mask = pipe.encode_prompt(prompt)
# inputs with prompt converted to embeddings
inputs = {
"prompt_embeds": prompt_embeds,
"prompt_attention_mask": prompt_attn_mask,
"negative_prompt": None,
"negative_prompt_embeds": negative_prompt_embeds,
"negative_prompt_attention_mask": neg_prompt_attn_mask,
"generator": generator,
"num_inference_steps": num_inference_steps,
"output_type": output_type,
"num_images_per_prompt": 2,
"use_resolution_binning": False,
}
# set all optional components to None
for optional_component in pipe._optional_components:
setattr(pipe, optional_component, None)
output = pipe(**inputs)[0]
with tempfile.TemporaryDirectory() as tmpdir:
pipe.save_pretrained(tmpdir)
pipe_loaded = self.pipeline_class.from_pretrained(tmpdir)
pipe_loaded.to(torch_device)
pipe_loaded.set_progress_bar_config(disable=None)
for optional_component in pipe._optional_components:
self.assertTrue(
getattr(pipe_loaded, optional_component) is None,
f"`{optional_component}` did not stay set to None after loading.",
)
inputs = self.get_dummy_inputs(torch_device)
generator = inputs["generator"]
num_inference_steps = inputs["num_inference_steps"]
output_type = inputs["output_type"]
# inputs with prompt converted to embeddings
inputs = {
"prompt_embeds": prompt_embeds,
"prompt_attention_mask": prompt_attn_mask,
"negative_prompt": None,
"negative_prompt_embeds": negative_prompt_embeds,
"negative_prompt_attention_mask": neg_prompt_attn_mask,
"generator": generator,
"num_inference_steps": num_inference_steps,
"output_type": output_type,
"num_images_per_prompt": 2,
"use_resolution_binning": False,
}
output_loaded = pipe_loaded(**inputs)[0]
max_diff = np.abs(to_np(output) - to_np(output_loaded)).max()
self.assertLess(max_diff, 1e-4)
def test_inference_with_multiple_images_per_prompt(self):
device = "cpu"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
inputs["num_images_per_prompt"] = 2
image = pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
self.assertEqual(image.shape, (2, 8, 8, 3))
expected_slice = np.array([0.6319, 0.3526, 0.3806, 0.6327, 0.4639, 0.483, 0.2583, 0.5331, 0.4852])
max_diff = np.abs(image_slice.flatten() - expected_slice).max()
self.assertLessEqual(max_diff, 1e-3)
def test_raises_warning_for_mask_feature(self):
device = "cpu"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
inputs.update({"mask_feature": True})
with self.assertWarns(FutureWarning) as warning_ctx:
_ = pipe(**inputs).images
assert "mask_feature" in str(warning_ctx.warning)
def test_inference_batch_single_identical(self):
self._test_inference_batch_single_identical(expected_max_diff=1e-3)
@slow
@require_torch_gpu
class PixArtAlphaPipelineIntegrationTests(unittest.TestCase):
ckpt_id_1024 = "PixArt-alpha/PixArt-XL-2-1024-MS"
ckpt_id_512 = "PixArt-alpha/PixArt-XL-2-512x512"
prompt = "A small cactus with a happy face in the Sahara desert."
def tearDown(self):
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def test_pixart_1024(self):
generator = torch.manual_seed(0)
pipe = PixArtAlphaPipeline.from_pretrained(self.ckpt_id_1024, torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = self.prompt
image = pipe(prompt, generator=generator, output_type="np").images
image_slice = image[0, -3:, -3:, -1]
expected_slice = np.array([0.1941, 0.2117, 0.2188, 0.1946, 0.218, 0.2124, 0.199, 0.2437, 0.2583])
max_diff = np.abs(image_slice.flatten() - expected_slice).max()
self.assertLessEqual(max_diff, 1e-3)
def test_pixart_512(self):
generator = torch.manual_seed(0)
pipe = PixArtAlphaPipeline.from_pretrained(self.ckpt_id_512, torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = self.prompt
image = pipe(prompt, generator=generator, output_type="np").images
image_slice = image[0, -3:, -3:, -1]
expected_slice = np.array([0.2637, 0.291, 0.2939, 0.207, 0.2512, 0.2783, 0.2168, 0.2324, 0.2817])
max_diff = np.abs(image_slice.flatten() - expected_slice).max()
self.assertLessEqual(max_diff, 1e-3)
def test_pixart_1024_without_resolution_binning(self):
generator = torch.manual_seed(0)
pipe = PixArtAlphaPipeline.from_pretrained(self.ckpt_id_1024, torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = self.prompt
height, width = 1024, 768
num_inference_steps = 10
image = pipe(
prompt,
height=height,
width=width,
generator=generator,
num_inference_steps=num_inference_steps,
output_type="np",
).images
image_slice = image[0, -3:, -3:, -1]
generator = torch.manual_seed(0)
no_res_bin_image = pipe(
prompt,
height=height,
width=width,
generator=generator,
num_inference_steps=num_inference_steps,
output_type="np",
use_resolution_binning=False,
).images
no_res_bin_image_slice = no_res_bin_image[0, -3:, -3:, -1]
assert not np.allclose(image_slice, no_res_bin_image_slice, atol=1e-4, rtol=1e-4)
def test_pixart_512_without_resolution_binning(self):
generator = torch.manual_seed(0)
pipe = PixArtAlphaPipeline.from_pretrained(self.ckpt_id_512, torch_dtype=torch.float16)
pipe.enable_model_cpu_offload()
prompt = self.prompt
height, width = 512, 768
num_inference_steps = 10
image = pipe(
prompt,
height=height,
width=width,
generator=generator,
num_inference_steps=num_inference_steps,
output_type="np",
).images
image_slice = image[0, -3:, -3:, -1]
generator = torch.manual_seed(0)
no_res_bin_image = pipe(
prompt,
height=height,
width=width,
generator=generator,
num_inference_steps=num_inference_steps,
output_type="np",
use_resolution_binning=False,
).images
no_res_bin_image_slice = no_res_bin_image[0, -3:, -3:, -1]
assert not np.allclose(image_slice, no_res_bin_image_slice, atol=1e-4, rtol=1e-4)
| diffusers/tests/pipelines/pixart/test_pixart.py/0 | {
"file_path": "diffusers/tests/pipelines/pixart/test_pixart.py",
"repo_id": "diffusers",
"token_count": 7103
} | 130 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import random
import unittest
import numpy as np
import torch
from PIL import Image
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
EulerAncestralDiscreteScheduler,
LMSDiscreteScheduler,
PNDMScheduler,
StableDiffusionInstructPix2PixPipeline,
UNet2DConditionModel,
)
from diffusers.image_processor import VaeImageProcessor
from diffusers.utils.testing_utils import (
enable_full_determinism,
floats_tensor,
load_image,
require_torch_gpu,
slow,
torch_device,
)
from ..pipeline_params import (
IMAGE_TO_IMAGE_IMAGE_PARAMS,
TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS,
TEXT_GUIDED_IMAGE_VARIATION_PARAMS,
TEXT_TO_IMAGE_CALLBACK_CFG_PARAMS,
)
from ..test_pipelines_common import PipelineKarrasSchedulerTesterMixin, PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class StableDiffusionInstructPix2PixPipelineFastTests(
PipelineLatentTesterMixin, PipelineKarrasSchedulerTesterMixin, PipelineTesterMixin, unittest.TestCase
):
pipeline_class = StableDiffusionInstructPix2PixPipeline
params = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {"height", "width", "cross_attention_kwargs"}
batch_params = TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS
image_params = IMAGE_TO_IMAGE_IMAGE_PARAMS
image_latents_params = IMAGE_TO_IMAGE_IMAGE_PARAMS
callback_cfg_params = TEXT_TO_IMAGE_CALLBACK_CFG_PARAMS.union({"image_latents"}) - {"negative_prompt_embeds"}
def get_dummy_components(self):
torch.manual_seed(0)
unet = UNet2DConditionModel(
block_out_channels=(32, 64),
layers_per_block=2,
sample_size=32,
in_channels=8,
out_channels=4,
down_block_types=("DownBlock2D", "CrossAttnDownBlock2D"),
up_block_types=("CrossAttnUpBlock2D", "UpBlock2D"),
cross_attention_dim=32,
)
scheduler = PNDMScheduler(skip_prk_steps=True)
torch.manual_seed(0)
vae = AutoencoderKL(
block_out_channels=[32, 64],
in_channels=3,
out_channels=3,
down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"],
up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"],
latent_channels=4,
)
torch.manual_seed(0)
text_encoder_config = CLIPTextConfig(
bos_token_id=0,
eos_token_id=2,
hidden_size=32,
intermediate_size=37,
layer_norm_eps=1e-05,
num_attention_heads=4,
num_hidden_layers=5,
pad_token_id=1,
vocab_size=1000,
)
text_encoder = CLIPTextModel(text_encoder_config)
tokenizer = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
components = {
"unet": unet,
"scheduler": scheduler,
"vae": vae,
"text_encoder": text_encoder,
"tokenizer": tokenizer,
"safety_checker": None,
"feature_extractor": None,
"image_encoder": None,
}
return components
def get_dummy_inputs(self, device, seed=0):
image = floats_tensor((1, 3, 32, 32), rng=random.Random(seed)).to(device)
image = image.cpu().permute(0, 2, 3, 1)[0]
image = Image.fromarray(np.uint8(image)).convert("RGB")
if str(device).startswith("mps"):
generator = torch.manual_seed(seed)
else:
generator = torch.Generator(device=device).manual_seed(seed)
inputs = {
"prompt": "A painting of a squirrel eating a burger",
"image": image,
"generator": generator,
"num_inference_steps": 2,
"guidance_scale": 6.0,
"image_guidance_scale": 1,
"output_type": "numpy",
}
return inputs
def test_stable_diffusion_pix2pix_default_case(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
sd_pipe = StableDiffusionInstructPix2PixPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = sd_pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 32, 32, 3)
expected_slice = np.array([0.7526, 0.3750, 0.4547, 0.6117, 0.5866, 0.5016, 0.4327, 0.5642, 0.4815])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_negative_prompt(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
sd_pipe = StableDiffusionInstructPix2PixPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
negative_prompt = "french fries"
output = sd_pipe(**inputs, negative_prompt=negative_prompt)
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 32, 32, 3)
expected_slice = np.array([0.7511, 0.3642, 0.4553, 0.6236, 0.5797, 0.5013, 0.4343, 0.5611, 0.4831])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_multiple_init_images(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
sd_pipe = StableDiffusionInstructPix2PixPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
inputs["prompt"] = [inputs["prompt"]] * 2
image = np.array(inputs["image"]).astype(np.float32) / 255.0
image = torch.from_numpy(image).unsqueeze(0).to(device)
image = image / 2 + 0.5
image = image.permute(0, 3, 1, 2)
inputs["image"] = image.repeat(2, 1, 1, 1)
image = sd_pipe(**inputs).images
image_slice = image[-1, -3:, -3:, -1]
assert image.shape == (2, 32, 32, 3)
expected_slice = np.array([0.5812, 0.5748, 0.5222, 0.5908, 0.5695, 0.7174, 0.6804, 0.5523, 0.5579])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_euler(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
components["scheduler"] = EulerAncestralDiscreteScheduler(
beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear"
)
sd_pipe = StableDiffusionInstructPix2PixPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = sd_pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
slice = [round(x, 4) for x in image_slice.flatten().tolist()]
print(",".join([str(x) for x in slice]))
assert image.shape == (1, 32, 32, 3)
expected_slice = np.array([0.7417, 0.3842, 0.4732, 0.5776, 0.5891, 0.5139, 0.4052, 0.5673, 0.4986])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-3
def test_inference_batch_single_identical(self):
super().test_inference_batch_single_identical(expected_max_diff=3e-3)
# Overwrite the default test_latents_inputs because pix2pix encode the image differently
def test_latents_input(self):
components = self.get_dummy_components()
pipe = StableDiffusionInstructPix2PixPipeline(**components)
pipe.image_processor = VaeImageProcessor(do_resize=False, do_normalize=False)
pipe = pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
out = pipe(**self.get_dummy_inputs_by_type(torch_device, input_image_type="pt"))[0]
vae = components["vae"]
inputs = self.get_dummy_inputs_by_type(torch_device, input_image_type="pt")
for image_param in self.image_latents_params:
if image_param in inputs.keys():
inputs[image_param] = vae.encode(inputs[image_param]).latent_dist.mode()
out_latents_inputs = pipe(**inputs)[0]
max_diff = np.abs(out - out_latents_inputs).max()
self.assertLess(max_diff, 1e-4, "passing latents as image input generate different result from passing image")
# Override the default test_callback_cfg because pix2pix create inputs for cfg differently
def test_callback_cfg(self):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe = pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
def callback_no_cfg(pipe, i, t, callback_kwargs):
if i == 1:
for k, w in callback_kwargs.items():
if k in self.callback_cfg_params:
callback_kwargs[k] = callback_kwargs[k].chunk(3)[0]
pipe._guidance_scale = 1.0
return callback_kwargs
inputs = self.get_dummy_inputs(torch_device)
inputs["guidance_scale"] = 1.0
inputs["num_inference_steps"] = 2
out_no_cfg = pipe(**inputs)[0]
inputs["guidance_scale"] = 7.5
inputs["callback_on_step_end"] = callback_no_cfg
inputs["callback_on_step_end_tensor_inputs"] = pipe._callback_tensor_inputs
out_callback_no_cfg = pipe(**inputs)[0]
assert out_no_cfg.shape == out_callback_no_cfg.shape
@slow
@require_torch_gpu
class StableDiffusionInstructPix2PixPipelineSlowTests(unittest.TestCase):
def tearDown(self):
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def get_inputs(self, seed=0):
generator = torch.manual_seed(seed)
image = load_image(
"https://huggingface.co/datasets/diffusers/test-arrays/resolve/main/stable_diffusion_pix2pix/example.jpg"
)
inputs = {
"prompt": "turn him into a cyborg",
"image": image,
"generator": generator,
"num_inference_steps": 3,
"guidance_scale": 7.5,
"image_guidance_scale": 1.0,
"output_type": "numpy",
}
return inputs
def test_stable_diffusion_pix2pix_default(self):
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
"timbrooks/instruct-pix2pix", safety_checker=None
)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing()
inputs = self.get_inputs()
image = pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1].flatten()
assert image.shape == (1, 512, 512, 3)
expected_slice = np.array([0.5902, 0.6015, 0.6027, 0.5983, 0.6092, 0.6061, 0.5765, 0.5785, 0.5555])
assert np.abs(expected_slice - image_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_k_lms(self):
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
"timbrooks/instruct-pix2pix", safety_checker=None
)
pipe.scheduler = LMSDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing()
inputs = self.get_inputs()
image = pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1].flatten()
assert image.shape == (1, 512, 512, 3)
expected_slice = np.array([0.6578, 0.6817, 0.6972, 0.6761, 0.6856, 0.6916, 0.6428, 0.6516, 0.6301])
assert np.abs(expected_slice - image_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_ddim(self):
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
"timbrooks/instruct-pix2pix", safety_checker=None
)
pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing()
inputs = self.get_inputs()
image = pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1].flatten()
assert image.shape == (1, 512, 512, 3)
expected_slice = np.array([0.3828, 0.3834, 0.3818, 0.3792, 0.3865, 0.3752, 0.3792, 0.3847, 0.3753])
assert np.abs(expected_slice - image_slice).max() < 1e-3
def test_stable_diffusion_pix2pix_intermediate_state(self):
number_of_steps = 0
def callback_fn(step: int, timestep: int, latents: torch.FloatTensor) -> None:
callback_fn.has_been_called = True
nonlocal number_of_steps
number_of_steps += 1
if step == 1:
latents = latents.detach().cpu().numpy()
assert latents.shape == (1, 4, 64, 64)
latents_slice = latents[0, -3:, -3:, -1]
expected_slice = np.array([-0.2463, -0.4644, -0.9756, 1.5176, 1.4414, 0.7866, 0.9897, 0.8521, 0.7983])
assert np.abs(latents_slice.flatten() - expected_slice).max() < 5e-2
elif step == 2:
latents = latents.detach().cpu().numpy()
assert latents.shape == (1, 4, 64, 64)
latents_slice = latents[0, -3:, -3:, -1]
expected_slice = np.array([-0.2644, -0.4626, -0.9653, 1.5176, 1.4551, 0.7686, 0.9805, 0.8452, 0.8115])
assert np.abs(latents_slice.flatten() - expected_slice).max() < 5e-2
callback_fn.has_been_called = False
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
"timbrooks/instruct-pix2pix", safety_checker=None, torch_dtype=torch.float16
)
pipe = pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing()
inputs = self.get_inputs()
pipe(**inputs, callback=callback_fn, callback_steps=1)
assert callback_fn.has_been_called
assert number_of_steps == 3
def test_stable_diffusion_pipeline_with_sequential_cpu_offloading(self):
torch.cuda.empty_cache()
torch.cuda.reset_max_memory_allocated()
torch.cuda.reset_peak_memory_stats()
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
"timbrooks/instruct-pix2pix", safety_checker=None, torch_dtype=torch.float16
)
pipe = pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing(1)
pipe.enable_sequential_cpu_offload()
inputs = self.get_inputs()
_ = pipe(**inputs)
mem_bytes = torch.cuda.max_memory_allocated()
# make sure that less than 2.2 GB is allocated
assert mem_bytes < 2.2 * 10**9
def test_stable_diffusion_pix2pix_pipeline_multiple_of_8(self):
inputs = self.get_inputs()
# resize to resolution that is divisible by 8 but not 16 or 32
inputs["image"] = inputs["image"].resize((504, 504))
model_id = "timbrooks/instruct-pix2pix"
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
model_id,
safety_checker=None,
)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
pipe.enable_attention_slicing()
output = pipe(**inputs)
image = output.images[0]
image_slice = image[255:258, 383:386, -1]
assert image.shape == (504, 504, 3)
expected_slice = np.array([0.2726, 0.2529, 0.2664, 0.2655, 0.2641, 0.2642, 0.2591, 0.2649, 0.2590])
assert np.abs(image_slice.flatten() - expected_slice).max() < 5e-3
| diffusers/tests/pipelines/stable_diffusion/test_stable_diffusion_instruction_pix2pix.py/0 | {
"file_path": "diffusers/tests/pipelines/stable_diffusion/test_stable_diffusion_instruction_pix2pix.py",
"repo_id": "diffusers",
"token_count": 7736
} | 131 |
# coding=utf-8
# Copyright 2023 HuggingFace Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import gc
import random
import unittest
import numpy as np
import torch
from parameterized import parameterized
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTextModelWithProjection, CLIPTokenizer
import diffusers
from diffusers import (
AutoencoderKL,
EulerDiscreteScheduler,
LCMScheduler,
MultiAdapter,
StableDiffusionXLAdapterPipeline,
T2IAdapter,
UNet2DConditionModel,
)
from diffusers.utils import load_image, logging
from diffusers.utils.testing_utils import (
enable_full_determinism,
floats_tensor,
numpy_cosine_similarity_distance,
require_torch_gpu,
slow,
torch_device,
)
from ..pipeline_params import TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS, TEXT_GUIDED_IMAGE_VARIATION_PARAMS
from ..test_pipelines_common import (
PipelineTesterMixin,
SDXLOptionalComponentsTesterMixin,
assert_mean_pixel_difference,
)
enable_full_determinism()
class StableDiffusionXLAdapterPipelineFastTests(
PipelineTesterMixin, SDXLOptionalComponentsTesterMixin, unittest.TestCase
):
pipeline_class = StableDiffusionXLAdapterPipeline
params = TEXT_GUIDED_IMAGE_VARIATION_PARAMS
batch_params = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
def get_dummy_components(self, adapter_type="full_adapter_xl", time_cond_proj_dim=None):
torch.manual_seed(0)
unet = UNet2DConditionModel(
block_out_channels=(32, 64),
layers_per_block=2,
sample_size=32,
in_channels=4,
out_channels=4,
down_block_types=("DownBlock2D", "CrossAttnDownBlock2D"),
up_block_types=("CrossAttnUpBlock2D", "UpBlock2D"),
# SD2-specific config below
attention_head_dim=(2, 4),
use_linear_projection=True,
addition_embed_type="text_time",
addition_time_embed_dim=8,
transformer_layers_per_block=(1, 2),
projection_class_embeddings_input_dim=80, # 6 * 8 + 32
cross_attention_dim=64,
time_cond_proj_dim=time_cond_proj_dim,
)
scheduler = EulerDiscreteScheduler(
beta_start=0.00085,
beta_end=0.012,
steps_offset=1,
beta_schedule="scaled_linear",
timestep_spacing="leading",
)
torch.manual_seed(0)
vae = AutoencoderKL(
block_out_channels=[32, 64],
in_channels=3,
out_channels=3,
down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"],
up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"],
latent_channels=4,
sample_size=128,
)
torch.manual_seed(0)
text_encoder_config = CLIPTextConfig(
bos_token_id=0,
eos_token_id=2,
hidden_size=32,
intermediate_size=37,
layer_norm_eps=1e-05,
num_attention_heads=4,
num_hidden_layers=5,
pad_token_id=1,
vocab_size=1000,
# SD2-specific config below
hidden_act="gelu",
projection_dim=32,
)
text_encoder = CLIPTextModel(text_encoder_config)
tokenizer = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
text_encoder_2 = CLIPTextModelWithProjection(text_encoder_config)
tokenizer_2 = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
if adapter_type == "full_adapter_xl":
adapter = T2IAdapter(
in_channels=3,
channels=[32, 64],
num_res_blocks=2,
downscale_factor=4,
adapter_type=adapter_type,
)
elif adapter_type == "multi_adapter":
adapter = MultiAdapter(
[
T2IAdapter(
in_channels=3,
channels=[32, 64],
num_res_blocks=2,
downscale_factor=4,
adapter_type="full_adapter_xl",
),
T2IAdapter(
in_channels=3,
channels=[32, 64],
num_res_blocks=2,
downscale_factor=4,
adapter_type="full_adapter_xl",
),
]
)
else:
raise ValueError(
f"Unknown adapter type: {adapter_type}, must be one of 'full_adapter_xl', or 'multi_adapter''"
)
components = {
"adapter": adapter,
"unet": unet,
"scheduler": scheduler,
"vae": vae,
"text_encoder": text_encoder,
"tokenizer": tokenizer,
"text_encoder_2": text_encoder_2,
"tokenizer_2": tokenizer_2,
# "safety_checker": None,
"feature_extractor": None,
"image_encoder": None,
}
return components
def get_dummy_components_with_full_downscaling(self, adapter_type="full_adapter_xl"):
"""Get dummy components with x8 VAE downscaling and 3 UNet down blocks.
These dummy components are intended to fully-exercise the T2I-Adapter
downscaling behavior.
"""
torch.manual_seed(0)
unet = UNet2DConditionModel(
block_out_channels=(32, 32, 64),
layers_per_block=2,
sample_size=32,
in_channels=4,
out_channels=4,
down_block_types=("DownBlock2D", "CrossAttnDownBlock2D", "CrossAttnDownBlock2D"),
up_block_types=("CrossAttnUpBlock2D", "CrossAttnUpBlock2D", "UpBlock2D"),
# SD2-specific config below
attention_head_dim=2,
use_linear_projection=True,
addition_embed_type="text_time",
addition_time_embed_dim=8,
transformer_layers_per_block=1,
projection_class_embeddings_input_dim=80, # 6 * 8 + 32
cross_attention_dim=64,
)
scheduler = EulerDiscreteScheduler(
beta_start=0.00085,
beta_end=0.012,
steps_offset=1,
beta_schedule="scaled_linear",
timestep_spacing="leading",
)
torch.manual_seed(0)
vae = AutoencoderKL(
block_out_channels=[32, 32, 32, 64],
in_channels=3,
out_channels=3,
down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D", "DownEncoderBlock2D", "DownEncoderBlock2D"],
up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D", "UpDecoderBlock2D", "UpDecoderBlock2D"],
latent_channels=4,
sample_size=128,
)
torch.manual_seed(0)
text_encoder_config = CLIPTextConfig(
bos_token_id=0,
eos_token_id=2,
hidden_size=32,
intermediate_size=37,
layer_norm_eps=1e-05,
num_attention_heads=4,
num_hidden_layers=5,
pad_token_id=1,
vocab_size=1000,
# SD2-specific config below
hidden_act="gelu",
projection_dim=32,
)
text_encoder = CLIPTextModel(text_encoder_config)
tokenizer = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
text_encoder_2 = CLIPTextModelWithProjection(text_encoder_config)
tokenizer_2 = CLIPTokenizer.from_pretrained("hf-internal-testing/tiny-random-clip")
if adapter_type == "full_adapter_xl":
adapter = T2IAdapter(
in_channels=3,
channels=[32, 32, 64],
num_res_blocks=2,
downscale_factor=16,
adapter_type=adapter_type,
)
elif adapter_type == "multi_adapter":
adapter = MultiAdapter(
[
T2IAdapter(
in_channels=3,
channels=[32, 32, 64],
num_res_blocks=2,
downscale_factor=16,
adapter_type="full_adapter_xl",
),
T2IAdapter(
in_channels=3,
channels=[32, 32, 64],
num_res_blocks=2,
downscale_factor=16,
adapter_type="full_adapter_xl",
),
]
)
else:
raise ValueError(
f"Unknown adapter type: {adapter_type}, must be one of 'full_adapter_xl', or 'multi_adapter''"
)
components = {
"adapter": adapter,
"unet": unet,
"scheduler": scheduler,
"vae": vae,
"text_encoder": text_encoder,
"tokenizer": tokenizer,
"text_encoder_2": text_encoder_2,
"tokenizer_2": tokenizer_2,
# "safety_checker": None,
"feature_extractor": None,
"image_encoder": None,
}
return components
def get_dummy_inputs(self, device, seed=0, height=64, width=64, num_images=1):
if num_images == 1:
image = floats_tensor((1, 3, height, width), rng=random.Random(seed)).to(device)
else:
image = [
floats_tensor((1, 3, height, width), rng=random.Random(seed)).to(device) for _ in range(num_images)
]
if str(device).startswith("mps"):
generator = torch.manual_seed(seed)
else:
generator = torch.Generator(device=device).manual_seed(seed)
inputs = {
"prompt": "A painting of a squirrel eating a burger",
"image": image,
"generator": generator,
"num_inference_steps": 2,
"guidance_scale": 5.0,
"output_type": "numpy",
}
return inputs
def test_stable_diffusion_adapter_default_case(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = sd_pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array(
[0.5752919, 0.6022097, 0.4728038, 0.49861962, 0.57084894, 0.4644975, 0.5193715, 0.5133664, 0.4729858]
)
assert np.abs(image_slice.flatten() - expected_slice).max() < 5e-3
@parameterized.expand(
[
# (dim=144) The internal feature map will be 9x9 after initial pixel unshuffling (downscaled x16).
(((4 * 2 + 1) * 16),),
# (dim=160) The internal feature map will be 5x5 after the first T2I down block (downscaled x32).
(((4 * 1 + 1) * 32),),
]
)
def test_multiple_image_dimensions(self, dim):
"""Test that the T2I-Adapter pipeline supports any input dimension that
is divisible by the adapter's `downscale_factor`. This test was added in
response to an issue where the T2I Adapter's downscaling padding
behavior did not match the UNet's behavior.
Note that we have selected `dim` values to produce odd resolutions at
each downscaling level.
"""
components = self.get_dummy_components_with_full_downscaling()
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe = sd_pipe.to(torch_device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device, height=dim, width=dim)
image = sd_pipe(**inputs).images
assert image.shape == (1, dim, dim, 3)
@parameterized.expand(["full_adapter", "full_adapter_xl", "light_adapter"])
def test_total_downscale_factor(self, adapter_type):
"""Test that the T2IAdapter correctly reports its total_downscale_factor."""
batch_size = 1
in_channels = 3
out_channels = [320, 640, 1280, 1280]
in_image_size = 512
adapter = T2IAdapter(
in_channels=in_channels,
channels=out_channels,
num_res_blocks=2,
downscale_factor=8,
adapter_type=adapter_type,
)
adapter.to(torch_device)
in_image = floats_tensor((batch_size, in_channels, in_image_size, in_image_size)).to(torch_device)
adapter_state = adapter(in_image)
# Assume that the last element in `adapter_state` has been downsampled the most, and check
# that it matches the `total_downscale_factor`.
expected_out_image_size = in_image_size // adapter.total_downscale_factor
assert adapter_state[-1].shape == (
batch_size,
out_channels[-1],
expected_out_image_size,
expected_out_image_size,
)
def test_save_load_optional_components(self):
return self._test_save_load_optional_components()
def test_adapter_sdxl_lcm(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components(time_cond_proj_dim=256)
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe.scheduler = LCMScheduler.from_config(sd_pipe.scheduler.config)
sd_pipe = sd_pipe.to(torch_device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
output = sd_pipe(**inputs)
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array([0.5425, 0.5385, 0.4964, 0.5045, 0.6149, 0.4974, 0.5469, 0.5332, 0.5426])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-2
def test_adapter_sdxl_lcm_custom_timesteps(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components(time_cond_proj_dim=256)
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe.scheduler = LCMScheduler.from_config(sd_pipe.scheduler.config)
sd_pipe = sd_pipe.to(torch_device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
del inputs["num_inference_steps"]
inputs["timesteps"] = [999, 499]
output = sd_pipe(**inputs)
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array([0.5425, 0.5385, 0.4964, 0.5045, 0.6149, 0.4974, 0.5469, 0.5332, 0.5426])
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-2
class StableDiffusionXLMultiAdapterPipelineFastTests(
StableDiffusionXLAdapterPipelineFastTests, PipelineTesterMixin, unittest.TestCase
):
def get_dummy_components(self, time_cond_proj_dim=None):
return super().get_dummy_components("multi_adapter", time_cond_proj_dim=time_cond_proj_dim)
def get_dummy_components_with_full_downscaling(self):
return super().get_dummy_components_with_full_downscaling("multi_adapter")
def get_dummy_inputs(self, device, seed=0, height=64, width=64):
inputs = super().get_dummy_inputs(device, seed, height, width, num_images=2)
inputs["adapter_conditioning_scale"] = [0.5, 0.5]
return inputs
def test_stable_diffusion_adapter_default_case(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components()
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe = sd_pipe.to(device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
image = sd_pipe(**inputs).images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array(
[0.5813032, 0.60995954, 0.47563356, 0.5056669, 0.57199144, 0.4631841, 0.5176794, 0.51252556, 0.47183886]
)
assert np.abs(image_slice.flatten() - expected_slice).max() < 5e-3
def test_inference_batch_consistent(
self, batch_sizes=[2, 4, 13], additional_params_copy_to_batched_inputs=["num_inference_steps"]
):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
logger = logging.get_logger(pipe.__module__)
logger.setLevel(level=diffusers.logging.FATAL)
# batchify inputs
for batch_size in batch_sizes:
batched_inputs = {}
for name, value in inputs.items():
if name in self.batch_params:
# prompt is string
if name == "prompt":
len_prompt = len(value)
# make unequal batch sizes
batched_inputs[name] = [value[: len_prompt // i] for i in range(1, batch_size + 1)]
# make last batch super long
batched_inputs[name][-1] = 100 * "very long"
elif name == "image":
batched_images = []
for image in value:
batched_images.append(batch_size * [image])
batched_inputs[name] = batched_images
else:
batched_inputs[name] = batch_size * [value]
elif name == "batch_size":
batched_inputs[name] = batch_size
else:
batched_inputs[name] = value
for arg in additional_params_copy_to_batched_inputs:
batched_inputs[arg] = inputs[arg]
batched_inputs["output_type"] = "np"
output = pipe(**batched_inputs)
assert len(output[0]) == batch_size
batched_inputs["output_type"] = "np"
output = pipe(**batched_inputs)[0]
assert output.shape[0] == batch_size
logger.setLevel(level=diffusers.logging.WARNING)
def test_num_images_per_prompt(self):
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe = pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
batch_sizes = [1, 2]
num_images_per_prompts = [1, 2]
for batch_size in batch_sizes:
for num_images_per_prompt in num_images_per_prompts:
inputs = self.get_dummy_inputs(torch_device)
for key in inputs.keys():
if key in self.batch_params:
if key == "image":
batched_images = []
for image in inputs[key]:
batched_images.append(batch_size * [image])
inputs[key] = batched_images
else:
inputs[key] = batch_size * [inputs[key]]
images = pipe(**inputs, num_images_per_prompt=num_images_per_prompt)[0]
assert images.shape[0] == batch_size * num_images_per_prompt
def test_inference_batch_single_identical(
self,
batch_size=3,
test_max_difference=None,
test_mean_pixel_difference=None,
relax_max_difference=False,
expected_max_diff=2e-3,
additional_params_copy_to_batched_inputs=["num_inference_steps"],
):
if test_max_difference is None:
# TODO(Pedro) - not sure why, but not at all reproducible at the moment it seems
# make sure that batched and non-batched is identical
test_max_difference = torch_device != "mps"
if test_mean_pixel_difference is None:
# TODO same as above
test_mean_pixel_difference = torch_device != "mps"
components = self.get_dummy_components()
pipe = self.pipeline_class(**components)
pipe.to(torch_device)
pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(torch_device)
logger = logging.get_logger(pipe.__module__)
logger.setLevel(level=diffusers.logging.FATAL)
# batchify inputs
batched_inputs = {}
batch_size = batch_size
for name, value in inputs.items():
if name in self.batch_params:
# prompt is string
if name == "prompt":
len_prompt = len(value)
# make unequal batch sizes
batched_inputs[name] = [value[: len_prompt // i] for i in range(1, batch_size + 1)]
# make last batch super long
batched_inputs[name][-1] = 100 * "very long"
elif name == "image":
batched_images = []
for image in value:
batched_images.append(batch_size * [image])
batched_inputs[name] = batched_images
else:
batched_inputs[name] = batch_size * [value]
elif name == "batch_size":
batched_inputs[name] = batch_size
elif name == "generator":
batched_inputs[name] = [self.get_generator(i) for i in range(batch_size)]
else:
batched_inputs[name] = value
for arg in additional_params_copy_to_batched_inputs:
batched_inputs[arg] = inputs[arg]
output_batch = pipe(**batched_inputs)
assert output_batch[0].shape[0] == batch_size
inputs["generator"] = self.get_generator(0)
output = pipe(**inputs)
logger.setLevel(level=diffusers.logging.WARNING)
if test_max_difference:
if relax_max_difference:
# Taking the median of the largest <n> differences
# is resilient to outliers
diff = np.abs(output_batch[0][0] - output[0][0])
diff = diff.flatten()
diff.sort()
max_diff = np.median(diff[-5:])
else:
max_diff = np.abs(output_batch[0][0] - output[0][0]).max()
assert max_diff < expected_max_diff
if test_mean_pixel_difference:
assert_mean_pixel_difference(output_batch[0][0], output[0][0])
def test_adapter_sdxl_lcm(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components(time_cond_proj_dim=256)
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe.scheduler = LCMScheduler.from_config(sd_pipe.scheduler.config)
sd_pipe = sd_pipe.to(torch_device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
output = sd_pipe(**inputs)
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array([0.5313, 0.5375, 0.4942, 0.5021, 0.6142, 0.4968, 0.5434, 0.5311, 0.5448])
debug = [str(round(i, 4)) for i in image_slice.flatten().tolist()]
print(",".join(debug))
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-2
def test_adapter_sdxl_lcm_custom_timesteps(self):
device = "cpu" # ensure determinism for the device-dependent torch.Generator
components = self.get_dummy_components(time_cond_proj_dim=256)
sd_pipe = StableDiffusionXLAdapterPipeline(**components)
sd_pipe.scheduler = LCMScheduler.from_config(sd_pipe.scheduler.config)
sd_pipe = sd_pipe.to(torch_device)
sd_pipe.set_progress_bar_config(disable=None)
inputs = self.get_dummy_inputs(device)
del inputs["num_inference_steps"]
inputs["timesteps"] = [999, 499]
output = sd_pipe(**inputs)
image = output.images
image_slice = image[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
expected_slice = np.array([0.5313, 0.5375, 0.4942, 0.5021, 0.6142, 0.4968, 0.5434, 0.5311, 0.5448])
debug = [str(round(i, 4)) for i in image_slice.flatten().tolist()]
print(",".join(debug))
assert np.abs(image_slice.flatten() - expected_slice).max() < 1e-2
@slow
@require_torch_gpu
class AdapterSDXLPipelineSlowTests(unittest.TestCase):
def tearDown(self):
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def test_canny_lora(self):
adapter = T2IAdapter.from_pretrained("TencentARC/t2i-adapter-lineart-sdxl-1.0", torch_dtype=torch.float16).to(
"cpu"
)
pipe = StableDiffusionXLAdapterPipeline.from_pretrained(
"stabilityai/stable-diffusion-xl-base-1.0",
adapter=adapter,
torch_dtype=torch.float16,
variant="fp16",
)
pipe.load_lora_weights("CiroN2022/toy-face", weight_name="toy_face_sdxl.safetensors")
pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=None)
generator = torch.Generator(device="cpu").manual_seed(0)
prompt = "toy"
image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/t2i_adapter/toy_canny.png"
)
images = pipe(prompt, image=image, generator=generator, output_type="np", num_inference_steps=3).images
assert images[0].shape == (768, 512, 3)
image_slice = images[0, -3:, -3:, -1].flatten()
expected_slice = np.array([0.4284, 0.4337, 0.4319, 0.4255, 0.4329, 0.4280, 0.4338, 0.4420, 0.4226])
assert numpy_cosine_similarity_distance(image_slice, expected_slice) < 1e-4
def test_download_ckpt_diff_format_is_same(self):
ckpt_path = (
"https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/blob/main/sd_xl_base_1.0.safetensors"
)
adapter = T2IAdapter.from_pretrained("TencentARC/t2i-adapter-lineart-sdxl-1.0", torch_dtype=torch.float16)
prompt = "toy"
image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/t2i_adapter/toy_canny.png"
)
pipe_single_file = StableDiffusionXLAdapterPipeline.from_single_file(
ckpt_path,
adapter=adapter,
torch_dtype=torch.float16,
)
pipe_single_file.enable_model_cpu_offload()
pipe_single_file.set_progress_bar_config(disable=None)
generator = torch.Generator(device="cpu").manual_seed(0)
images_single_file = pipe_single_file(
prompt, image=image, generator=generator, output_type="np", num_inference_steps=3
).images
generator = torch.Generator(device="cpu").manual_seed(0)
pipe = StableDiffusionXLAdapterPipeline.from_pretrained(
"stabilityai/stable-diffusion-xl-base-1.0",
adapter=adapter,
torch_dtype=torch.float16,
)
pipe.enable_model_cpu_offload()
images = pipe(prompt, image=image, generator=generator, output_type="np", num_inference_steps=3).images
assert images_single_file[0].shape == (768, 512, 3)
assert images[0].shape == (768, 512, 3)
max_diff = numpy_cosine_similarity_distance(images[0].flatten(), images_single_file[0].flatten())
assert max_diff < 5e-3
| diffusers/tests/pipelines/stable_diffusion_xl/test_stable_diffusion_xl_adapter.py/0 | {
"file_path": "diffusers/tests/pipelines/stable_diffusion_xl/test_stable_diffusion_xl_adapter.py",
"repo_id": "diffusers",
"token_count": 14125
} | 132 |
from diffusers.utils.testing_utils import require_onnxruntime
@require_onnxruntime
class OnnxPipelineTesterMixin:
"""
This mixin is designed to be used with unittest.TestCase classes.
It provides a set of common tests for each ONNXRuntime pipeline, e.g. saving and loading the pipeline,
equivalence of dict and tuple outputs, etc.
"""
pass
| diffusers/tests/pipelines/test_pipelines_onnx_common.py/0 | {
"file_path": "diffusers/tests/pipelines/test_pipelines_onnx_common.py",
"repo_id": "diffusers",
"token_count": 118
} | 133 |
import torch
from diffusers import CMStochasticIterativeScheduler
from .test_schedulers import SchedulerCommonTest
class CMStochasticIterativeSchedulerTest(SchedulerCommonTest):
scheduler_classes = (CMStochasticIterativeScheduler,)
num_inference_steps = 10
def get_scheduler_config(self, **kwargs):
config = {
"num_train_timesteps": 201,
"sigma_min": 0.002,
"sigma_max": 80.0,
}
config.update(**kwargs)
return config
# Override test_step_shape to add CMStochasticIterativeScheduler-specific logic regarding timesteps
# Problem is that we don't know two timesteps that will always be in the timestep schedule from only the scheduler
# config; scaled sigma_max is always in the timestep schedule, but sigma_min is in the sigma schedule while scaled
# sigma_min is not in the timestep schedule
def test_step_shape(self):
num_inference_steps = 10
scheduler_config = self.get_scheduler_config()
scheduler = self.scheduler_classes[0](**scheduler_config)
scheduler.set_timesteps(num_inference_steps)
timestep_0 = scheduler.timesteps[0]
timestep_1 = scheduler.timesteps[1]
sample = self.dummy_sample
residual = 0.1 * sample
output_0 = scheduler.step(residual, timestep_0, sample).prev_sample
output_1 = scheduler.step(residual, timestep_1, sample).prev_sample
self.assertEqual(output_0.shape, sample.shape)
self.assertEqual(output_0.shape, output_1.shape)
def test_timesteps(self):
for timesteps in [10, 50, 100, 1000]:
self.check_over_configs(num_train_timesteps=timesteps)
def test_clip_denoised(self):
for clip_denoised in [True, False]:
self.check_over_configs(clip_denoised=clip_denoised)
def test_full_loop_no_noise_onestep(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
num_inference_steps = 1
scheduler.set_timesteps(num_inference_steps)
timesteps = scheduler.timesteps
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
for i, t in enumerate(timesteps):
# 1. scale model input
scaled_sample = scheduler.scale_model_input(sample, t)
# 2. predict noise residual
residual = model(scaled_sample, t)
# 3. predict previous sample x_t-1
pred_prev_sample = scheduler.step(residual, t, sample, generator=generator).prev_sample
sample = pred_prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 192.7614) < 1e-2
assert abs(result_mean.item() - 0.2510) < 1e-3
def test_full_loop_no_noise_multistep(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
timesteps = [106, 0]
scheduler.set_timesteps(timesteps=timesteps)
timesteps = scheduler.timesteps
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
for t in timesteps:
# 1. scale model input
scaled_sample = scheduler.scale_model_input(sample, t)
# 2. predict noise residual
residual = model(scaled_sample, t)
# 3. predict previous sample x_t-1
pred_prev_sample = scheduler.step(residual, t, sample, generator=generator).prev_sample
sample = pred_prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 347.6357) < 1e-2
assert abs(result_mean.item() - 0.4527) < 1e-3
def test_full_loop_with_noise(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
num_inference_steps = 10
t_start = 8
scheduler.set_timesteps(num_inference_steps)
timesteps = scheduler.timesteps
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
noise = self.dummy_noise_deter
timesteps = scheduler.timesteps[t_start * scheduler.order :]
sample = scheduler.add_noise(sample, noise, timesteps[:1])
for t in timesteps:
# 1. scale model input
scaled_sample = scheduler.scale_model_input(sample, t)
# 2. predict noise residual
residual = model(scaled_sample, t)
# 3. predict previous sample x_t-1
pred_prev_sample = scheduler.step(residual, t, sample, generator=generator).prev_sample
sample = pred_prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 763.9186) < 1e-2, f" expected result sum 763.9186, but get {result_sum}"
assert abs(result_mean.item() - 0.9947) < 1e-3, f" expected result mean 0.9947, but get {result_mean}"
def test_custom_timesteps_increasing_order(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
timesteps = [39, 30, 12, 15, 0]
with self.assertRaises(ValueError, msg="`timesteps` must be in descending order."):
scheduler.set_timesteps(timesteps=timesteps)
def test_custom_timesteps_passing_both_num_inference_steps_and_timesteps(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
timesteps = [39, 30, 12, 1, 0]
num_inference_steps = len(timesteps)
with self.assertRaises(ValueError, msg="Can only pass one of `num_inference_steps` or `timesteps`."):
scheduler.set_timesteps(num_inference_steps=num_inference_steps, timesteps=timesteps)
def test_custom_timesteps_too_large(self):
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
timesteps = [scheduler.config.num_train_timesteps]
with self.assertRaises(
ValueError,
msg="`timesteps` must start before `self.config.train_timesteps`: {scheduler.config.num_train_timesteps}}",
):
scheduler.set_timesteps(timesteps=timesteps)
| diffusers/tests/schedulers/test_scheduler_consistency_model.py/0 | {
"file_path": "diffusers/tests/schedulers/test_scheduler_consistency_model.py",
"repo_id": "diffusers",
"token_count": 3029
} | 134 |
import torch
from diffusers import KDPM2AncestralDiscreteScheduler
from diffusers.utils.testing_utils import torch_device
from .test_schedulers import SchedulerCommonTest
class KDPM2AncestralDiscreteSchedulerTest(SchedulerCommonTest):
scheduler_classes = (KDPM2AncestralDiscreteScheduler,)
num_inference_steps = 10
def get_scheduler_config(self, **kwargs):
config = {
"num_train_timesteps": 1100,
"beta_start": 0.0001,
"beta_end": 0.02,
"beta_schedule": "linear",
}
config.update(**kwargs)
return config
def test_timesteps(self):
for timesteps in [10, 50, 100, 1000]:
self.check_over_configs(num_train_timesteps=timesteps)
def test_betas(self):
for beta_start, beta_end in zip([0.00001, 0.0001, 0.001], [0.0002, 0.002, 0.02]):
self.check_over_configs(beta_start=beta_start, beta_end=beta_end)
def test_schedules(self):
for schedule in ["linear", "scaled_linear"]:
self.check_over_configs(beta_schedule=schedule)
def test_full_loop_no_noise(self):
if torch_device == "mps":
return
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
scheduler.set_timesteps(self.num_inference_steps)
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
sample = sample.to(torch_device)
for i, t in enumerate(scheduler.timesteps):
sample = scheduler.scale_model_input(sample, t)
model_output = model(sample, t)
output = scheduler.step(model_output, t, sample, generator=generator)
sample = output.prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 13849.3877) < 1e-2
assert abs(result_mean.item() - 18.0331) < 5e-3
def test_prediction_type(self):
for prediction_type in ["epsilon", "v_prediction"]:
self.check_over_configs(prediction_type=prediction_type)
def test_full_loop_with_v_prediction(self):
if torch_device == "mps":
return
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config(prediction_type="v_prediction")
scheduler = scheduler_class(**scheduler_config)
scheduler.set_timesteps(self.num_inference_steps)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
sample = sample.to(torch_device)
generator = torch.manual_seed(0)
for i, t in enumerate(scheduler.timesteps):
sample = scheduler.scale_model_input(sample, t)
model_output = model(sample, t)
output = scheduler.step(model_output, t, sample, generator=generator)
sample = output.prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 328.9970) < 1e-2
assert abs(result_mean.item() - 0.4284) < 1e-3
def test_full_loop_device(self):
if torch_device == "mps":
return
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
scheduler.set_timesteps(self.num_inference_steps, device=torch_device)
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter.to(torch_device) * scheduler.init_noise_sigma
for t in scheduler.timesteps:
sample = scheduler.scale_model_input(sample, t)
model_output = model(sample, t)
output = scheduler.step(model_output, t, sample, generator=generator)
sample = output.prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 13849.3818) < 1e-1
assert abs(result_mean.item() - 18.0331) < 1e-3
def test_full_loop_with_noise(self):
if torch_device == "mps":
return
scheduler_class = self.scheduler_classes[0]
scheduler_config = self.get_scheduler_config()
scheduler = scheduler_class(**scheduler_config)
scheduler.set_timesteps(self.num_inference_steps)
generator = torch.manual_seed(0)
model = self.dummy_model()
sample = self.dummy_sample_deter * scheduler.init_noise_sigma
# add noise
t_start = self.num_inference_steps - 2
noise = self.dummy_noise_deter
noise = noise.to(sample.device)
timesteps = scheduler.timesteps[t_start * scheduler.order :]
sample = scheduler.add_noise(sample, noise, timesteps[:1])
for i, t in enumerate(timesteps):
sample = scheduler.scale_model_input(sample, t)
model_output = model(sample, t)
output = scheduler.step(model_output, t, sample, generator=generator)
sample = output.prev_sample
result_sum = torch.sum(torch.abs(sample))
result_mean = torch.mean(torch.abs(sample))
assert abs(result_sum.item() - 93087.0312) < 1e-2, f" expected result sum 93087.0312, but get {result_sum}"
assert abs(result_mean.item() - 121.2071) < 5e-3, f" expected result mean 121.2071, but get {result_mean}"
| diffusers/tests/schedulers/test_scheduler_kdpm2_ancestral.py/0 | {
"file_path": "diffusers/tests/schedulers/test_scheduler_kdpm2_ancestral.py",
"repo_id": "diffusers",
"token_count": 2516
} | 135 |
# coding=utf-8
# Copyright 2023 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import importlib
import inspect
import os
import re
import warnings
from collections import OrderedDict
from difflib import get_close_matches
from pathlib import Path
from diffusers.models.auto import get_values
from diffusers.utils import ENV_VARS_TRUE_VALUES, is_flax_available, is_torch_available
# All paths are set with the intent you should run this script from the root of the repo with the command
# python utils/check_repo.py
PATH_TO_DIFFUSERS = "src/diffusers"
PATH_TO_TESTS = "tests"
PATH_TO_DOC = "docs/source/en"
# Update this list with models that are supposed to be private.
PRIVATE_MODELS = [
"DPRSpanPredictor",
"RealmBertModel",
"T5Stack",
"TFDPRSpanPredictor",
]
# Update this list for models that are not tested with a comment explaining the reason it should not be.
# Being in this list is an exception and should **not** be the rule.
IGNORE_NON_TESTED = PRIVATE_MODELS.copy() + [
# models to ignore for not tested
"OPTDecoder", # Building part of bigger (tested) model.
"DecisionTransformerGPT2Model", # Building part of bigger (tested) model.
"SegformerDecodeHead", # Building part of bigger (tested) model.
"PLBartEncoder", # Building part of bigger (tested) model.
"PLBartDecoder", # Building part of bigger (tested) model.
"PLBartDecoderWrapper", # Building part of bigger (tested) model.
"BigBirdPegasusEncoder", # Building part of bigger (tested) model.
"BigBirdPegasusDecoder", # Building part of bigger (tested) model.
"BigBirdPegasusDecoderWrapper", # Building part of bigger (tested) model.
"DetrEncoder", # Building part of bigger (tested) model.
"DetrDecoder", # Building part of bigger (tested) model.
"DetrDecoderWrapper", # Building part of bigger (tested) model.
"M2M100Encoder", # Building part of bigger (tested) model.
"M2M100Decoder", # Building part of bigger (tested) model.
"Speech2TextEncoder", # Building part of bigger (tested) model.
"Speech2TextDecoder", # Building part of bigger (tested) model.
"LEDEncoder", # Building part of bigger (tested) model.
"LEDDecoder", # Building part of bigger (tested) model.
"BartDecoderWrapper", # Building part of bigger (tested) model.
"BartEncoder", # Building part of bigger (tested) model.
"BertLMHeadModel", # Needs to be setup as decoder.
"BlenderbotSmallEncoder", # Building part of bigger (tested) model.
"BlenderbotSmallDecoderWrapper", # Building part of bigger (tested) model.
"BlenderbotEncoder", # Building part of bigger (tested) model.
"BlenderbotDecoderWrapper", # Building part of bigger (tested) model.
"MBartEncoder", # Building part of bigger (tested) model.
"MBartDecoderWrapper", # Building part of bigger (tested) model.
"MegatronBertLMHeadModel", # Building part of bigger (tested) model.
"MegatronBertEncoder", # Building part of bigger (tested) model.
"MegatronBertDecoder", # Building part of bigger (tested) model.
"MegatronBertDecoderWrapper", # Building part of bigger (tested) model.
"PegasusEncoder", # Building part of bigger (tested) model.
"PegasusDecoderWrapper", # Building part of bigger (tested) model.
"DPREncoder", # Building part of bigger (tested) model.
"ProphetNetDecoderWrapper", # Building part of bigger (tested) model.
"RealmBertModel", # Building part of bigger (tested) model.
"RealmReader", # Not regular model.
"RealmScorer", # Not regular model.
"RealmForOpenQA", # Not regular model.
"ReformerForMaskedLM", # Needs to be setup as decoder.
"Speech2Text2DecoderWrapper", # Building part of bigger (tested) model.
"TFDPREncoder", # Building part of bigger (tested) model.
"TFElectraMainLayer", # Building part of bigger (tested) model (should it be a TFModelMixin ?)
"TFRobertaForMultipleChoice", # TODO: fix
"TrOCRDecoderWrapper", # Building part of bigger (tested) model.
"SeparableConv1D", # Building part of bigger (tested) model.
"FlaxBartForCausalLM", # Building part of bigger (tested) model.
"FlaxBertForCausalLM", # Building part of bigger (tested) model. Tested implicitly through FlaxRobertaForCausalLM.
"OPTDecoderWrapper",
]
# Update this list with test files that don't have a tester with a `all_model_classes` variable and which don't
# trigger the common tests.
TEST_FILES_WITH_NO_COMMON_TESTS = [
"models/decision_transformer/test_modeling_decision_transformer.py",
"models/camembert/test_modeling_camembert.py",
"models/mt5/test_modeling_flax_mt5.py",
"models/mbart/test_modeling_mbart.py",
"models/mt5/test_modeling_mt5.py",
"models/pegasus/test_modeling_pegasus.py",
"models/camembert/test_modeling_tf_camembert.py",
"models/mt5/test_modeling_tf_mt5.py",
"models/xlm_roberta/test_modeling_tf_xlm_roberta.py",
"models/xlm_roberta/test_modeling_flax_xlm_roberta.py",
"models/xlm_prophetnet/test_modeling_xlm_prophetnet.py",
"models/xlm_roberta/test_modeling_xlm_roberta.py",
"models/vision_text_dual_encoder/test_modeling_vision_text_dual_encoder.py",
"models/vision_text_dual_encoder/test_modeling_flax_vision_text_dual_encoder.py",
"models/decision_transformer/test_modeling_decision_transformer.py",
]
# Update this list for models that are not in any of the auto MODEL_XXX_MAPPING. Being in this list is an exception and
# should **not** be the rule.
IGNORE_NON_AUTO_CONFIGURED = PRIVATE_MODELS.copy() + [
# models to ignore for model xxx mapping
"DPTForDepthEstimation",
"DecisionTransformerGPT2Model",
"GLPNForDepthEstimation",
"ViltForQuestionAnswering",
"ViltForImagesAndTextClassification",
"ViltForImageAndTextRetrieval",
"ViltForMaskedLM",
"XGLMEncoder",
"XGLMDecoder",
"XGLMDecoderWrapper",
"PerceiverForMultimodalAutoencoding",
"PerceiverForOpticalFlow",
"SegformerDecodeHead",
"FlaxBeitForMaskedImageModeling",
"PLBartEncoder",
"PLBartDecoder",
"PLBartDecoderWrapper",
"BeitForMaskedImageModeling",
"CLIPTextModel",
"CLIPVisionModel",
"TFCLIPTextModel",
"TFCLIPVisionModel",
"FlaxCLIPTextModel",
"FlaxCLIPVisionModel",
"FlaxWav2Vec2ForCTC",
"DetrForSegmentation",
"DPRReader",
"FlaubertForQuestionAnswering",
"FlavaImageCodebook",
"FlavaTextModel",
"FlavaImageModel",
"FlavaMultimodalModel",
"GPT2DoubleHeadsModel",
"LukeForMaskedLM",
"LukeForEntityClassification",
"LukeForEntityPairClassification",
"LukeForEntitySpanClassification",
"OpenAIGPTDoubleHeadsModel",
"RagModel",
"RagSequenceForGeneration",
"RagTokenForGeneration",
"RealmEmbedder",
"RealmForOpenQA",
"RealmScorer",
"RealmReader",
"TFDPRReader",
"TFGPT2DoubleHeadsModel",
"TFOpenAIGPTDoubleHeadsModel",
"TFRagModel",
"TFRagSequenceForGeneration",
"TFRagTokenForGeneration",
"Wav2Vec2ForCTC",
"HubertForCTC",
"SEWForCTC",
"SEWDForCTC",
"XLMForQuestionAnswering",
"XLNetForQuestionAnswering",
"SeparableConv1D",
"VisualBertForRegionToPhraseAlignment",
"VisualBertForVisualReasoning",
"VisualBertForQuestionAnswering",
"VisualBertForMultipleChoice",
"TFWav2Vec2ForCTC",
"TFHubertForCTC",
"MaskFormerForInstanceSegmentation",
]
# Update this list for models that have multiple model types for the same
# model doc
MODEL_TYPE_TO_DOC_MAPPING = OrderedDict(
[
("data2vec-text", "data2vec"),
("data2vec-audio", "data2vec"),
("data2vec-vision", "data2vec"),
]
)
# This is to make sure the transformers module imported is the one in the repo.
spec = importlib.util.spec_from_file_location(
"diffusers",
os.path.join(PATH_TO_DIFFUSERS, "__init__.py"),
submodule_search_locations=[PATH_TO_DIFFUSERS],
)
diffusers = spec.loader.load_module()
def check_model_list():
"""Check the model list inside the transformers library."""
# Get the models from the directory structure of `src/diffusers/models/`
models_dir = os.path.join(PATH_TO_DIFFUSERS, "models")
_models = []
for model in os.listdir(models_dir):
model_dir = os.path.join(models_dir, model)
if os.path.isdir(model_dir) and "__init__.py" in os.listdir(model_dir):
_models.append(model)
# Get the models from the directory structure of `src/transformers/models/`
models = [model for model in dir(diffusers.models) if not model.startswith("__")]
missing_models = sorted(set(_models).difference(models))
if missing_models:
raise Exception(
f"The following models should be included in {models_dir}/__init__.py: {','.join(missing_models)}."
)
# If some modeling modules should be ignored for all checks, they should be added in the nested list
# _ignore_modules of this function.
def get_model_modules():
"""Get the model modules inside the transformers library."""
_ignore_modules = [
"modeling_auto",
"modeling_encoder_decoder",
"modeling_marian",
"modeling_mmbt",
"modeling_outputs",
"modeling_retribert",
"modeling_utils",
"modeling_flax_auto",
"modeling_flax_encoder_decoder",
"modeling_flax_utils",
"modeling_speech_encoder_decoder",
"modeling_flax_speech_encoder_decoder",
"modeling_flax_vision_encoder_decoder",
"modeling_transfo_xl_utilities",
"modeling_tf_auto",
"modeling_tf_encoder_decoder",
"modeling_tf_outputs",
"modeling_tf_pytorch_utils",
"modeling_tf_utils",
"modeling_tf_transfo_xl_utilities",
"modeling_tf_vision_encoder_decoder",
"modeling_vision_encoder_decoder",
]
modules = []
for model in dir(diffusers.models):
# There are some magic dunder attributes in the dir, we ignore them
if not model.startswith("__"):
model_module = getattr(diffusers.models, model)
for submodule in dir(model_module):
if submodule.startswith("modeling") and submodule not in _ignore_modules:
modeling_module = getattr(model_module, submodule)
if inspect.ismodule(modeling_module):
modules.append(modeling_module)
return modules
def get_models(module, include_pretrained=False):
"""Get the objects in module that are models."""
models = []
model_classes = (diffusers.ModelMixin, diffusers.TFModelMixin, diffusers.FlaxModelMixin)
for attr_name in dir(module):
if not include_pretrained and ("Pretrained" in attr_name or "PreTrained" in attr_name):
continue
attr = getattr(module, attr_name)
if isinstance(attr, type) and issubclass(attr, model_classes) and attr.__module__ == module.__name__:
models.append((attr_name, attr))
return models
def is_a_private_model(model):
"""Returns True if the model should not be in the main init."""
if model in PRIVATE_MODELS:
return True
# Wrapper, Encoder and Decoder are all privates
if model.endswith("Wrapper"):
return True
if model.endswith("Encoder"):
return True
if model.endswith("Decoder"):
return True
return False
def check_models_are_in_init():
"""Checks all models defined in the library are in the main init."""
models_not_in_init = []
dir_transformers = dir(diffusers)
for module in get_model_modules():
models_not_in_init += [
model[0] for model in get_models(module, include_pretrained=True) if model[0] not in dir_transformers
]
# Remove private models
models_not_in_init = [model for model in models_not_in_init if not is_a_private_model(model)]
if len(models_not_in_init) > 0:
raise Exception(f"The following models should be in the main init: {','.join(models_not_in_init)}.")
# If some test_modeling files should be ignored when checking models are all tested, they should be added in the
# nested list _ignore_files of this function.
def get_model_test_files():
"""Get the model test files.
The returned files should NOT contain the `tests` (i.e. `PATH_TO_TESTS` defined in this script). They will be
considered as paths relative to `tests`. A caller has to use `os.path.join(PATH_TO_TESTS, ...)` to access the files.
"""
_ignore_files = [
"test_modeling_common",
"test_modeling_encoder_decoder",
"test_modeling_flax_encoder_decoder",
"test_modeling_flax_speech_encoder_decoder",
"test_modeling_marian",
"test_modeling_tf_common",
"test_modeling_tf_encoder_decoder",
]
test_files = []
# Check both `PATH_TO_TESTS` and `PATH_TO_TESTS/models`
model_test_root = os.path.join(PATH_TO_TESTS, "models")
model_test_dirs = []
for x in os.listdir(model_test_root):
x = os.path.join(model_test_root, x)
if os.path.isdir(x):
model_test_dirs.append(x)
for target_dir in [PATH_TO_TESTS] + model_test_dirs:
for file_or_dir in os.listdir(target_dir):
path = os.path.join(target_dir, file_or_dir)
if os.path.isfile(path):
filename = os.path.split(path)[-1]
if "test_modeling" in filename and os.path.splitext(filename)[0] not in _ignore_files:
file = os.path.join(*path.split(os.sep)[1:])
test_files.append(file)
return test_files
# This is a bit hacky but I didn't find a way to import the test_file as a module and read inside the tester class
# for the all_model_classes variable.
def find_tested_models(test_file):
"""Parse the content of test_file to detect what's in all_model_classes"""
# This is a bit hacky but I didn't find a way to import the test_file as a module and read inside the class
with open(os.path.join(PATH_TO_TESTS, test_file), "r", encoding="utf-8", newline="\n") as f:
content = f.read()
all_models = re.findall(r"all_model_classes\s+=\s+\(\s*\(([^\)]*)\)", content)
# Check with one less parenthesis as well
all_models += re.findall(r"all_model_classes\s+=\s+\(([^\)]*)\)", content)
if len(all_models) > 0:
model_tested = []
for entry in all_models:
for line in entry.split(","):
name = line.strip()
if len(name) > 0:
model_tested.append(name)
return model_tested
def check_models_are_tested(module, test_file):
"""Check models defined in module are tested in test_file."""
# XxxModelMixin are not tested
defined_models = get_models(module)
tested_models = find_tested_models(test_file)
if tested_models is None:
if test_file.replace(os.path.sep, "/") in TEST_FILES_WITH_NO_COMMON_TESTS:
return
return [
f"{test_file} should define `all_model_classes` to apply common tests to the models it tests. "
+ "If this intentional, add the test filename to `TEST_FILES_WITH_NO_COMMON_TESTS` in the file "
+ "`utils/check_repo.py`."
]
failures = []
for model_name, _ in defined_models:
if model_name not in tested_models and model_name not in IGNORE_NON_TESTED:
failures.append(
f"{model_name} is defined in {module.__name__} but is not tested in "
+ f"{os.path.join(PATH_TO_TESTS, test_file)}. Add it to the all_model_classes in that file."
+ "If common tests should not applied to that model, add its name to `IGNORE_NON_TESTED`"
+ "in the file `utils/check_repo.py`."
)
return failures
def check_all_models_are_tested():
"""Check all models are properly tested."""
modules = get_model_modules()
test_files = get_model_test_files()
failures = []
for module in modules:
test_file = [file for file in test_files if f"test_{module.__name__.split('.')[-1]}.py" in file]
if len(test_file) == 0:
failures.append(f"{module.__name__} does not have its corresponding test file {test_file}.")
elif len(test_file) > 1:
failures.append(f"{module.__name__} has several test files: {test_file}.")
else:
test_file = test_file[0]
new_failures = check_models_are_tested(module, test_file)
if new_failures is not None:
failures += new_failures
if len(failures) > 0:
raise Exception(f"There were {len(failures)} failures:\n" + "\n".join(failures))
def get_all_auto_configured_models():
"""Return the list of all models in at least one auto class."""
result = set() # To avoid duplicates we concatenate all model classes in a set.
if is_torch_available():
for attr_name in dir(diffusers.models.auto.modeling_auto):
if attr_name.startswith("MODEL_") and attr_name.endswith("MAPPING_NAMES"):
result = result | set(get_values(getattr(diffusers.models.auto.modeling_auto, attr_name)))
if is_flax_available():
for attr_name in dir(diffusers.models.auto.modeling_flax_auto):
if attr_name.startswith("FLAX_MODEL_") and attr_name.endswith("MAPPING_NAMES"):
result = result | set(get_values(getattr(diffusers.models.auto.modeling_flax_auto, attr_name)))
return list(result)
def ignore_unautoclassed(model_name):
"""Rules to determine if `name` should be in an auto class."""
# Special white list
if model_name in IGNORE_NON_AUTO_CONFIGURED:
return True
# Encoder and Decoder should be ignored
if "Encoder" in model_name or "Decoder" in model_name:
return True
return False
def check_models_are_auto_configured(module, all_auto_models):
"""Check models defined in module are each in an auto class."""
defined_models = get_models(module)
failures = []
for model_name, _ in defined_models:
if model_name not in all_auto_models and not ignore_unautoclassed(model_name):
failures.append(
f"{model_name} is defined in {module.__name__} but is not present in any of the auto mapping. "
"If that is intended behavior, add its name to `IGNORE_NON_AUTO_CONFIGURED` in the file "
"`utils/check_repo.py`."
)
return failures
def check_all_models_are_auto_configured():
"""Check all models are each in an auto class."""
missing_backends = []
if not is_torch_available():
missing_backends.append("PyTorch")
if not is_flax_available():
missing_backends.append("Flax")
if len(missing_backends) > 0:
missing = ", ".join(missing_backends)
if os.getenv("TRANSFORMERS_IS_CI", "").upper() in ENV_VARS_TRUE_VALUES:
raise Exception(
"Full quality checks require all backends to be installed (with `pip install -e .[dev]` in the "
f"Transformers repo, the following are missing: {missing}."
)
else:
warnings.warn(
"Full quality checks require all backends to be installed (with `pip install -e .[dev]` in the "
f"Transformers repo, the following are missing: {missing}. While it's probably fine as long as you "
"didn't make any change in one of those backends modeling files, you should probably execute the "
"command above to be on the safe side."
)
modules = get_model_modules()
all_auto_models = get_all_auto_configured_models()
failures = []
for module in modules:
new_failures = check_models_are_auto_configured(module, all_auto_models)
if new_failures is not None:
failures += new_failures
if len(failures) > 0:
raise Exception(f"There were {len(failures)} failures:\n" + "\n".join(failures))
_re_decorator = re.compile(r"^\s*@(\S+)\s+$")
def check_decorator_order(filename):
"""Check that in the test file `filename` the slow decorator is always last."""
with open(filename, "r", encoding="utf-8", newline="\n") as f:
lines = f.readlines()
decorator_before = None
errors = []
for i, line in enumerate(lines):
search = _re_decorator.search(line)
if search is not None:
decorator_name = search.groups()[0]
if decorator_before is not None and decorator_name.startswith("parameterized"):
errors.append(i)
decorator_before = decorator_name
elif decorator_before is not None:
decorator_before = None
return errors
def check_all_decorator_order():
"""Check that in all test files, the slow decorator is always last."""
errors = []
for fname in os.listdir(PATH_TO_TESTS):
if fname.endswith(".py"):
filename = os.path.join(PATH_TO_TESTS, fname)
new_errors = check_decorator_order(filename)
errors += [f"- {filename}, line {i}" for i in new_errors]
if len(errors) > 0:
msg = "\n".join(errors)
raise ValueError(
"The parameterized decorator (and its variants) should always be first, but this is not the case in the"
f" following files:\n{msg}"
)
def find_all_documented_objects():
"""Parse the content of all doc files to detect which classes and functions it documents"""
documented_obj = []
for doc_file in Path(PATH_TO_DOC).glob("**/*.rst"):
with open(doc_file, "r", encoding="utf-8", newline="\n") as f:
content = f.read()
raw_doc_objs = re.findall(r"(?:autoclass|autofunction):: transformers.(\S+)\s+", content)
documented_obj += [obj.split(".")[-1] for obj in raw_doc_objs]
for doc_file in Path(PATH_TO_DOC).glob("**/*.md"):
with open(doc_file, "r", encoding="utf-8", newline="\n") as f:
content = f.read()
raw_doc_objs = re.findall(r"\[\[autodoc\]\]\s+(\S+)\s+", content)
documented_obj += [obj.split(".")[-1] for obj in raw_doc_objs]
return documented_obj
# One good reason for not being documented is to be deprecated. Put in this list deprecated objects.
DEPRECATED_OBJECTS = [
"AutoModelWithLMHead",
"BartPretrainedModel",
"DataCollator",
"DataCollatorForSOP",
"GlueDataset",
"GlueDataTrainingArguments",
"LineByLineTextDataset",
"LineByLineWithRefDataset",
"LineByLineWithSOPTextDataset",
"PretrainedBartModel",
"PretrainedFSMTModel",
"SingleSentenceClassificationProcessor",
"SquadDataTrainingArguments",
"SquadDataset",
"SquadExample",
"SquadFeatures",
"SquadV1Processor",
"SquadV2Processor",
"TFAutoModelWithLMHead",
"TFBartPretrainedModel",
"TextDataset",
"TextDatasetForNextSentencePrediction",
"Wav2Vec2ForMaskedLM",
"Wav2Vec2Tokenizer",
"glue_compute_metrics",
"glue_convert_examples_to_features",
"glue_output_modes",
"glue_processors",
"glue_tasks_num_labels",
"squad_convert_examples_to_features",
"xnli_compute_metrics",
"xnli_output_modes",
"xnli_processors",
"xnli_tasks_num_labels",
"TFTrainer",
"TFTrainingArguments",
]
# Exceptionally, some objects should not be documented after all rules passed.
# ONLY PUT SOMETHING IN THIS LIST AS A LAST RESORT!
UNDOCUMENTED_OBJECTS = [
"AddedToken", # This is a tokenizers class.
"BasicTokenizer", # Internal, should never have been in the main init.
"CharacterTokenizer", # Internal, should never have been in the main init.
"DPRPretrainedReader", # Like an Encoder.
"DummyObject", # Just picked by mistake sometimes.
"MecabTokenizer", # Internal, should never have been in the main init.
"ModelCard", # Internal type.
"SqueezeBertModule", # Internal building block (should have been called SqueezeBertLayer)
"TFDPRPretrainedReader", # Like an Encoder.
"TransfoXLCorpus", # Internal type.
"WordpieceTokenizer", # Internal, should never have been in the main init.
"absl", # External module
"add_end_docstrings", # Internal, should never have been in the main init.
"add_start_docstrings", # Internal, should never have been in the main init.
"cached_path", # Internal used for downloading models.
"convert_tf_weight_name_to_pt_weight_name", # Internal used to convert model weights
"logger", # Internal logger
"logging", # External module
"requires_backends", # Internal function
]
# This list should be empty. Objects in it should get their own doc page.
SHOULD_HAVE_THEIR_OWN_PAGE = [
# Benchmarks
"PyTorchBenchmark",
"PyTorchBenchmarkArguments",
"TensorFlowBenchmark",
"TensorFlowBenchmarkArguments",
]
def ignore_undocumented(name):
"""Rules to determine if `name` should be undocumented."""
# NOT DOCUMENTED ON PURPOSE.
# Constants uppercase are not documented.
if name.isupper():
return True
# ModelMixins / Encoders / Decoders / Layers / Embeddings / Attention are not documented.
if (
name.endswith("ModelMixin")
or name.endswith("Decoder")
or name.endswith("Encoder")
or name.endswith("Layer")
or name.endswith("Embeddings")
or name.endswith("Attention")
):
return True
# Submodules are not documented.
if os.path.isdir(os.path.join(PATH_TO_DIFFUSERS, name)) or os.path.isfile(
os.path.join(PATH_TO_DIFFUSERS, f"{name}.py")
):
return True
# All load functions are not documented.
if name.startswith("load_tf") or name.startswith("load_pytorch"):
return True
# is_xxx_available functions are not documented.
if name.startswith("is_") and name.endswith("_available"):
return True
# Deprecated objects are not documented.
if name in DEPRECATED_OBJECTS or name in UNDOCUMENTED_OBJECTS:
return True
# MMBT model does not really work.
if name.startswith("MMBT"):
return True
if name in SHOULD_HAVE_THEIR_OWN_PAGE:
return True
return False
def check_all_objects_are_documented():
"""Check all models are properly documented."""
documented_objs = find_all_documented_objects()
modules = diffusers._modules
objects = [c for c in dir(diffusers) if c not in modules and not c.startswith("_")]
undocumented_objs = [c for c in objects if c not in documented_objs and not ignore_undocumented(c)]
if len(undocumented_objs) > 0:
raise Exception(
"The following objects are in the public init so should be documented:\n - "
+ "\n - ".join(undocumented_objs)
)
check_docstrings_are_in_md()
check_model_type_doc_match()
def check_model_type_doc_match():
"""Check all doc pages have a corresponding model type."""
model_doc_folder = Path(PATH_TO_DOC) / "model_doc"
model_docs = [m.stem for m in model_doc_folder.glob("*.md")]
model_types = list(diffusers.models.auto.configuration_auto.MODEL_NAMES_MAPPING.keys())
model_types = [MODEL_TYPE_TO_DOC_MAPPING[m] if m in MODEL_TYPE_TO_DOC_MAPPING else m for m in model_types]
errors = []
for m in model_docs:
if m not in model_types and m != "auto":
close_matches = get_close_matches(m, model_types)
error_message = f"{m} is not a proper model identifier."
if len(close_matches) > 0:
close_matches = "/".join(close_matches)
error_message += f" Did you mean {close_matches}?"
errors.append(error_message)
if len(errors) > 0:
raise ValueError(
"Some model doc pages do not match any existing model type:\n"
+ "\n".join(errors)
+ "\nYou can add any missing model type to the `MODEL_NAMES_MAPPING` constant in "
"models/auto/configuration_auto.py."
)
# Re pattern to catch :obj:`xx`, :class:`xx`, :func:`xx` or :meth:`xx`.
_re_rst_special_words = re.compile(r":(?:obj|func|class|meth):`([^`]+)`")
# Re pattern to catch things between double backquotes.
_re_double_backquotes = re.compile(r"(^|[^`])``([^`]+)``([^`]|$)")
# Re pattern to catch example introduction.
_re_rst_example = re.compile(r"^\s*Example.*::\s*$", flags=re.MULTILINE)
def is_rst_docstring(docstring):
"""
Returns `True` if `docstring` is written in rst.
"""
if _re_rst_special_words.search(docstring) is not None:
return True
if _re_double_backquotes.search(docstring) is not None:
return True
if _re_rst_example.search(docstring) is not None:
return True
return False
def check_docstrings_are_in_md():
"""Check all docstrings are in md"""
files_with_rst = []
for file in Path(PATH_TO_DIFFUSERS).glob("**/*.py"):
with open(file, "r") as f:
code = f.read()
docstrings = code.split('"""')
for idx, docstring in enumerate(docstrings):
if idx % 2 == 0 or not is_rst_docstring(docstring):
continue
files_with_rst.append(file)
break
if len(files_with_rst) > 0:
raise ValueError(
"The following files have docstrings written in rst:\n"
+ "\n".join([f"- {f}" for f in files_with_rst])
+ "\nTo fix this run `doc-builder convert path_to_py_file` after installing `doc-builder`\n"
"(`pip install git+https://github.com/huggingface/doc-builder`)"
)
def check_repo_quality():
"""Check all models are properly tested and documented."""
print("Checking all models are included.")
check_model_list()
print("Checking all models are public.")
check_models_are_in_init()
print("Checking all models are properly tested.")
check_all_decorator_order()
check_all_models_are_tested()
print("Checking all objects are properly documented.")
check_all_objects_are_documented()
print("Checking all models are in at least one auto class.")
check_all_models_are_auto_configured()
if __name__ == "__main__":
check_repo_quality()
| diffusers/utils/check_repo.py/0 | {
"file_path": "diffusers/utils/check_repo.py",
"repo_id": "diffusers",
"token_count": 12371
} | 136 |
<jupyter_start><jupyter_text>DreamBooth Hackathon 🏆 Welcome to the DreamBooth Hackathon! In this competition, you'll **personalise a Stable Diffusion model by fine-tuning it on a handful of your own images.** To do so, we'll use a technique called [_DreamBooth_](https://arxiv.org/abs/2208.12242), which allows one to implant a subject (e.g. your pet or favourite dish) into the output domain of the model such that it can be synthesized with a _unique identifier_ in the prompt.Let's dive in! PrerequisitesBefore diving into this notebook, you should read the:* [Unit 3 README](https://github.com/huggingface/diffusion-models-class/blob/main/unit3/README.md) that contains a deep dive into Stable Diffusion* DreamBooth [blog post](https://dreambooth.github.io/) to get a sense of what's possible with this technique* Hugging Face [blog post](https://huggingface.co/blog/dreambooth) on best practices for fine-tuning Stable Diffusion with DreamBooth 🚨 **Note:** the code in **this notebook requires at least 14GB of GPU vRAM** and is a simplified version of the [official training script](https://github.com/huggingface/diffusers/tree/main/examples/dreambooth) provided in 🤗 Diffusers. It produces decent models for most applications, but we recommend experimenting with the advanced features like class preservation loss & fine-tuning the text encoder if you have at least 24GB vRAM available. Check out the 🤗 Diffusers [docs](https://huggingface.co/docs/diffusers/training/dreambooth) for more details. What is DreamBooth? DreamBooth is a technique to teach new concepts to Stable Diffusion using a specialized form of fine-tuning. If you're on Twitter or Reddit, you may have seen people using this technique to create (often hilarious) avatars of themselves. For example, here's what [Andrej Karpathy](https://karpathy.ai/) would look like as a cowboy (you may need to run the cell to see the output):<jupyter_code>%%html
<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Stableboost auto-suggests a few hundred prompts by default but you can generate additional variations for any one prompt that seems to be giving fun/interesting results, or adjust it in any way: <a href="https://t.co/qWmadiXftP">pic.twitter.com/qWmadiXftP</a></p>— Andrej Karpathy (@karpathy) <a href="https://twitter.com/karpathy/status/1600578187141840896?ref_src=twsrc%5Etfw">December 7, 2022</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script><jupyter_output><empty_output><jupyter_text>The way DreamBooth works is as follows:* Collect around 10-20 input images of a subject (e.g. your dog) and define a unique identifier [V] the refers to the subject. This identifier is usually some made up word like `flffydog` which is implanted in different text prompts at inference time to place the subject in different contexts.* Fine-tune the diffusion model by providing the images together with a text prompt like "A photo of a [V] dog" that contains the unique identifier and class name (i.e. "dog" in this example)* (Optionally) Apply a special _class-specific prior preservation loss_, which leverages the semantic prior that the model has on the class and encourages it to generate diverse instances belong to the subject's class by injecting the class name in the text prompt. In practice, this step is only really needed for human faces and can be skipped for the themes we'll be exploring in this hackathon.An overview of the DreamBooth technique is shown in the image below: What can DreamBooth do?Besides putting your subject in interesting locations, DreamBooth can be used for _**text-guided view synthesis**_, where the subject is viewed from different viewpoints as shown in the example below: DreamBooth can also be used to modify properties of the subject, such as colour or mixing up animal species! Now that we've seen some of the cool things DreamBooth can do, let's start training our own models! Step 1: Setup If you're running this notebook on Google Colab or Kaggle, run the cell below to install the required libraries:<jupyter_code>%pip install -qqU diffusers transformers bitsandbytes accelerate ftfy datasets<jupyter_output><empty_output><jupyter_text>If you're running on Kaggle, you'll need to install the latest PyTorch version to work with 🤗 Accelerate:<jupyter_code># Uncomment and run if using Kaggle's notebooks. You may need to restart the notebook afterwards
# %pip install -U torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116<jupyter_output><empty_output><jupyter_text>To be able to push your model to the Hub and make it appear on the [DreamBooth Leaderboard](https://huggingface.co/spaces/dreambooth-hackathon/leaderboard), there are a few more steps to follow. First you have to create an [access token](https://huggingface.co/docs/hub/security-tokens) with _**write access**_ from your Hugging Face account and then execute the following cell and input your token:<jupyter_code>from huggingface_hub import notebook_login
notebook_login()<jupyter_output><empty_output><jupyter_text>The final step is to install Git LFS:<jupyter_code>%%capture
!sudo apt -qq install git-lfs
!git config --global credential.helper store<jupyter_output><empty_output><jupyter_text>Step 2: Pick a theme This competition is composed of 5 _themes_, where each theme will collect models belong to the following categories:* **Animal 🐨:** Use this theme to generate images of your pet or favourite animal hanging out in the Acropolis, swimming, or flying in space.* **Science 🔬:** Use this theme to generate cool synthetic images of galaxies, proteins, or any domain of the natural and medical sciences.* **Food 🍔:** Use this theme to tune Stable Diffusion on your favourite dish or cuisine.* **Landscape 🏔:** Use this theme to generate beautiful landscapes of your faourite mountain, lake, or garden.* **Wildcard 🔥:** Use this theme to go wild and create Stable Diffusion models for any category of your choosing!We'll be **giving out prizes to the top 3 most liked models per theme**, and you're encouraged to submit as many models as you want! Run the cell below to create a dropdown widget where you can select the theme you wish to submit to:<jupyter_code>import ipywidgets as widgets
theme = "animal"
drop_down = widgets.Dropdown(
options=["animal", "science", "food", "landscape", "wildcard"],
description="Pick a theme",
disabled=False,
)
def dropdown_handler(change):
global theme
theme = change.new
drop_down.observe(dropdown_handler, names="value")
display(drop_down)
print(f"You've selected the {theme} theme!")<jupyter_output>You've selected the animal theme!<jupyter_text>Step 3: Create an image dataset and upload it to the Hub Once you've picked a theme, the next step is to **create a dataset of images for that theme** and upload it to the Hugging Face Hub:* You'll need around **10-20 images of the subject** that you wish to implant in the model. These can be photos you've taken or downloaded from platforms like [Unsplash](https://unsplash.com/). Alternatively, you can take a look at any of the [image datasets](https://huggingface.co/datasets?task_categories=task_categories:image-classification&sort=downloads) on the Hugging Face Hub for inspiration.* For best results, we recommend using images of your subject from **different angles and perspectives** Once you've collected your images in a folder, you can upload them to the Hub by using the UI to drag and drop your images. See [this guide](https://huggingface.co/docs/datasets/upload_datasetupload-with-the-hub-ui) for more details, or watch the video below:<jupyter_code>from IPython.display import YouTubeVideo
YouTubeVideo("HaN6qCr_Afc")<jupyter_output><empty_output><jupyter_text>Alternatively, you can load your dataset locally using the `imagefolder` feature of 🤗 Datasets and then push it to the Hub:```pythonfrom datasets import load_datasetdataset = load_dataset("imagefolder", data_dir="your_folder_of_images") Remove the dummy label columndataset = dataset.remove_columns("label") Push to Hubdataset.push_to_hub("dreambooth-hackathon-images")``` Once you've created your dataset, you can download it by using the `load_dataset()` function as follows:<jupyter_code>from datasets import load_dataset
dataset_id = "lewtun/corgi" # CHANGE THIS TO YOUR {hub_username}/{dataset_id}
dataset = load_dataset(dataset_id, split="train")
dataset<jupyter_output>Using custom data configuration lewtun--corgi-387a0d84d49c152d
Found cached dataset parquet (/home/lewis_huggingface_co/.cache/huggingface/datasets/lewtun___parquet/lewtun--corgi-387a0d84d49c152d/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)<jupyter_text>Now that we have our dataset, let's define a helper function to view a few of the images:<jupyter_code>from PIL import Image
def image_grid(imgs, rows, cols):
assert len(imgs) == rows * cols
w, h = imgs[0].size
grid = Image.new("RGB", size=(cols * w, rows * h))
grid_w, grid_h = grid.size
for i, img in enumerate(imgs):
grid.paste(img, box=(i % cols * w, i // cols * h))
return grid
num_samples = 4
image_grid(dataset["image"][:num_samples], rows=1, cols=num_samples)<jupyter_output><empty_output><jupyter_text>If this looks good, you can move onto the next step - creating a PyTorch dataset for training with DreamBooth. Step 3: Create a training dataset To create a training set for our images we need a few components:* An _instance prompt_ that is used to prime the model at the start of training. In most cases, using "a photo of [identifier] [class noun]" works quite well, e.g. "a photo of ccorgi dog" for our cute Corgi pictures. * **Note:** it is recommended that you pick a unique / made up word like `ccorgi` to describe your subject. This will ensure a common word in the model's vocabulary isn't overwritten.* A _tokenizer_ to convert the instance prompt into input IDs that can be fed to the text encoder of Stable Diffusion* A set of _image transforms_, notably resizing the images to a common shape and normalizing the pixel values to a common mean and standard distribution.With this in mind, let's start by defining the instance prompt:<jupyter_code>name_of_your_concept = "ccorgi" # CHANGE THIS ACCORDING TO YOUR SUBJECT
type_of_thing = "dog" # CHANGE THIS ACCORDING TO YOUR SUBJECT
instance_prompt = f"a photo of {name_of_your_concept} {type_of_thing}"
print(f"Instance prompt: {instance_prompt}")<jupyter_output>Instance prompt: a photo of ccorgi dog<jupyter_text>Next, we need to create a PyTorch `Dataset` object that implements the `__len__` and `__getitem__` dunder methods:<jupyter_code>from torch.utils.data import Dataset
from torchvision import transforms
class DreamBoothDataset(Dataset):
def __init__(self, dataset, instance_prompt, tokenizer, size=512):
self.dataset = dataset
self.instance_prompt = instance_prompt
self.tokenizer = tokenizer
self.size = size
self.transforms = transforms.Compose(
[
transforms.Resize(size),
transforms.CenterCrop(size),
transforms.ToTensor(),
transforms.Normalize([0.5], [0.5]),
]
)
def __len__(self):
return len(self.dataset)
def __getitem__(self, index):
example = {}
image = self.dataset[index]["image"]
example["instance_images"] = self.transforms(image)
example["instance_prompt_ids"] = self.tokenizer(
self.instance_prompt,
padding="do_not_pad",
truncation=True,
max_length=self.tokenizer.model_max_length,
).input_ids
return example<jupyter_output><empty_output><jupyter_text>Great, let's now check this works by loading the CLIP tokenizer associated with the text encoder of the original Stable Diffusion model, and then creating the training dataset:<jupyter_code>from transformers import CLIPTokenizer
# The Stable Diffusion checkpoint we'll fine-tune
model_id = "CompVis/stable-diffusion-v1-4"
tokenizer = CLIPTokenizer.from_pretrained(
model_id,
subfolder="tokenizer",
)
train_dataset = DreamBoothDataset(dataset, instance_prompt, tokenizer)
train_dataset[0]<jupyter_output><empty_output><jupyter_text>Step 4: Define a data collator Now that we have a training dataset, the next thing we need is to define a _data collator_. A data collator is a function that collects elements in a batch of data and applies some logic to form a single tensor we can provide to the model. If you'd to learn more, you can check out this video from the [Hugging Face Course](hf.co/course):<jupyter_code>YouTubeVideo("-RPeakdlHYo")<jupyter_output><empty_output><jupyter_text>For DreamBooth, our data collator need to provide the model with the input IDs from the tokenizer and the pixel values from the images as a stacked tensor. The function below does the trick:<jupyter_code>import torch
def collate_fn(examples):
input_ids = [example["instance_prompt_ids"] for example in examples]
pixel_values = [example["instance_images"] for example in examples]
pixel_values = torch.stack(pixel_values)
pixel_values = pixel_values.to(memory_format=torch.contiguous_format).float()
input_ids = tokenizer.pad(
{"input_ids": input_ids}, padding=True, return_tensors="pt"
).input_ids
batch = {
"input_ids": input_ids,
"pixel_values": pixel_values,
}
return batch<jupyter_output><empty_output><jupyter_text>Step 5: Load the components of the Stable Diffusion pipeline We nearly have all the pieces ready for training! As you saw in the Unit 3 notebook on Stable Diffusion, the pipeline is composed of several models:* A text encoder that converts the prompts into text embeddings. Here we're using CLIP since it's the encoder used to train Stable Diffusion v1-4* A VAE or variational autoencoder that converts the images to compressed representations (i.e. latents) and decompresses them at inference time* A UNet that applies the denoising operation on the latent of the VAEWe can load all these components using the 🤗 Diffusers and 🤗 Transformers libraries as follows:<jupyter_code>from diffusers import AutoencoderKL, UNet2DConditionModel
from transformers import CLIPFeatureExtractor, CLIPTextModel
text_encoder = CLIPTextModel.from_pretrained(model_id, subfolder="text_encoder")
vae = AutoencoderKL.from_pretrained(model_id, subfolder="vae")
unet = UNet2DConditionModel.from_pretrained(model_id, subfolder="unet")
feature_extractor = CLIPFeatureExtractor.from_pretrained("openai/clip-vit-base-patch32")<jupyter_output><empty_output><jupyter_text>Step 6: Fine-tune the model Now comes the fun part - training our model with DreamBooth! As shown in the [Hugging Face's blog post](https://huggingface.co/blog/dreambooth), the most essential hyperparameters to tweak are the learning rate and number of training steps.In general, you'll get better results with a lower learning rate at the expense of needing to increase the number of training steps. The values below are a good starting point, but you may need to adjust them according to your dataset:<jupyter_code>learning_rate = 2e-06
max_train_steps = 400<jupyter_output><empty_output><jupyter_text>Next, let's wrap the other hyperparameters we need in a `Namespace` object to make it easier to configure the training run:<jupyter_code>from argparse import Namespace
args = Namespace(
pretrained_model_name_or_path=model_id,
resolution=512, # Reduce this if you want to save some memory
train_dataset=train_dataset,
instance_prompt=instance_prompt,
learning_rate=learning_rate,
max_train_steps=max_train_steps,
train_batch_size=1,
gradient_accumulation_steps=1, # Increase this if you want to lower memory usage
max_grad_norm=1.0,
gradient_checkpointing=True, # set this to True to lower the memory usage.
use_8bit_adam=True, # use 8bit optimizer from bitsandbytes
seed=3434554,
sample_batch_size=2,
output_dir="my-dreambooth", # where to save the pipeline
)<jupyter_output><empty_output><jupyter_text>The final step is to define a `training_function()` function that wraps the training logic and can be passed to 🤗 Accelerate to handle training on 1 or more GPUs. If this is the first time you're using 🤗 Accelerate, check out this video to get a quick overview of what it can do:<jupyter_code>YouTubeVideo("s7dy8QRgjJ0")<jupyter_output><empty_output><jupyter_text>The details should look familiar to what we saw in Units 1 & 2 when we trained our own diffusion models from scratch:<jupyter_code>import math
import torch.nn.functional as F
from accelerate import Accelerator
from accelerate.utils import set_seed
from diffusers import DDPMScheduler, PNDMScheduler, StableDiffusionPipeline
from diffusers.pipelines.stable_diffusion import StableDiffusionSafetyChecker
from torch.utils.data import DataLoader
from tqdm.auto import tqdm
def training_function(text_encoder, vae, unet):
accelerator = Accelerator(
gradient_accumulation_steps=args.gradient_accumulation_steps,
)
set_seed(args.seed)
if args.gradient_checkpointing:
unet.enable_gradient_checkpointing()
# Use 8-bit Adam for lower memory usage or to fine-tune the model in 16GB GPUs
if args.use_8bit_adam:
import bitsandbytes as bnb
optimizer_class = bnb.optim.AdamW8bit
else:
optimizer_class = torch.optim.AdamW
optimizer = optimizer_class(
unet.parameters(), # only optimize unet
lr=args.learning_rate,
)
noise_scheduler = DDPMScheduler(
beta_start=0.00085,
beta_end=0.012,
beta_schedule="scaled_linear",
num_train_timesteps=1000,
)
train_dataloader = DataLoader(
args.train_dataset,
batch_size=args.train_batch_size,
shuffle=True,
collate_fn=collate_fn,
)
unet, optimizer, train_dataloader = accelerator.prepare(
unet, optimizer, train_dataloader
)
# Move text_encode and vae to gpu
text_encoder.to(accelerator.device)
vae.to(accelerator.device)
# We need to recalculate our total training steps as the size of the training dataloader may have changed.
num_update_steps_per_epoch = math.ceil(
len(train_dataloader) / args.gradient_accumulation_steps
)
num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
# Train!
total_batch_size = (
args.train_batch_size
* accelerator.num_processes
* args.gradient_accumulation_steps
)
# Only show the progress bar once on each machine.
progress_bar = tqdm(
range(args.max_train_steps), disable=not accelerator.is_local_main_process
)
progress_bar.set_description("Steps")
global_step = 0
for epoch in range(num_train_epochs):
unet.train()
for step, batch in enumerate(train_dataloader):
with accelerator.accumulate(unet):
# Convert images to latent space
with torch.no_grad():
latents = vae.encode(batch["pixel_values"]).latent_dist.sample()
latents = latents * 0.18215
# Sample noise that we'll add to the latents
noise = torch.randn(latents.shape).to(latents.device)
bsz = latents.shape[0]
# Sample a random timestep for each image
timesteps = torch.randint(
0,
noise_scheduler.config.num_train_timesteps,
(bsz,),
device=latents.device,
).long()
# Add noise to the latents according to the noise magnitude at each timestep
# (this is the forward diffusion process)
noisy_latents = noise_scheduler.add_noise(latents, noise, timesteps)
# Get the text embedding for conditioning
with torch.no_grad():
encoder_hidden_states = text_encoder(batch["input_ids"])[0]
# Predict the noise residual
noise_pred = unet(
noisy_latents, timesteps, encoder_hidden_states
).sample
loss = (
F.mse_loss(noise_pred, noise, reduction="none")
.mean([1, 2, 3])
.mean()
)
accelerator.backward(loss)
if accelerator.sync_gradients:
accelerator.clip_grad_norm_(unet.parameters(), args.max_grad_norm)
optimizer.step()
optimizer.zero_grad()
# Checks if the accelerator has performed an optimization step behind the scenes
if accelerator.sync_gradients:
progress_bar.update(1)
global_step += 1
logs = {"loss": loss.detach().item()}
progress_bar.set_postfix(**logs)
if global_step >= args.max_train_steps:
break
accelerator.wait_for_everyone()
# Create the pipeline using using the trained modules and save it.
if accelerator.is_main_process:
print(f"Loading pipeline and saving to {args.output_dir}...")
scheduler = PNDMScheduler(
beta_start=0.00085,
beta_end=0.012,
beta_schedule="scaled_linear",
skip_prk_steps=True,
steps_offset=1,
)
pipeline = StableDiffusionPipeline(
text_encoder=text_encoder,
vae=vae,
unet=accelerator.unwrap_model(unet),
tokenizer=tokenizer,
scheduler=scheduler,
safety_checker=StableDiffusionSafetyChecker.from_pretrained(
"CompVis/stable-diffusion-safety-checker"
),
feature_extractor=feature_extractor,
)
pipeline.save_pretrained(args.output_dir)<jupyter_output><empty_output><jupyter_text>Now that we have the function defined, let's train it! Depending on the size of your dataset and type of GPU, this can take anywhere from 5 minutes to 1 hour to run:<jupyter_code>from accelerate import notebook_launcher
num_of_gpus = 1 # CHANGE THIS TO MATCH THE NUMBER OF GPUS YOU HAVE
notebook_launcher(
training_function, args=(text_encoder, vae, unet), num_processes=num_of_gpus
)<jupyter_output>Launching training on one GPU.<jupyter_text>If you're running on a single GPU, you can free up some memory for the next section by copying the code below into a new cell and running it. For multi-GPU machines, 🤗 Accelerate doesn't allow _any_ cell to directly access the GPU with `torch.cuda`, so we don't recommend using this trick in those cases:```pythonwith torch.no_grad(): torch.cuda.empty_cache()``` Step 7: Run inference and inspect generations Now that we've trained the model, let's generate some images with it to see how it fares! First we'll load the pipeline from the output directory we save the model to:<jupyter_code>pipe = StableDiffusionPipeline.from_pretrained(
args.output_dir,
torch_dtype=torch.float16,
).to("cuda")<jupyter_output><empty_output><jupyter_text>Next, let's generate a few images. The `prompt` variable will later be used to set the default on the Hugging Face Hub widget, so experiment a bit to find a good one. You might also want to try creating elaborate prompts with [CLIP Interrogator](https://huggingface.co/spaces/pharma/CLIP-Interrogator):<jupyter_code># Pick a funny prompt here and it will be used as the widget's default
# when we push to the Hub in the next section
prompt = f"a photo of {name_of_your_concept} {type_of_thing} in the Acropolis"
# Tune the guidance to control how closely the generations follow the prompt.
# Values between 7-11 usually work best
guidance_scale = 7
num_cols = 2
all_images = []
for _ in range(num_cols):
images = pipe(prompt, guidance_scale=guidance_scale).images
all_images.extend(images)
image_grid(all_images, 1, num_cols)<jupyter_output><empty_output><jupyter_text>Step 8: Push your model to the Hub If you're happy with you model, the final step is to push it to the Hub and view it on the [DreamBooth Leaderboard](https://huggingface.co/spaces/dreambooth-hackathon/leaderboard)!First, you'll need to define a name for your model repo. By default, we use the unique identifier and class name, but feel free to change this if you want:<jupyter_code># Create a name for your model on the Hub. No spaces allowed.
model_name = f"{name_of_your_concept}-{type_of_thing}"<jupyter_output><empty_output><jupyter_text>Next, add a brief description on the type of model you've trained or any other information you'd like to share:<jupyter_code># Describe the theme and model you've trained
description = f"""
This is a Stable Diffusion model fine-tuned on `{type_of_thing}` images for the {theme} theme.
"""<jupyter_output><empty_output><jupyter_text>Finally, run the cell below to create a repo on the Hub and push all our files with a nice model card to boot:<jupyter_code># Code to upload a pipeline saved locally to the hub
from huggingface_hub import HfApi, ModelCard, create_repo, get_full_repo_name
# Set up repo and upload files
hub_model_id = get_full_repo_name(model_name)
create_repo(hub_model_id)
api = HfApi()
api.upload_folder(folder_path=args.output_dir, path_in_repo="", repo_id=hub_model_id)
content = f"""
---
license: creativeml-openrail-m
tags:
- pytorch
- diffusers
- stable-diffusion
- text-to-image
- diffusion-models-class
- dreambooth-hackathon
- {theme}
widget:
- text: {prompt}
---
# DreamBooth model for the {name_of_your_concept} concept trained by {api.whoami()["name"]} on the {dataset_id} dataset.
This is a Stable Diffusion model fine-tuned on the {name_of_your_concept} concept with DreamBooth. It can be used by modifying the `instance_prompt`: **{instance_prompt}**
This model was created as part of the DreamBooth Hackathon 🔥. Visit the [organisation page](https://huggingface.co/dreambooth-hackathon) for instructions on how to take part!
## Description
{description}
## Usage
```python
from diffusers import StableDiffusionPipeline
pipeline = StableDiffusionPipeline.from_pretrained('{hub_model_id}')
image = pipeline().images[0]
image
```
"""
card = ModelCard(content)
hub_url = card.push_to_hub(hub_model_id)
print(f"Upload successful! Model can be found here: {hub_url}")
print(
f"View your submission on the public leaderboard here: https://huggingface.co/spaces/dreambooth-hackathon/leaderboard"
)<jupyter_output>Upload successful! Model can be found here: https://huggingface.co/lewtun/test-dogs/blob/main/README.md
View your submission on the public leaderboard here: https://huggingface.co/spaces/dreambooth-hackathon/leaderboard | diffusion-models-class/units/en/events/dreambooth.ipynb/0 | {
"file_path": "diffusion-models-class/units/en/events/dreambooth.ipynb",
"repo_id": "diffusion-models-class",
"token_count": 9183
} | 137 |
<jupyter_start><jupyter_text>*FineTuning* et guidageDans ce *notebook*, nous allons couvrir deux approches principales pour adapter les modèles de diffusion existants :* Avec le *finetuning*, nous entraînons de nouveau les modèles existants sur de nouvelles données dans le but de modifier le résultat qu'ils produisent.* Avec **guidage**, nous prenons un modèle existant et dirigeons le processus de génération au moment de l'inférence pour un contrôle supplémentaire. Ce que vous apprendrez :A la fin de ce *notebook*, vous saurez comment :- Créer une boucle d'échantillonnage et générer des échantillons plus rapidement à l'aide d'un nouveau planificateur- *Finetuner* un modèle de diffusion existant sur de nouvelles données, y compris : - Utiliser l'accumulation du gradient pour contourner certains des problèmes liés aux petits batchs. - Enregistrer les échantillons dans [Weights and Biases] (https://wandb.ai/site) pendant l'entraînement pour suivre la progression (via le script d'exemple joint). - Sauvegarder le pipeline résultant et le télécharger sur le Hub- Guider le processus d'échantillonnage avec des fonctions de perte supplémentaires pour ajouter un contrôle sur les modèles existants, y compris : - Explorer différentes approches de guidage avec une simple perte basée sur la couleur - Utiliser CLIP pour guider la génération à l'aide d'un prompt de texte - Partager une boucle d'échantillonnage personnalisée en utilisant Gradio et 🤗 Spaces.❓ Si vous avez des questions, merci de les poster sur le canal `diffusion-models-class` du [serveur Discord d'Hugging Face](https://huggingface.co/join/discord). Configuration et importationsPour enregistrer vos modèles *finetunés* sur le Hub d'Hugging Face, vous devrez vous connecter avec un *token* qui a un accès en écriture. Le code ci-dessous vous invite à le faire et vous renvoie à la page des *tokens* de votre compte. Vous aurez également besoin d'un compte Weights and Biases si vous souhaitez utiliser le script d'entraînement pour enregistrer des échantillons au fur et à mesure que le modèle s'entraîne. Là encore, le code devrait vous inviter à vous connecter là où c'est nécessaire.A part cela, la seule chose à faire est d'installer quelques dépendances, d'importer tout ce dont nous aurons besoin et de spécifier l'appareil que nous utiliserons :<jupyter_code>!pip install -qq diffusers datasets accelerate wandb open-clip-torch
# Code pour se connecter au Hub d'Hugging Face, nécessaire pour partager les modèles
# Assurez-vous d'utiliser un *token* avec un accès WRITE (écriture)
from huggingface_hub import notebook_login
notebook_login()
import numpy as np
import torch
import torch.nn.functional as F
import torchvision
from datasets import load_dataset
from diffusers import DDIMScheduler, DDPMPipeline
from matplotlib import pyplot as plt
from PIL import Image
from torchvision import transforms
from tqdm.auto import tqdm
device = (
"mps"
if torch.backends.mps.is_available()
else "cuda"
if torch.cuda.is_available()
else "cpu"
)<jupyter_output><empty_output><jupyter_text>Chargement d'un pipeline pré-entraînéPour commencer ce *notebook*, chargeons un pipeline existant et voyons ce que nous pouvons en faire :<jupyter_code>image_pipe = DDPMPipeline.from_pretrained("google/ddpm-celebahq-256")
image_pipe.to(device);<jupyter_output><empty_output><jupyter_text>La génération d'images est aussi simple que l'exécution de la méthode [`__call__`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/ddpm/pipeline_ddpm.pyL42) du pipeline en l'appelant comme une fonction :<jupyter_code>images = image_pipe().images
images[0]<jupyter_output><empty_output><jupyter_text>Sympathique, mais LENT ! Avant d'aborder les sujets principaux du jour, jetons un coup d'œil à la boucle d'échantillonnage proprement dite et voyons comment nous pouvons utiliser un échantillonneur plus sophistiqué pour l'accélérer. Échantillonnage plus rapide avec DDIMÀ chaque étape, le modèle est nourri d'une entrée bruyante et il lui est demandé de prédire le bruit (et donc une estimation de ce à quoi l'image entièrement débruitée pourrait ressembler). Au départ, ces prédictions ne sont pas très bonnes, c'est pourquoi nous décomposons le processus en plusieurs étapes. Cependant, l'utilisation de plus de 1000 étapes s'est avérée inutile, et une multitude de recherches récentes ont exploré la manière d'obtenir de bons échantillons avec le moins d'étapes possible. Dans la bibliothèque 🤗 *Diffusers*, ces méthodes d'échantillonnage sont gérées par un planificateur, qui doit effectuer chaque mise à jour via la fonction `step()`. Pour générer une image, on commence par un bruit aléatoire $x$. Ensuite, pour chaque pas de temps dans le planificateur de bruit, nous introduisons l'entrée bruitée $x$ dans le modèle et transmettons la prédiction résultante à la fonction `step()`. Celle-ci renvoie une sortie avec un attribut `prev_sample`. "previous" parce que nous revenons en arrière dans le temps, d'un niveau de bruit élevé à un niveau de bruit faible (à l'inverse du processus de diffusion vers l'avant).Voyons cela en action ! Tout d'abord, nous chargeons un planificateur, ici un `DDIMScheduler` basé sur le papier [*Denoising Diffusion Implicit Models*](https://arxiv.org/abs/2010.02502) qui peut donner des échantillons décents en beaucoup moins d'étapes que l'implémentation originale du DDPM :<jupyter_code># Créer un nouveau planificateur et définir le nombre d'étapes d'inférence
scheduler = DDIMScheduler.from_pretrained("google/ddpm-celebahq-256")
scheduler.set_timesteps(num_inference_steps=40)<jupyter_output><empty_output><jupyter_text>Vous pouvez constater que ce modèle effectue 40 étapes au total, chaque saut équivalant à 25 étapes du programme original de 1000 étapes :<jupyter_code>scheduler.timesteps<jupyter_output><empty_output><jupyter_text>Créons 4 images aléatoires et exécutons la boucle d'échantillonnage, en visualisant à la fois le $x$ actuel et la version débruitée prédite au fur et à mesure de l'avancement du processus :<jupyter_code># Le point de départ aléatoire
x = torch.randn(4, 3, 256, 256).to(device) # Batch de 4 images à 3 canaux de 256 x 256 px
# Boucle sur les pas de temps d'échantillonnage
for i, t in tqdm(enumerate(scheduler.timesteps)):
# Préparer l'entrée du modèle
model_input = scheduler.scale_model_input(x, t)
# Obtenir la prédiction
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
# Calculer la forme que devrait prendre l'échantillon mis à jour à l'aide du planificateur
scheduler_output = scheduler.step(noise_pred, t, x)
# Mise à jour de x
x = scheduler_output.prev_sample
# Occasionnellement, afficher à la fois x et les images débruitées prédites
if i % 10 == 0 or i == len(scheduler.timesteps) - 1:
fig, axs = plt.subplots(1, 2, figsize=(12, 5))
grid = torchvision.utils.make_grid(x, nrow=4).permute(1, 2, 0)
axs[0].imshow(grid.cpu().clip(-1, 1) * 0.5 + 0.5)
axs[0].set_title(f"Current x (step {i})")
pred_x0 = (
scheduler_output.pred_original_sample
) # Non disponible pour tous les planificateurs
grid = torchvision.utils.make_grid(pred_x0, nrow=4).permute(1, 2, 0)
axs[1].imshow(grid.cpu().clip(-1, 1) * 0.5 + 0.5)
axs[1].set_title(f"Predicted denoised images (step {i})")
plt.show()<jupyter_output><empty_output><jupyter_text>Comme vous pouvez le voir, les prédictions initiales ne sont pas très bonnes, mais au fur et à mesure que le processus se poursuit, les résultats prédits deviennent de plus en plus précis. Si vous êtes curieux de savoir ce qui se passe à l'intérieur de la fonction `step()`, inspectez le code (bien commenté) avec :<jupyter_code># ??scheduler.step<jupyter_output><empty_output><jupyter_text>Vous pouvez également insérer ce nouveau planificateur à la place du planificateur original fourni avec le pipeline, et échantillonner de la manière suivante :<jupyter_code>image_pipe.scheduler = scheduler
images = image_pipe(num_inference_steps=40).images
images[0]<jupyter_output><empty_output><jupyter_text>Très bien, nous pouvons maintenant obtenir des échantillons dans un délai raisonnable ! Cela devrait accélérer les choses au fur et à mesure que nous avançons dans le reste de ce *notebook* :) FinetuningEt maintenant, le plus amusant ! Étant donné ce pipeline pré-entraîné, comment pouvons-nous réentraîner le modèle pour générer des images sur la base de nouvelles données d'entraînement ?Il s'avère que cela est presque identique à entraîner un modèle à partir de zéro (comme nous l'avons vu dans l'unité 1), sauf que nous commençons avec le modèle existant. Voyons cela en action et abordons quelques considérations supplémentaires au fur et à mesure. Tout d'abord, le jeu de données : vous pouvez essayer ce [jeu de données de visages vintage](https://huggingface.co/datasets/Norod78/Vintage-Faces-FFHQAligned) ou ces [visages animés](https://huggingface.co/datasets/huggan/anime-faces) pour quelque chose de plus proche des données d'entraînement originales de ce modèle de visages. Mais pour le plaisir, utilisons plutôt le même petit jeu de données de papillons que nous avons utilisé pour nous entraîner à partir de zéro dans l'unité 1. Exécutez le code ci-dessous pour télécharger le jeu de données papillons et créer un chargeur de données à partir duquel nous pouvons échantillonner un batch d'images :<jupyter_code># Pas sur Colab ? Les commentaires avec #@ permettent de modifier l'interface utilisateur comme les titres ou les entrées
# mais peuvent être ignorés si vous travaillez sur une plateforme différente.
dataset_name = "huggan/smithsonian_butterflies_subset" # @param
dataset = load_dataset(dataset_name, split="train")
image_size = 256 # @param
batch_size = 4 # @param
preprocess = transforms.Compose(
[
transforms.Resize((image_size, image_size)),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize([0.5], [0.5]),
]
)
def transform(examples):
images = [preprocess(image.convert("RGB")) for image in examples["image"]]
return {"images": images}
dataset.set_transform(transform)
train_dataloader = torch.utils.data.DataLoader(
dataset, batch_size=batch_size, shuffle=True
)
print("Previewing batch:")
batch = next(iter(train_dataloader))
grid = torchvision.utils.make_grid(batch["images"], nrow=4)
plt.imshow(grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5);<jupyter_output>Using custom data configuration huggan--smithsonian_butterflies_subset-7665b1021a37404c
Found cached dataset parquet (/home/lewis_huggingface_co/.cache/huggingface/datasets/huggan___parquet/huggan--smithsonian_butterflies_subset-7665b1021a37404c/0.0.0/2a3b91fbd88a2c90d1dbbb32b460cf621d31bd5b05b934492fdef7d8d6f236ec)<jupyter_text>**Considération 1 :** notre taille de batch ici (4) est assez petite, puisque nous entraînons sur une grande taille d'image (256 pixels) en utilisant un modèle assez grand et que nous manquerons de RAM du GPU si nous augmentons trop la taille du batch. Vous pouvez réduire la taille de l'image pour accélérer les choses et permettre des batchs plus importants, mais ces modèles ont été conçus et entraînés à l'origine pour une génération de 256 pixels. Passons maintenant à la boucle d'entraînement. Nous allons mettre à jour les poids du modèle pré-entraîné en fixant la cible d'optimisation à `image_pipe.unet.parameters()`. Le reste est presque identique à l'exemple de boucle d'entraînement de l'unité 1. Cela prend environ 10 minutes à exécuter sur Colab, c'est donc le bon moment pour prendre un café ou un thé pendant que vous attendez :<jupyter_code>num_epochs = 2 # @param
lr = 1e-5 # 2param
grad_accumulation_steps = 2 # @param
optimizer = torch.optim.AdamW(image_pipe.unet.parameters(), lr=lr)
losses = []
for epoch in range(num_epochs):
for step, batch in tqdm(enumerate(train_dataloader), total=len(train_dataloader)):
clean_images = batch["images"].to(device)
# bruit à ajouter aux images
noise = torch.randn(clean_images.shape).to(clean_images.device)
bs = clean_images.shape[0]
# un pas de temps aléatoire pour chaque image
timesteps = torch.randint(
0,
image_pipe.scheduler.num_train_timesteps,
(bs,),
device=clean_images.device,
).long()
# Ajouter du bruit aux images propres en fonction de la magnitude du bruit à chaque pas de temps
# (il s'agit du processus de diffusion vers l'avant)
noisy_images = image_pipe.scheduler.add_noise(clean_images, noise, timesteps)
# Obtenir la prédiction du modèle pour le bruit
noise_pred = image_pipe.unet(noisy_images, timesteps, return_dict=False)[0]
# Comparez la prédiction avec le bruit réel :
loss = F.mse_loss(
noise_pred, noise
) # NB : essayer de prédire le bruit (eps) pas (noisy_ims-clean_ims) ou juste (clean_ims)
# Stocker pour un plot ultérieur
losses.append(loss.item())
# Mettre à jour les paramètres du modèle avec l'optimiseur sur la base de cette perte
loss.backward(loss)
# Accumulation des gradients
if (step + 1) % grad_accumulation_steps == 0:
optimizer.step()
optimizer.zero_grad()
print(
f"Epoch {epoch} average loss: {sum(losses[-len(train_dataloader):])/len(train_dataloader)}"
)
# Tracer la courbe de perte :
plt.plot(losses)<jupyter_output><empty_output><jupyter_text>**Considération 2 :** notre signal de perte est extrêmement bruyant, puisque nous ne travaillons qu'avec quatre exemples à des niveaux de bruit aléatoires pour chaque étape. Ce n'est pas idéal pour l'entraînement. Une solution consiste à utiliser un taux d'apprentissage extrêmement faible pour limiter la taille de la mise à jour à chaque étape. Ce serait encore mieux si nous pouvions trouver un moyen d'obtenir les mêmes avantages qu'en utilisant une taille de batch plus importante sans que les besoins en mémoire ne montent en flèche...Entrez dans [l'accumulation des gradients](https://kozodoi.me/python/deep%20learning/pytorch/tutorial/2021/02/19/gradient-accumulation.html:~:text=Simplement%20speaking%2C%20gradient%20accumulation%20means,may%20find%20 this%20tutorial%20use.). Si nous appelons `loss.backward()` plusieurs fois avant d'exécuter `optimizer.step()` et `optimizer.zero_grad()`, PyTorch accumule (somme) les gradients, fusionnant effectivement le signal de plusieurs batchs pour donner une seule (meilleure) estimation qui est ensuite utilisée pour mettre à jour les paramètres. Il en résulte moins de mises à jour totales, tout comme nous le verrions si nous utilisions une taille de batch plus importante. C'est quelque chose que de nombreux *frameworks* gèrent pour vous (par exemple, 🤗 [Accelerate rend cela facile](https://huggingface.co/docs/accelerate/usage_guides/gradient_accumulation)), mais il est agréable de le voir mis en œuvre à partir de zéro car il s'agit d'une technique utile pour traiter l'entraînement sous les contraintes de mémoire du GPU ! Comme vous pouvez le voir dans le code ci-dessus (après le commentaire Gradient accumulation), il n'y a pas vraiment besoin de beaucoup de code. **Exercice :** Voyez si vous pouvez ajouter l'accumulation des gradients à la boucle d'entraînement de l'unité 1.Comment se comporte-t-elle ? Réfléchissez à la manière dont vous pourriez ajuster le taux d'apprentissage en fonction du nombre d'étapes d'accumulation des gradients ; devrait-il rester identique à auparavant ? **Considération 3 :** Cela prend encore beaucoup de temps, et afficher une mise à jour d'une ligne à chaque époque n'est pas suffisant pour nous donner une bonne idée de ce qui se passe. Nous devrions probablement :- Générer quelques échantillons de temps en temps pour examiner visuellement la performance qualitativement au fur et à mesure que le modèle s'entraîne.- Enregistrer des éléments tels que la perte et les générations d'échantillons pendant l'entraînement, peut-être en utilisant quelque chose comme Weights and Biases ou Tensorboard.Nous avons créé un script rapide (finetune_model.py) qui reprend le code d'entraînement ci-dessus et y ajoute une fonctionnalité minimale de *logging*. Vous pouvez voir les [logs d'un entraînement ci-dessous](https://wandb.ai/johnowhitaker/dm_finetune/runs/2upaa341) :<jupyter_code>%wandb johnowhitaker/dm_finetune/2upaa341 # Vous aurez besoin d'un compte W&B pour que cela fonctionne - sautez si vous ne voulez pas vous connecter.<jupyter_output><empty_output><jupyter_text>Il est amusant de voir comment les échantillons générés changent au fur et à mesure que l'entraînement progresse. Même si la perte ne semble pas s'améliorer beaucoup, on peut voir une progression du domaine original (images de chambres à coucher) vers les nouvelles données d'entraînement (wikiart). A la fin de ce *notebook* se trouve un code commenté pour *finetuné* un modèle en utilisant ce script comme alternative à l'exécution de la cellule ci-dessus. **Exercice :** Voyez si vous pouvez modifier l'exemple officiel de script d'entraînement que nous avons vu dans l'unité 1 pour commencer avec un modèle pré-entraîné plutôt que d'entraîner à partir de zéro.Comparez-le au script minimal dont le lien figure ci-dessus ; quelles sont les fonctionnalités supplémentaires qui manquent au script minimal ? En générant quelques images avec ce modèle, nous pouvons voir que ces visages ont déjà l'air très étranges !<jupyter_code>x = torch.randn(8, 3, 256, 256).to(device) # Batch de 8
for i, t in tqdm(enumerate(scheduler.timesteps)):
model_input = scheduler.scale_model_input(x, t)
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
x = scheduler.step(noise_pred, t, x).prev_sample
grid = torchvision.utils.make_grid(x, nrow=4)
plt.imshow(grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5);<jupyter_output><empty_output><jupyter_text>**Considération 4 :** Le *finetuning* peut être tout à fait imprévisible ! Si nous entraînions plus longtemps, nous pourrions voir des papillons parfaits. Mais les étapes intermédiaires peuvent être extrêmement intéressantes en elles-mêmes, surtout si vos intérêts sont plutôt artistiques ! Entraînez sur des périodes très courtes ou très longues et faites varier le taux d'apprentissage pour voir comment cela affecte les types de résultats produits par le modèle final. Code pour *finetuner* un modèle en utilisant le script d'exemple minimal que nous avons utilisé sur le modèle de démonstration WikiArtSi vous souhaitez entraîner un modèle similaire à celui que nous avons créé sur WikiArt, vous pouvez décommenter et exécuter les cellules ci-dessous. Comme cela prend un certain temps et peut épuiser la mémoire de votre GPU, nous vous conseillons de le faire après avoir parcouru le reste de ce *notebook*.<jupyter_code>## Pour télécharger le script de finetuning :
# !wget https://github.com/huggingface/diffusion-models-class/raw/main/unit2/finetune_model.py
## Pour exécuter le script, entraînant le modèle de visage sur des visages vintage
## (l'idéal est d'exécuter ce script dans un terminal) :
# !python finetune_model.py --image_size 128 --batch_size 8 --num_epochs 16\
# --grad_accumulation_steps 2 --start_model "google/ddpm-celebahq-256"\
# --dataset_name "Norod78/Vintage-Faces-FFHQAligned" --wandb_project 'dm-finetune'\
# --log_samples_every 100 --save_model_every 1000 --model_save_name 'vintageface'<jupyter_output><empty_output><jupyter_text>Sauvegarde et chargement des pipelines *finetunés*Maintenant que nous avons *finetuné* le UNet dans notre modèle de diffusion, sauvegardons-le dans un dossier local en exécutant :<jupyter_code>image_pipe.save_pretrained("my-finetuned-model")<jupyter_output><empty_output><jupyter_text>Comme nous l'avons vu dans l'unité 1, cela permet de sauvegarder la configuration, le modèle et le planificateur :<jupyter_code>!ls {"my-finetuned-model"}<jupyter_output>model_index.json scheduler unet<jupyter_text>Ensuite, vous pouvez suivre les mêmes étapes que celles décrites dans le *notebook* d'introduction à *Diffusers* de l'unité 1 pour pousser le modèle vers le Hub en vue d'une utilisation ultérieure :<jupyter_code># Code pour télécharger un pipeline sauvegardé localement vers le Hub
from huggingface_hub import HfApi, ModelCard, create_repo, get_full_repo_name
# Mise en place du repo et téléchargement des fichiers
model_name = "ddpm-celebahq-finetuned-butterflies-2epochs" # @param Le nom que vous souhaitez lui donner sur le Hub
local_folder_name = "my-finetuned-model" # @param Créé par le script ou par vous via image_pipe.save_pretrained('save_name')
description = "Describe your model here" # @param
hub_model_id = get_full_repo_name(model_name)
create_repo(hub_model_id)
api = HfApi()
api.upload_folder(
folder_path=f"{local_folder_name}/scheduler", path_in_repo="", repo_id=hub_model_id
)
api.upload_folder(
folder_path=f"{local_folder_name}/unet", path_in_repo="", repo_id=hub_model_id
)
api.upload_file(
path_or_fileobj=f"{local_folder_name}/model_index.json",
path_in_repo="model_index.json",
repo_id=hub_model_id,
)
# Ajouter une carte modèle (facultatif mais sympa !)
content = f"""
---
license: mit
tags:
- pytorch
- diffusers
- unconditional-image-generation
- diffusion-models-class
---
# Example Fine-Tuned Model for Unit 2 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class)
{description}
## Usage
```python
from diffusers import DDPMPipeline
pipeline = DDPMPipeline.from_pretrained('{hub_model_id}')
image = pipeline().images[0]
image
```
"""
card = ModelCard(content)
card.push_to_hub(hub_model_id)<jupyter_output><empty_output><jupyter_text>Félicitations, vous avez maintenant *finetuné* votre premier modèle de diffusion !Pour le reste de ce notebook, nous utiliserons un [modèle](https://huggingface.co/johnowhitaker/sd-class-wikiart-from-bedrooms) que nous avons *finetuné* à partir d'un modèle entraîné sur [LSUN bedrooms](https://huggingface.co/google/ddpm-bedroom-256) environ une fois sur le [WikiArt dataset](https://huggingface.co/datasets/huggan/wikiart). Si vous préférez, vous pouvez sauter cette cellule et utiliser le pipeline faces/butterflies que nous avons *finetuné* dans la section précédente ou en charger un depuis le Hub à la place :<jupyter_code># Chargement du pipeline pré-entraîné
pipeline_name = "johnowhitaker/sd-class-wikiart-from-bedrooms"
image_pipe = DDPMPipeline.from_pretrained(pipeline_name).to(device)
# Échantillon d'images avec un planificateur DDIM sur 40 étapes
scheduler = DDIMScheduler.from_pretrained(pipeline_name)
scheduler.set_timesteps(num_inference_steps=40)
# Point de départ aléatoire (batch de 8 images)
x = torch.randn(8, 3, 256, 256).to(device)
# Boucle d'échantillonnage minimale
for i, t in tqdm(enumerate(scheduler.timesteps)):
model_input = scheduler.scale_model_input(x, t)
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
x = scheduler.step(noise_pred, t, x).prev_sample
# Voir les résultats
grid = torchvision.utils.make_grid(x, nrow=4)
plt.imshow(grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5);<jupyter_output><empty_output><jupyter_text>**Considération 5** : Il est souvent difficile de savoir si le *finetuné* fonctionne bien, et ce que l'on entend par "bonnes performances" peut varier selon le cas d'utilisation. Par exemple, si vous *finetuné* un modèle conditionné par du texte comme Stable Diffusion sur un petit jeu de données, vous voudrez probablement qu'il conserve la plus grande partie de son apprentissage original afin de pouvoir comprendre des prompts arbitraires non couverts par votre nouveau jeu de données, tout en s'adaptant pour mieux correspondre au style de vos nouvelles données d'entraînement. Cela pourrait signifier l'utilisation d'un faible taux d'apprentissage avec quelque chose comme la moyenne exponentielle du modèle, comme démontré dans cet excellent [article de blog](https://lambdalabs.com/blog/how-to-fine-tune-stable-diffusion-how-we-made-the-text-to-pokemon-model-at-lambda) sur la création d'une version Pokemon de Stable Diffusion. Dans une autre situation, vous pouvez vouloir ré-entraîner complètement un modèle sur de nouvelles données (comme notre exemple chambre -> wikiart), auquel cas un taux d'apprentissage plus élevé et un entraînement plus poussé s'avèrent judicieux. Même si le [graphique de la perte] (https://wandb.ai/johnowhitaker/dm_finetune/runs/2upaa341) ne montre pas beaucoup d'amélioration, les échantillons s'éloignent clairement des données d'origine et s'orientent vers des résultats plus "artistiques", bien qu'ils restent pour la plupart incohérents.Ce qui nous amène à la section suivante, où nous examinons comment nous pourrions ajouter des conseils supplémentaires à un tel modèle pour mieux contrôler les résultats. GuidageQue faire si l'on souhaite exercer un certain contrôle sur les échantillons générés ? Par exemple, supposons que nous voulions biaiser les images générées pour qu'elles soient d'une couleur spécifique. Comment procéder ? C'est là qu'intervient le guidage, une technique qui permet d'ajouter un contrôle supplémentaire au processus d'échantillonnage. La première étape consiste à créer notre fonction de conditionnement : une mesure (perte) que nous souhaitons minimiser. En voici une pour l'exemple de la couleur, qui compare les pixels d'une image à une couleur cible (par défaut, une sorte de sarcelle claire) et renvoie l'erreur moyenne :<jupyter_code>def color_loss(images, target_color=(0.1, 0.9, 0.5)):
"""Étant donné une couleur cible (R, G, B), retourner une perte correspondant à la distance moyenne entre
les pixels de l'image et cette couleur. Par défaut, il s'agit d'une couleur sarcelle claire : (0.1, 0.9, 0.5)"""
target = (
torch.tensor(target_color).to(images.device) * 2 - 1
) # Map target color to (-1, 1)
target = target[
None, :, None, None
] # Obtenir la forme nécessaire pour fonctionner avec les images (b, c, h, w)
error = torch.abs(
images - target
).mean() # Différence absolue moyenne entre les pixels de l'image et la couleur cible
return error<jupyter_output><empty_output><jupyter_text>Ensuite, nous allons créer une version modifiée de la boucle d'échantillonnage où, à chaque étape, nous ferons ce qui suit :- Créer une nouvelle version de `x` avec `requires_grad = True`- Calculer la version débruitée (`x0`)- Introduire la version prédite `x0` dans notre fonction de perte- Trouver le gradient de cette fonction de perte par rapport à `x`- Utiliser ce gradient de conditionnement pour modifier `x` avant d'utiliser le planificateur, en espérant pousser x dans une direction qui conduira à une perte plus faible selon notre fonction d'orientation.Il existe deux variantes que vous pouvez explorer. Dans la première, nous fixons `requires_grad` sur `x` après avoir obtenu notre prédiction de bruit du UNet, ce qui est plus efficace en termes de mémoire (puisque nous n'avons pas à retracer les gradients à travers le modèle de diffusion), mais donne un gradient moins précis. Dans le second cas, nous définissons d'abord `requires_grad` sur `x`, puis nous le faisons passer par l'unet et nous calculons le `x0` prédit.<jupyter_code># Variante 1 : méthode rapide
# L'échelle de guidance détermine l'intensité de l'effet
guidance_loss_scale = 40 # Envisagez de modifier cette valeur à 5, ou à 100
x = torch.randn(8, 3, 256, 256).to(device)
for i, t in tqdm(enumerate(scheduler.timesteps)):
# Préparer l'entrée du modèle
model_input = scheduler.scale_model_input(x, t)
# Prédire le bruit résiduel
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
# Fixer x.requires_grad à True
x = x.detach().requires_grad_()
# Obtenir la valeur prédite x0
x0 = scheduler.step(noise_pred, t, x).pred_original_sample
# Calculer la perte
loss = color_loss(x0) * guidance_loss_scale
if i % 10 == 0:
print(i, "loss:", loss.item())
# Obtenir le gradient
cond_grad = -torch.autograd.grad(loss, x)[0]
# Modifier x en fonction de ce gradient
x = x.detach() + cond_grad
# Le planificateur
x = scheduler.step(noise_pred, t, x).prev_sample
# Voir le résultat
grid = torchvision.utils.make_grid(x, nrow=4)
im = grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5
Image.fromarray(np.array(im * 255).astype(np.uint8))<jupyter_output><empty_output><jupyter_text>Cette deuxième option nécessite presque le double de RAM GPU pour fonctionner, même si nous ne générons qu'un batch de quatre images au lieu de huit. Voyez si vous pouvez repérer la différence et réfléchissez à la raison pour laquelle cette méthode est plus "précise" :<jupyter_code># Variante 2 : définir x.requires_grad avant de calculer les prédictions du modèle
guidance_loss_scale = 40
x = torch.randn(4, 3, 256, 256).to(device)
for i, t in tqdm(enumerate(scheduler.timesteps)):
# Définir requires_grad avant la passe avant du modèle
x = x.detach().requires_grad_()
model_input = scheduler.scale_model_input(x, t)
# prédire (avec grad cette fois)
noise_pred = image_pipe.unet(model_input, t)["sample"]
# Obtenir la valeur prédite x0 :
x0 = scheduler.step(noise_pred, t, x).pred_original_sample
# Calculer la perte
loss = color_loss(x0) * guidance_loss_scale
if i % 10 == 0:
print(i, "loss:", loss.item())
# Obtenir le gradient
cond_grad = -torch.autograd.grad(loss, x)[0]
# Modifier x en fonction de ce gradient
x = x.detach() + cond_grad
# Le planificateur
x = scheduler.step(noise_pred, t, x).prev_sample
grid = torchvision.utils.make_grid(x, nrow=4)
im = grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5
Image.fromarray(np.array(im * 255).astype(np.uint8))<jupyter_output><empty_output><jupyter_text>Dans la seconde variante, les besoins en mémoire sont plus importants et l'effet est moins prononcé, de sorte que vous pouvez penser qu'elle est inférieure. Cependant, les résultats sont sans doute plus proches des types d'images sur lesquels le modèle a été entraîné, et vous pouvez toujours augmenter l'échelle de guidage pour obtenir un effet plus important. L'approche que vous utiliserez dépendra en fin de compte de ce qui fonctionne le mieux sur le plan expérimental. **Exercice :** choisissez votre couleur préférée et recherchez ses valeurs dans l'espace RGB.Modifiez la ligne `color_loss()` dans la cellule ci-dessus pour recevoir ces nouvelles valeurs RGB et examinez les résultats ; correspondent-ils à ce que vous attendez ? Guidage avec CLIPGuider vers une couleur nous donne un peu de contrôle, mais que se passerait-il si nous pouvions simplement taper un texte décrivant ce que nous voulons ?[CLIP](https://openai.com/research/clip) est un modèle créé par OpenAI qui nous permet de comparer des images à des légendes textuelles. C'est extrêmement puissant, car cela nous permet de quantifier à quel point une image correspond à un prompt. Et comme le processus est différentiable, nous pouvons l'utiliser comme fonction de perte pour guider notre modèle de diffusion !Nous n'entrerons pas dans les détails ici. L'approche de base est la suivante :- Enchâsser le prompt pour obtenir un enchâssement CLIP à 512 dimensions- Pour chaque étape du processus du modèle de diffusion : - Créer plusieurs variantes de l'image débruitée prédite (le fait d'avoir plusieurs variantes permet d'obtenir un signal de perte plus propre). - Pour chacune d'entre elles, enchâsser l'image avec CLIP et comparez cet enchâssement avec celui du prompt (à l'aide d'une mesure appelée "distance du grand cercle").- Calculer le gradient de cette perte par rapport à l'image bruyante actuelle x et utiliser ce gradient pour modifier x avant de le mettre à jour avec le planificateur.Pour une explication plus approfondie de CLIP, consultez cette [leçon sur le sujet] (https://johnowhitaker.github.io/tglcourse/clip.html) ou ce [rapport sur le projet OpenCLIP] (https://wandb.ai/johnowhitaker/openclip-benchmarking/reports/Exploring-OpenCLIP--VmlldzoyOTIzNzIz) que nous utilisons pour charger le modèle CLIP. Exécutez la cellule suivante pour charger un modèle CLIP :<jupyter_code>import open_clip
clip_model, _, preprocess = open_clip.create_model_and_transforms(
"ViT-B-32", pretrained="openai"
)
clip_model.to(device)
# Transformations pour redimensionner et augmenter une image + normalisation pour correspondre aux données entraînées par CLIP
tfms = torchvision.transforms.Compose(
[
torchvision.transforms.RandomResizedCrop(224), # CROP aléatoire à chaque fois
torchvision.transforms.RandomAffine(
5
), # Une augmentation aléatoire possible : biaiser l'image
torchvision.transforms.RandomHorizontalFlip(), # Vous pouvez ajouter des augmentations supplémentaires si vous le souhaitez
torchvision.transforms.Normalize(
mean=(0.48145466, 0.4578275, 0.40821073),
std=(0.26862954, 0.26130258, 0.27577711),
),
]
)
# Et définir une fonction de perte qui prend une image, l'enchâsse et la compare avec les caractéristiques textuelles du prompt
def clip_loss(image, text_features):
image_features = clip_model.encode_image(
tfms(image)
) # Note : applique les transformations ci-dessus
input_normed = torch.nn.functional.normalize(image_features.unsqueeze(1), dim=2)
embed_normed = torch.nn.functional.normalize(text_features.unsqueeze(0), dim=2)
dists = (
input_normed.sub(embed_normed).norm(dim=2).div(2).arcsin().pow(2).mul(2)
) # Distance du grand cercle
return dists.mean()<jupyter_output>100%|████████████████████████████████████████| 354M/354M [00:02<00:00, 120MiB/s]<jupyter_text>Une fois la fonction de perte définie, notre boucle d'échantillonnage guidé ressemble aux exemples précédents, en remplaçant `color_loss()` par notre nouvelle fonction de perte basée sur CLIP :<jupyter_code>prompt = "Red Rose (still life), red flower painting" # @param
# Explorer en changeant ça
guidance_scale = 8 # @param
n_cuts = 4 # @param
# Plus d'étapes -> plus de temps pour que le guidage ait un effet
scheduler.set_timesteps(50)
# Nous enchâssons un prompt avec CLIP comme cible
text = open_clip.tokenize([prompt]).to(device)
with torch.no_grad(), torch.cuda.amp.autocast():
text_features = clip_model.encode_text(text)
x = torch.randn(4, 3, 256, 256).to(
device
) # L'utilisation de la RAM est élevée, vous ne voulez peut-être qu'une seule image à la fois.
for i, t in tqdm(enumerate(scheduler.timesteps)):
model_input = scheduler.scale_model_input(x, t)
# prédire le bruit résiduel
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
cond_grad = 0
for cut in range(n_cuts):
# nécessite un grad sur x
x = x.detach().requires_grad_()
# Obtenir le x0 prédit
x0 = scheduler.step(noise_pred, t, x).pred_original_sample
# Calculer la perte
loss = clip_loss(x0, text_features) * guidance_scale
# Obtenir le gradient (échelle par n_cuts puisque nous voulons la moyenne)
cond_grad -= torch.autograd.grad(loss, x)[0] / n_cuts
if i % 25 == 0:
print("Step:", i, ", Guidance loss:", loss.item())
# Modifier x en fonction de ce gradient
alpha_bar = scheduler.alphas_cumprod[i]
x = (
x.detach() + cond_grad * alpha_bar.sqrt()
) # Note the additional scaling factor here!
# Le planificateur
x = scheduler.step(noise_pred, t, x).prev_sample
grid = torchvision.utils.make_grid(x.detach(), nrow=4)
im = grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5
Image.fromarray(np.array(im * 255).astype(np.uint8))<jupyter_output><empty_output><jupyter_text>Cela ressemble un peu à des roses ! Ce n'est pas parfait, mais si vous jouez avec les paramètres, vous pouvez obtenir des images agréables. Si vous examinez le code ci-dessus, vous verrez que nous mettons à l'échelle le gradient de conditionnement par un facteur de `alpha_bar.sqrt()`. Il existe des théories sur la "bonne" manière d'échelonner ces gradients, mais en pratique, vous pouvez expérimenter. Pour certains types de guidage, vous voudrez peut-être que la plupart des effets soient concentrés dans les premières étapes, pour d'autres (par exemple, une perte de style axée sur les textures), vous préférerez peut-être qu'ils n'interviennent que vers la fin du processus de génération. Quelques programmes possibles sont présentés ci-dessous :<jupyter_code>plt.plot([1 for a in scheduler.alphas_cumprod], label="no scaling")
plt.plot([a for a in scheduler.alphas_cumprod], label="alpha_bar")
plt.plot([a.sqrt() for a in scheduler.alphas_cumprod], label="alpha_bar.sqrt()")
plt.plot(
[(1 - a).sqrt() for a in scheduler.alphas_cumprod], label="(1-alpha_bar).sqrt()"
)
plt.legend()
plt.title("Possible guidance scaling schedules");<jupyter_output><empty_output><jupyter_text>Expérimentez avec différents plannificateurs, échelles de guidage et toute autre astuce à laquelle vous pouvez penser (l'écrêtage des gradients dans une certaine plage est une modification populaire) pour voir jusqu'à quel point vous pouvez obtenir ce résultat ! N'oubliez pas non plus d'essayer d'intervertir d'autres modèles. Peut-être le modèle de visages que nous avons chargé au début ; pouvez-vous le guider de manière fiable pour produire un visage masculin ? Que se passe-t-il si vous combinez le guidage CLIP avec la perte de couleur que nous avons utilisée plus tôt ? Etc.Si vous consultez [quelques codes pour la diffusion guidée par CLIP en pratique](https://huggingface.co/spaces/EleutherAI/clip-guided-diffusion/blob/main/app.py), vous verrez une approche plus complexe avec une meilleure classe pour choisir des découpes aléatoires dans les images et de nombreux ajustements supplémentaires de la fonction de perte pour de meilleures performances. Avant l'apparition des modèles de diffusion conditionnés par le texte, il s'agissait du meilleur système de conversion texte-image qui soit ! La petite version de notre jouet peut encore être améliorée, mais elle capture l'idée principale : grâce au guidage et aux capacités étonnantes de CLIP, nous pouvons ajouter le contrôle du texte à un modèle de diffusion inconditionnel 🎨. Partager une boucle d'échantillonnage personnalisée en tant que démo GradioVous avez peut-être trouvé une perte amusante pour guider la génération et vous souhaitez maintenant partager avec le monde entier votre modèle *finetuné* et cette stratégie d'échantillonnage personnalisée... Entrez dans [Gradio](https://gradio.app/). Gradio est un outil gratuit et open-source qui permet aux utilisateurs de créer et de partager facilement des modèles interactifs d'apprentissage automatique via une simple interface web. Avec Gradio, les utilisateurs peuvent construire des interfaces personnalisées pour leurs modèles d'apprentissage automatique, qui peuvent ensuite être partagés avec d'autres par le biais d'une URL unique. Il est également intégré à 🤗 Spaces, ce qui permet d'héberger facilement des démos et de les partager avec d'autres.Nous placerons notre logique de base dans une fonction qui prend certaines entrées et produit une image en sortie. Cette fonction peut ensuite être enveloppée dans une interface simple qui permet à l'utilisateur de spécifier certains paramètres (qui sont transmis en tant qu'entrées à la fonction principale de génération). De nombreux [composants](https://gradio.app/docs/components) sont disponibles ; pour cet exemple, nous utiliserons un curseur pour l'échelle d'orientation et un sélecteur de couleurs pour définir la couleur cible.<jupyter_code>!pip install -q gradio
import gradio as gr
from PIL import Image, ImageColor
# La fonction qui fait le gros du travail
def generate(color, guidance_loss_scale):
target_color = ImageColor.getcolor(color, "RGB") # Couleur cible en RGB
target_color = [a / 255 for a in target_color] # Rééchelonner de (0, 255) à (0, 1)
x = torch.randn(1, 3, 256, 256).to(device)
for i, t in tqdm(enumerate(scheduler.timesteps)):
model_input = scheduler.scale_model_input(x, t)
with torch.no_grad():
noise_pred = image_pipe.unet(model_input, t)["sample"]
x = x.detach().requires_grad_()
x0 = scheduler.step(noise_pred, t, x).pred_original_sample
loss = color_loss(x0, target_color) * guidance_loss_scale
cond_grad = -torch.autograd.grad(loss, x)[0]
x = x.detach() + cond_grad
x = scheduler.step(noise_pred, t, x).prev_sample
grid = torchvision.utils.make_grid(x, nrow=4)
im = grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5
im = Image.fromarray(np.array(im * 255).astype(np.uint8))
im.save("test.jpeg")
return im
# Voir la documentation de gradio pour les types d'entrées et de sorties disponibles.
inputs = [
gr.ColorPicker(label="color", value="55FFAA"), # Ajoutez ici toutes les entrées dont vous avez besoin
gr.Slider(label="guidance_scale", minimum=0, maximum=30, value=3),
]
outputs = gr.Image(label="result")
# Et l'interface minimale
demo = gr.Interface(
fn=generate,
inputs=inputs,
outputs=outputs,
examples=[
["#BB2266", 3],
["#44CCAA", 5], # Vous pouvez fournir des exemples d'entrées pour aider les gens à démarrer
],
)
demo.launch(debug=True) # debug=True vous permet de voir les erreurs et les sorties dans Colab<jupyter_output><empty_output> | diffusion-models-class/units/fr/unit2/finetuning_and_guidance.ipynb/0 | {
"file_path": "diffusion-models-class/units/fr/unit2/finetuning_and_guidance.ipynb",
"repo_id": "diffusion-models-class",
"token_count": 15878
} | 138 |
<jupyter_start><jupyter_text>Recherche sémantique avec FAISS (PyTorch) Installez les bibliothèques 🤗 Transformers et 🤗 Datasets pour exécuter ce *notebook*.<jupyter_code>!pip install datasets evaluate transformers[sentencepiece]
!pip install faiss-gpu
from huggingface_hub import hf_hub_url
data_files = hf_hub_url(
repo_id="lewtun/github-issues",
filename="datasets-issues-with-comments.jsonl",
repo_type="dataset",
)
from datasets import load_dataset
issues_dataset = load_dataset("json", data_files=data_files, split="train")
issues_dataset
issues_dataset = issues_dataset.filter(
lambda x: (x["is_pull_request"] == False and len(x["comments"]) > 0)
)
issues_dataset
columns = issues_dataset.column_names
columns_to_keep = ["title", "body", "html_url", "comments"]
columns_to_remove = set(columns_to_keep).symmetric_difference(columns)
issues_dataset = issues_dataset.remove_columns(columns_to_remove)
issues_dataset
issues_dataset.set_format("pandas")
df = issues_dataset[:]
df["comments"][0].tolist()
comments_df = df.explode("comments", ignore_index=True)
comments_df.head(4)
from datasets import Dataset
comments_dataset = Dataset.from_pandas(comments_df)
comments_dataset
comments_dataset = comments_dataset.map(
lambda x: {"comment_length": len(x["comments"].split())}
)
comments_dataset = comments_dataset.filter(lambda x: x["comment_length"] > 15)
comments_dataset
def concatenate_text(examples):
return {
"text": examples["title"]
+ " \n "
+ examples["body"]
+ " \n "
+ examples["comments"]
}
comments_dataset = comments_dataset.map(concatenate_text)
from transformers import AutoTokenizer, AutoModel
model_ckpt = "sentence-transformers/multi-qa-mpnet-base-dot-v1"
tokenizer = AutoTokenizer.from_pretrained(model_ckpt)
model = AutoModel.from_pretrained(model_ckpt)
import torch
device = torch.device("cuda")
model.to(device)
def cls_pooling(model_output):
return model_output.last_hidden_state[:, 0]
def get_embeddings(text_list):
encoded_input = tokenizer(
text_list, padding=True, truncation=True, return_tensors="pt"
)
encoded_input = {k: v.to(device) for k, v in encoded_input.items()}
model_output = model(**encoded_input)
return cls_pooling(model_output)
embedding = get_embeddings(comments_dataset["text"][0])
embedding.shape
embeddings_dataset = comments_dataset.map(
lambda x: {"embeddings": get_embeddings(x["text"]).detach().cpu().numpy()[0]}
)
embeddings_dataset.add_faiss_index(column="embeddings")
question = "How can I load a dataset offline?"
question_embedding = get_embeddings([question]).cpu().detach().numpy()
question_embedding.shape
scores, samples = embeddings_dataset.get_nearest_examples(
"embeddings", question_embedding, k=5
)
import pandas as pd
samples_df = pd.DataFrame.from_dict(samples)
samples_df["scores"] = scores
samples_df.sort_values("scores", ascending=False, inplace=True)
for _, row in samples_df.iterrows():
print(f"COMMENT: {row.comments}")
print(f"SCORE: {row.scores}")
print(f"TITLE: {row.title}")
print(f"URL: {row.html_url}")
print("=" * 50)
print()<jupyter_output><empty_output> | notebooks/course/fr/chapter5/section6_pt.ipynb/0 | {
"file_path": "notebooks/course/fr/chapter5/section6_pt.ipynb",
"repo_id": "notebooks",
"token_count": 1233
} | 139 |
<jupyter_start><jupyter_text>Finetuner un modèle de language masqué (TensorFlow) Installez les bibliothèques 🤗 *Datasets* et 🤗 *Transformers* pour exécuter ce *notebook*.<jupyter_code>!pip install datasets transformers[sentencepiece]
!apt install git-lfs<jupyter_output><empty_output><jupyter_text>Vous aurez besoin de configurer git, adaptez votre email et votre nom dans la cellule suivante.<jupyter_code>!git config --global user.email "[email protected]"
!git config --global user.name "Your Name"<jupyter_output><empty_output><jupyter_text>Vous devrez également être connecté au Hub d'Hugging Face. Exécutez ce qui suit et entrez vos informations d'identification.<jupyter_code>from huggingface_hub import notebook_login
notebook_login()
from transformers import TFAutoModelForMaskedLM
model_checkpoint = "camembert-base"
model = TFAutoModelForMaskedLM.from_pretrained(model_checkpoint)
model.summary()
text = "C'est une grande <mask>."
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)
inputs = tokenizer(text, return_tensors="np")
inputs
import numpy as np
import tensorflow as tf
inputs = tokenizer(text, return_tensors="np")
token_logits = model(**inputs).logits
# Trouver l'emplacement du [MASK] et extraire ses logits
mask_token_index = np.argwhere(inputs["input_ids"] == tokenizer.mask_token_id)[0, 1]
mask_token_logits = token_logits[0, mask_token_index, :]
# Choisir les <mask> candidats avec les logits les plus élevés
# Nous annulons le tableau avant argsort pour obtenir le plus grand, pas le plus petit, logits
top_5_tokens = np.argsort(-mask_token_logits)[:5].tolist()
for token in top_5_tokens:
print(f">>> {text.replace(tokenizer.mask_token, tokenizer.decode([token]))}")
from datasets import load_dataset
imdb_dataset = load_dataset("allocine")
imdb_dataset
sample = imdb_dataset["train"].shuffle(seed=42).select(range(3))
for row in sample:
print(f"\n'>>> Review: {row['review']}'")
print(f"'>>> Label: {row['label']}'")
def tokenize_function(examples):
result = tokenizer(examples["review"])
if tokenizer.is_fast:
result["word_ids"] = [result.word_ids(i) for i in range(len(result["input_ids"]))]
return result
# Utilisez batched=True pour activer le multithreading rapide !
tokenized_datasets = imdb_dataset.map(
tokenize_function, batched=True, remove_columns=["review", "label"]
)
tokenized_datasets
tokenizer.model_max_length
chunk_size = 128
# Le découpage produit une liste de listes pour chaque caractéristique
tokenized_samples = tokenized_datasets["train"][:3]
for idx, sample in enumerate(tokenized_samples["input_ids"]):
print(f"'>>> Review {idx} length: {len(sample)}'")
concatenated_examples = {
k: sum(tokenized_samples[k], []) for k in tokenized_samples.keys()
}
total_length = len(concatenated_examples["input_ids"])
print(f"'>>> Concatenated reviews length: {total_length}'")
chunks = {
k: [t[i : i + chunk_size] for i in range(0, total_length, chunk_size)]
for k, t in concatenated_examples.items()
}
for chunk in chunks["input_ids"]:
print(f"'>>> Chunk length: {len(chunk)}'")
def group_texts(examples):
# Concaténation de tous les textes
concatenated_examples = {k: sum(examples[k], []) for k in examples.keys()}
# Calculer la longueur des textes concaténés
total_length = len(concatenated_examples[list(examples.keys())[0]])
# Nous laissons tomber le dernier morceau s'il est plus petit que chunk_size
total_length = (total_length // chunk_size) * chunk_size
# Fractionnement par morceaux de max_len
result = {
k: [t[i : i + chunk_size] for i in range(0, total_length, chunk_size)]
for k, t in concatenated_examples.items()
}
# Créer une nouvelle colonne d'étiquettes
result["labels"] = result["input_ids"].copy()
return result
lm_datasets = tokenized_datasets.map(group_texts, batched=True)
lm_datasets
tokenizer.decode(lm_datasets["train"][1]["input_ids"])
from transformers import DataCollatorForLanguageModeling
data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm_probability=0.15)
samples = [lm_datasets["train"][i] for i in range(2)]
for sample in samples:
_ = sample.pop("word_ids")
for chunk in data_collator(samples)["input_ids"]:
print(f"\n'>>> {tokenizer.decode(chunk)}'")
import collections
import numpy as np
from transformers.data.data_collator import tf_default_data_collator
wwm_probability = 0.2
def whole_word_masking_data_collator(features):
for feature in features:
word_ids = feature.pop("word_ids")
# Création d'une correspondance entre les mots et les indices des tokens correspondants
mapping = collections.defaultdict(list)
current_word_index = -1
current_word = None
for idx, word_id in enumerate(word_ids):
if word_id is not None:
if word_id != current_word:
current_word = word_id
current_word_index += 1
mapping[current_word_index].append(idx)
# Masquer des mots de façon aléatoire
mask = np.random.binomial(1, wwm_probability, (len(mapping),))
input_ids = feature["input_ids"]
labels = feature["labels"]
new_labels = [-100] * len(labels)
for word_id in np.where(mask)[0]:
word_id = word_id.item()
for idx in mapping[word_id]:
new_labels[idx] = labels[idx]
input_ids[idx] = tokenizer.mask_token_id
feature["labels"] = new_labels
return tf_default_data_collator(features)
samples = [lm_datasets["train"][i] for i in range(2)]
batch = whole_word_masking_data_collator(samples)
for chunk in batch["input_ids"]:
print(f"\n'>>> {tokenizer.decode(chunk)}'")
train_size = 10_000
test_size = int(0.1 * train_size)
downsampled_dataset = lm_datasets["train"].train_test_split(
train_size=train_size, test_size=test_size, seed=42
)
downsampled_dataset
from huggingface_hub import notebook_login
notebook_login()
tf_train_dataset = model.prepare_tf_dataset(
downsampled_dataset["train"],
collate_fn=data_collator,
shuffle=True,
batch_size=32,
)
tf_eval_dataset = model.prepare_tf_dataset(
downsampled_dataset["test"],
collate_fn=data_collator,
shuffle=False,
batch_size=32,
)
from transformers import create_optimizer
from transformers.keras_callbacks import PushToHubCallback
import tensorflow as tf
num_train_steps = len(tf_train_dataset)
optimizer, schedule = create_optimizer(
init_lr=2e-5,
num_warmup_steps=1_000,
num_train_steps=num_train_steps,
weight_decay_rate=0.01,
)
model.compile(optimizer=optimizer)
# Entraîner en mixed-precision float16
tf.keras.mixed_precision.set_global_policy("mixed_float16")
callback = PushToHubCallback(
output_dir=f"{model_name}-finetuned-allocine", tokenizer=tokenizer
)
import math
eval_loss = model.evaluate(tf_eval_dataset)
print(f"Perplexity: {math.exp(eval_loss):.2f}")
model.fit(tf_train_dataset, validation_data=tf_eval_dataset, callbacks=[callback])
eval_loss = model.evaluate(tf_eval_dataset)
print(f"Perplexity: {math.exp(eval_loss):.2f}")
from transformers import pipeline
mask_filler = pipeline(
"fill-mask", model="camembert-base-finetuned-allocine", framework="tf"
)
preds = mask_filler(text)
for pred in preds:
print(f">>> {pred['sequence']}")<jupyter_output><empty_output> | notebooks/course/fr/chapter7/section3_tf.ipynb/0 | {
"file_path": "notebooks/course/fr/chapter7/section3_tf.ipynb",
"repo_id": "notebooks",
"token_count": 2949
} | 140 |
<jupyter_start><jupyter_text>Comprendre la classe Interface Installez les bibliothèques 🤗 Transformers et 🤗 Gradio pour exécuter ce *notebook*.<jupyter_code>!pip install datasets transformers[sentencepiece]
!pip install gradio
import numpy as np
import gradio as gr
def reverse_audio(audio):
sr, data = audio
reversed_audio = (sr, np.flipud(data))
return reversed_audio
mic = gr.Audio(source="microphone", type="numpy", label="Parler ici...")
gr.Interface(reverse_audio, mic, "audio").launch()
import numpy as np
import gradio as gr
notes = ["C", "C#", "D", "D#", "E", "F", "F#", "G", "G#", "A", "A#", "B"]
def generate_tone(note, octave, duration):
sr = 48000
a4_freq, tones_from_a4 = 440, 12 * (octave - 4) + (note - 9)
frequency = a4_freq * 2 ** (tones_from_a4 / 12)
duration = int(duration)
audio = np.linspace(0, duration, duration * sr)
audio = (20000 * np.sin(audio * (2 * np.pi * frequency))).astype(np.int16)
return (sr, audio)
gr.Interface(
generate_tone,
[
gr.Dropdown(notes, type="index"),
gr.Slider(minimum=4, maximum=6, step=1),
gr.Textbox(type="number", value=1, label="Durée en secondes"),
],
"audio",
).launch()
from transformers import pipeline
import gradio as gr
model = pipeline("automatic-speech-recognition",model="facebook/wav2vec2-large-xlsr-53-french")
def transcribe_audio(mic=None, file=None):
if mic is not None:
audio = mic
elif file is not None:
audio = file
else:
return "Vous devez fournir soit un enregistrement micro ou un fichier"
transcription = model(audio)["text"]
return transcription
gr.Interface(
fn=transcribe_audio,
inputs=[
gr.Audio(source="microphone", type="filepath", optional=True),
gr.Audio(source="upload", type="filepath", optional=True),
],
outputs="text",
).launch()<jupyter_output><empty_output> | notebooks/course/fr/chapter9/section3.ipynb/0 | {
"file_path": "notebooks/course/fr/chapter9/section3.ipynb",
"repo_id": "notebooks",
"token_count": 759
} | 141 |
<jupyter_start><jupyter_text>Image super-resolution using Latent Diffusion This colab notebook shows how to use the Latent Diffusion image super-resolution model using 🧨 [diffusers](https://github.com/huggingface/diffusers) libray.The model was originally released in [Latent Diffusion repo](https://github.com/CompVis/latent-diffusion). It's a simple, 4x super-resolution model diffusion model. This model is not conditioned on text. Install the Deps<jupyter_code>!pip install -qq diffusers==0.11.1 accelerate<jupyter_output>Installing build dependencies ... [?25l[?25hdone
Getting requirements to build wheel ... [?25l[?25hdone
Preparing wheel metadata ... [?25l[?25hdone
[K |████████████████████████████████| 175 kB 5.1 MB/s
[K |████████████████████████████████| 182 kB 46.5 MB/s
[?25h Building wheel for diffusers (PEP 517) ... [?25l[?25hdone<jupyter_text>Imports<jupyter_code>import torch
from PIL import Image
import requests
from io import BytesIO
from diffusers import LDMSuperResolutionPipeline<jupyter_output><empty_output><jupyter_text>Load the pipeline<jupyter_code>device = "cuda"
pipe = LDMSuperResolutionPipeline.from_pretrained( "CompVis/ldm-super-resolution-4x-openimages")
pipe = pipe.to(device)<jupyter_output><empty_output><jupyter_text>Get the image for demo<jupyter_code># let's download an image
url = "https://i.pinimg.com/236x/af/84/56/af8456faa55d76bd9afa18cd2fd72d58.jpg"
response = requests.get(url)
low_res_img = Image.open(BytesIO(response.content)).convert("RGB")
low_res_img = low_res_img.resize((128, 128))
low_res_img<jupyter_output><empty_output><jupyter_text>Run pipeline to upscale the image<jupyter_code># run pipeline in inference (sample random noise and denoise)
upscaled_image = pipe(low_res_img, num_inference_steps=100, eta=1).images[0]
upscaled_image<jupyter_output><empty_output> | notebooks/diffusers/latent_diffusion_upscaler.ipynb/0 | {
"file_path": "notebooks/diffusers/latent_diffusion_upscaler.ipynb",
"repo_id": "notebooks",
"token_count": 656
} | 142 |
# adapted from https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/pytorch/image_captioning.ipynb
# This example demonstrates normal finetuning (w/o peft) - for the sake of keeping the memory
# requirements small it freezes the original pre-trained text and image layers to keep the memory
# requirements to just 40GB. If you have multiple GPUs then you can remove the unfreeze part to
# finetune the whole model. Alternatively use the PEFT solution as shown in
# IDEFICS_finetuning_demo.ipynb notebook which requires only 20GB to finetune the whole model.
import torch
import torchvision.transforms as transforms
from datasets import load_dataset
from PIL import Image
from transformers import IdeficsForVisionText2Text, AutoProcessor, Trainer, TrainingArguments
device = "cuda" if torch.cuda.is_available() else "cpu"
checkpoint = "HuggingFaceM4/idefics-9b"
# checkpoint = "HuggingFaceM4/tiny-random-idefics"
processor = AutoProcessor.from_pretrained(checkpoint)
model = IdeficsForVisionText2Text.from_pretrained(checkpoint, torch_dtype=torch.bfloat16).to(device)
# freeze the original text and vision models and finetune only the layers added by IDEFICS
# you can unfreeze the whole model, but it'll require multiple gpus to finetune
model.model.freeze_text_layers()
model.model.freeze_vision_layers()
# help util
def check_inference():
url = "https://huggingface.co/datasets/sayakpaul/sample-datasets/resolve/main/pokemon.png"
prompts = [
url,
"Question: What's on the picture? Answer:",
]
inputs = processor(prompts, return_tensors="pt").to(device)
generated_ids = model.generate(**inputs, max_length=150)
generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(generated_text)
# check generation before finetuning
check_inference()
# well, actually it looks like the model is already aware of pokemon - but this dataset will refine it further
# finetune the model on the pokemon types dataset
ds = load_dataset("GabeHD/pokemon-type-captions")
ds = ds["train"].train_test_split(test_size=0.1)
train_ds = ds["train"]
eval_ds = ds["test"]
def convert_to_rgb(image):
# `image.convert("RGB")` would only work for .jpg images, as it creates a wrong background
# for transparent images. The call to `alpha_composite` handles this case
if image.mode == "RGB":
return image
image_rgba = image.convert("RGBA")
background = Image.new("RGBA", image_rgba.size, (255, 255, 255))
alpha_composite = Image.alpha_composite(background, image_rgba)
alpha_composite = alpha_composite.convert("RGB")
return alpha_composite
def ds_transforms(example_batch):
image_size = processor.image_processor.image_size
image_mean = processor.image_processor.image_mean
image_std = processor.image_processor.image_std
image_transform = transforms.Compose([
convert_to_rgb,
transforms.RandomResizedCrop((image_size, image_size), scale=(0.9, 1.0), interpolation=transforms.InterpolationMode.BICUBIC),
transforms.ToTensor(),
transforms.Normalize(mean=image_mean, std=image_std),
])
prompts = []
for i in range(len(example_batch)):
prompts.append(
[
example_batch["image"][i],
f"Question: What's on the picture? Answer: {example_batch['text'][i]}\n",
],
)
inputs = processor(prompts, transform=image_transform, return_tensors="pt").to(device)
inputs["labels"] = inputs["input_ids"]
return inputs
train_ds.set_transform(ds_transforms)
eval_ds.set_transform(ds_transforms)
model_name = checkpoint.split("/")[1]
# this setup requires about 40GB of gpu memory
training_args = TrainingArguments(
output_dir=f"{model_name}-pokemon",
learning_rate=5e-6,
num_train_epochs=10,
bf16=True,
per_device_train_batch_size=32,
per_device_eval_batch_size=32,
gradient_accumulation_steps=2,
dataloader_pin_memory=False,
save_total_limit=3,
evaluation_strategy="steps",
save_strategy="steps",
save_steps=1000, # don't save until ready...
eval_steps=40,
logging_steps=40,
remove_unused_columns=False,
push_to_hub=False,
label_names=["labels"],
load_best_model_at_end=True,
report_to=None,
)
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_ds,
eval_dataset=eval_ds,
)
trainer.train()
# check generation again after finetuning
check_inference()
# after finetuning ideally we want generate to produce something like: a drawing of a pink and blue pokemon
| notebooks/examples/idefics/finetune_image_captioning.py/0 | {
"file_path": "notebooks/examples/idefics/finetune_image_captioning.py",
"repo_id": "notebooks",
"token_count": 1670
} | 143 |
<jupyter_start><jupyter_text>If you're opening this Notebook on colab, you will probably need to install 🤗 Transformers as well as some other libraries. Uncomment the following cell and run it.<jupyter_code># Install
!pip install -q biopython transformers datasets huggingface_hub accelerate peft<jupyter_output>[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m3.1/3.1 MB[0m [31m43.0 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m56.8/56.8 kB[0m [31m7.4 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m7.2/7.2 MB[0m [31m104.9 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m486.2/486.2 kB[0m [31m43.5 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m236.8/236.8 kB[0m [31m23.1 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m227.6/227.6 kB[0m [31m21.9 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m7.8/7.8 MB[0m [31m49.6 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━[0m [32m1.3/1.3 MB[0m [31m37.2 MB/s[0m eta [36m0:00:00[0m
[2K [90m━━━━━━━━━━━━━━━━━━━━━━━━━━[...]<jupyter_text>If you're opening this notebook locally, make sure your environment has an install from the last version of those libraries.To be able to share your model with the community and generate results like the one shown in the picture below via the inference API, there are a few more steps to follow.First you have to login to the huggingface hub<jupyter_code>from huggingface_hub import notebook_login
notebook_login()<jupyter_output>Token will not been saved to git credential helper. Pass `add_to_git_credential=True` if you want to set the git credential as well.
Token is valid (permission: write).
Your token has been saved to /root/.cache/huggingface/token
Login successful<jupyter_text>Then you need to install Git-LFS. Uncomment the following instructions:<jupyter_code>!apt install git-lfs<jupyter_output>Reading package lists... Done
Building dependency tree
Reading state information... Done
git-lfs is already the newest version (2.9.2-1).
0 upgraded, 0 newly installed, 0 to remove and 13 not upgraded.<jupyter_text>We also quickly upload some telemetry - this tells us which examples and software versions are getting used so we know where to prioritize our maintenance efforts. We don't collect (or care about) any personally identifiable information, but if you'd prefer not to be counted, feel free to skip this step or delete this cell entirely.<jupyter_code>from transformers.utils import send_example_telemetry
send_example_telemetry("nucleotide_transformer_dna_sequence_modeling_with_lora_notebook", framework="pytorch")<jupyter_output><empty_output><jupyter_text>**Fine-Tuning the Nucleotide-transformer with LoRA** The **Nucleotide Transformer** paper [Dalla-torre et al, 2023](https://www.biorxiv.org/content/10.1101/2023.01.11.523679v2) introduces 4 genomics foundational models developed by **InstaDeep**. These transformers, of various sizes and trained on different datasets, allow powerful representations of DNA sequences that allow to tackle a very diverse set of problems such as chromatin accessibility, deleteriousness prediction, promoter and enhancer prediction etc... These representations can be extracted from the transformer and used as proxies of the DNA sequences (this is called probing) or the transformer can be trained further on a specific task (this is called finetuning). This notebook allows you to fine-tune one of these models.[LoRA: Low-Rank Adaptation of Large Language Models](https://arxiv.org/abs/2106.09685) is one of the state of the art parameter-efficient finetuning methods that is explained in details in this [blog post](https://huggingface.co/blog/lora). Any transformer model can be finetuned using this method with very little effort using the 🤗 Transformers library, which is why it is used in this notebook instead of the [IA³ technique](https://arxiv.org/abs/2205.05638) presented in the original paper.The model we are going to use is the [500M Human Ref model](https://huggingface.co/InstaDeepAI/nucleotide-transformer-500m-1000g), which is a 500M parameters transformer pre-trained on the human reference genome, per the training methodology presented in the Nucleotide Transformer Paper. It is one of the 4 models introduced, all available on the [Instadeep HuggingFace page](https://huggingface.co/InstaDeepAI):```| Model name | Num layers | Num parameters | Training dataset ||---------------------|------------|----------------|------------------------|| `500M Human Ref` | 24 | 500M | Human reference genome || `500M 1000G` | 24 | 500M | 1000G genomes || `2.5B 1000G` | 32 | 2.5B | 1000G genomes || `2.5B Multispecies` | 32 | 2.5B | Multi-species dataset |```Note that even though the finetuning is done with a parameter-efficient method, using the larger checkpoints will still require more GPU memory and produce longer finetuning timesIn the following, we showcase the nucleotide transformer ability to classify genomic sequences as two of the most basic genomic motifs: **promoters** and **enhancers types**. Both of them are classification task, but the enhancers types task is much more challenging with its 3 classes.These two tasks are very basic, but the nucleotide transformers have been shown to beat/match state of the art models on much more complex tasks such as [DeepSEA](https://www.nature.com/articles/nmeth.3547), which, given a DNA sequence, predicts 919 chromatin profiles from a diverse set of human cells and tissues from a single sequence or [DeepSTARR](https://www.nature.com/articles/s41588-022-01048-5), which predicts an enhancer's activity. **Importing required packages and setting up PEFT model** **Import and install**<jupyter_code># Imports
from transformers import AutoTokenizer, AutoModelForMaskedLM, TrainingArguments, Trainer, AutoModelForSequenceClassification
import torch
from sklearn.metrics import matthews_corrcoef, f1_score
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
import numpy as np
# Define the working device
device = torch.device("cuda")<jupyter_output><empty_output><jupyter_text>**Prepare and create the model for fine-tuning** The nucleotide transformer will be fine-tuned on two **classification tasks**: promoter and enhancer types classification.The `AutoModelForSequenceClassification` module automatically loads the model and adds a simple classification head on top of the final embeddings.<jupyter_code>num_labels_promoter = 2
# Load the model
model = AutoModelForSequenceClassification.from_pretrained("InstaDeepAI/nucleotide-transformer-500m-human-ref", num_labels=num_labels_promoter)
model = model.to(device)<jupyter_output>Some weights of the model checkpoint at InstaDeepAI/nucleotide-transformer-500m-human-ref were not used when initializing EsmForSequenceClassification: ['lm_head.layer_norm.weight', 'lm_head.bias', 'lm_head.dense.bias', 'lm_head.decoder.weight', 'lm_head.dense.weight', 'lm_head.layer_norm.bias']
- This IS expected if you are initializing EsmForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing EsmForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
Some weights of EsmForSequenceClassification were not initialized from the model checkpoint at InstaDeepAI/nucleotide-transformer-500m-human-ref and are newly initialized: ['classifier.out_proj.bias', 'classifier[...]<jupyter_text>The LoRA parameters are now added to the model, and the parameters that will be finetuned are indicated.<jupyter_code>from peft import LoraConfig, TaskType
peft_config = LoraConfig(
task_type=TaskType.SEQ_CLS, inference_mode=False, r=1, lora_alpha= 32, lora_dropout=0.1, target_modules= ["query", "value"],
#modules_to_save=["intermediate"] # modules that are not frozen and updated during the training
)
from peft import get_peft_model
lora_classifier = get_peft_model(model, peft_config) # transform our classifier into a peft model
lora_classifier.print_trainable_parameters()
lora_classifier.to(device) # Put the model on the GPU<jupyter_output>trainable params: 3407364 || all params: 482205925 || trainable%: 0.7066201021897439<jupyter_text>**First task : Promoter prediction** Promoter prediction is a **sequence classification** problem, in which the DNA sequence is predicted to be either a promoter or not.A promoter is a region of DNA where transcription of a gene is initiated. Promoters are a vital component of expression vectors because they control the binding of RNA polymerase to DNA. RNA polymerase transcribes DNA to mRNA which is ultimately translated into a functional protein This task was introduced in [DeePromoter](https://www.frontiersin.org/articles/10.3389/fgene.2019.00286/full), where a set of TATA and non-TATA promoters was gathered. A negative sequence was generated from each promoter, by randomly sampling subsets of the sequence, to guarantee that some obvious motifs were present both in the positive and negative dataset. **Dataset loading and preparation**<jupyter_code>from datasets import load_dataset, Dataset
# Load the promoter dataset from the InstaDeep Hugging Face ressources
dataset_name = "promoter_all"
train_dataset_promoter = load_dataset(
"InstaDeepAI/nucleotide_transformer_downstream_tasks",
dataset_name,
split="train",
streaming= False,
)
test_dataset_promoter = load_dataset(
"InstaDeepAI/nucleotide_transformer_downstream_tasks",
dataset_name,
split="test",
streaming= False,
)
# Get training data
train_sequences_promoter = train_dataset_promoter['sequence']
train_labels_promoter = train_dataset_promoter['label']
# Split the dataset into a training and a validation dataset
train_sequences_promoter, validation_sequences_promoter, train_labels_promoter, validation_labels_promoter = train_test_split(train_sequences_promoter,
train_labels_promoter, test_size=0.05, random_state=42)
# Get test data
test_sequences_promoter = test_dataset_promoter['sequence']
test_labels_promoter = test_dataset_promoter['label']<jupyter_output><empty_output><jupyter_text>Let us have a look at the data. If we extract the last sequence of the dataset, we see that it is indeed a promoter, as its label is 1. Furthermore, we can also see that it is a TATA promoter, as the TATA motif is present at the 221th nucleotide of the sequence!<jupyter_code>idx_sequence = -1
sequence, label = train_sequences_promoter[idx_sequence], train_labels_promoter[idx_sequence]
print(f"The DNA sequence is {sequence}.")
print(f"Its associated label is label {label}.")
idx_TATA = sequence.find("TATA")
print(f"This promoter is a TATA promoter, as the TATA motif is present at the {idx_TATA}th nucleotide.")<jupyter_output>The DNA sequence is CACACCAGACAAAATTTGGTTAATTTGCGCCCAATATTCATTACTTTGACCTAACCTTTGTTCTGAAGGCCGTGTACAAGGACAAGGCCCTGAGATTATTGCAACAGTAACTTGAAAAACTTTCAGAAGTCTATTCTGTAGGATTAAAGGAATGCTGAGACTATTCAAGTTTGAAGTCCTGGGGGTGGGGAAAAATAAAAAACCTGTGCTAGAAAGCTTAGTATAGCATGTAACTTTAGAGTCCTGTGGAGTCCTGAGTCTCCCACAGACCAGAACAGTCATTTAAAAGTTTTCAGGAAA.
Its associated label is label 1.
This promoter is a TATA promoter, as the TATA motif is present at the 221th nucleotide.<jupyter_text>**Tokenizing the datasets** All inputs to neural nets must be numerical. The process of converting strings into numerical indices suitable for a neural net is called **tokenization**.<jupyter_code># Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained("InstaDeepAI/nucleotide-transformer-500m-human-ref")
# Promoter dataset
ds_train_promoter = Dataset.from_dict({"data": train_sequences_promoter,'labels':train_labels_promoter})
ds_validation_promoter = Dataset.from_dict({"data": validation_sequences_promoter,'labels':validation_labels_promoter})
ds_test_promoter = Dataset.from_dict({"data": test_sequences_promoter,'labels':test_labels_promoter})
def tokenize_function(examples):
outputs = tokenizer(examples["data"])
return outputs
# Creating tokenized promoter dataset
tokenized_datasets_train_promoter = ds_train_promoter.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)
tokenized_datasets_validation_promoter = ds_validation_promoter.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)
tokenized_datasets_test_promoter = ds_test_promoter.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)<jupyter_output><empty_output><jupyter_text>**Fine-tuning and evaluation** We initialize our `TrainingArguments`. These control the various training hyperparameters, and will be passed to our `Trainer`.The hyperparameters used for the IA³ method in the paper do not provide good performance for the LoRa method. Mainly, LoRA introduces more trainable parameters, therefore requiring a smaller learning rate. We here use a learning rate of 5.10⁻⁴, which enables us to get close to the [**paper's**](https://www.biorxiv.org/content/10.1101/2023.01.11.523679v1.full.pdf) performance.<jupyter_code>batch_size = 8
model_name='nucleotide-transformer'
args_promoter = TrainingArguments(
f"{model_name}-finetuned-lora-NucleotideTransformer",
remove_unused_columns=False,
evaluation_strategy="steps",
save_strategy="steps",
learning_rate=5e-4,
per_device_train_batch_size=batch_size,
gradient_accumulation_steps= 1,
per_device_eval_batch_size= 64,
num_train_epochs= 2,
logging_steps= 100,
load_best_model_at_end=True, # Keep the best model according to the evaluation
metric_for_best_model="f1_score",
label_names=["labels"],
dataloader_drop_last=True,
max_steps= 1000
)<jupyter_output><empty_output><jupyter_text>Next, we define the metric we will use to evaluate our models and write a `compute_metrics` function. We can load this from the `scikit-learn` library.<jupyter_code># Define the metric for the evaluation using the f1 score
def compute_metrics_f1_score(eval_pred):
"""Computes F1 score for binary classification"""
predictions = np.argmax(eval_pred.predictions, axis=-1)
references = eval_pred.label_ids
r={'f1_score': f1_score(references, predictions)}
return r
trainer = Trainer(
model.to(device),
args_promoter,
train_dataset= tokenized_datasets_train_promoter,
eval_dataset= tokenized_datasets_validation_promoter,
tokenizer=tokenizer,
compute_metrics=compute_metrics_f1_score,
)<jupyter_output><empty_output><jupyter_text>We can now finetune our model by just calling the `train` method:<jupyter_code>train_results = trainer.train()<jupyter_output>/usr/local/lib/python3.10/dist-packages/transformers/optimization.py:411: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning
warnings.warn(<jupyter_text>Note that the finetuning is done with a small batch size (8). The training time can be reduced by increasing the batch size, as it leverages parallelism in the GPU. **Validation F1 score**<jupyter_code>curve_evaluation_f1_score =[[a['step'],a['eval_f1_score']] for a in trainer.state.log_history if 'eval_f1_score' in a.keys()]
eval_f1_score = [c[1] for c in curve_evaluation_f1_score]
steps = [c[0] for c in curve_evaluation_f1_score]
plt.plot(steps, eval_f1_score, 'b', label='Validation F1 score')
plt.title('Validation F1 score for promoter prediction')
plt.xlabel('Number of training steps performed')
plt.ylabel('Validation F1 score')
plt.legend()
plt.show()<jupyter_output><empty_output><jupyter_text>**F1 score on the test dataset**<jupyter_code># Compute the F1 score on the test dataset :
print(f"F1 score on the test dataset: {trainer.predict(tokenized_datasets_test_promoter).metrics['test_f1_score']}")<jupyter_output><empty_output><jupyter_text>For the promoter prediction task, we reproduced the experiment carried out in the [**article**](https://www.biorxiv.org/content/10.1101/2023.01.11.523679v1.full.pdf) by adapting the learning rate to the LoRa method. A F1 score of **0.937** is obtained after just **1000 training steps**. To get closer to the **0.954** score obtained in the nucleotide transformer paper after 10,000 training steps, we surely need to train for longer! **Second task : Enhancer prediction** In this section, we fine-tune the nucleotide transformer model on **enhancer type prediction**, which consists in classifying a DNA sequence as **strong**, **weak** or **non enhancer**.In genetics, an enhancer is a short (50–1500 bp) region of DNA that can be bound by proteins (activators) to increase the likelihood that transcription of a particular gene will occur.[A deep learning framework for enhancer prediction using word embedding and sequence generation](https://www.sciencedirect.com/science/article/abs/pii/S0301462222000643) introduced the dataset used here by augmenting an original set of enhancers with 6000 synthetic enhancers and 6000 syntheticnon-enhancers produced through a generative model. **Dataset loading and preparation**<jupyter_code>from datasets import load_dataset, Dataset
# Load the enhancers dataset from the InstaDeep Hugging Face ressources
dataset_name = "enhancers_types"
train_dataset_enhancers = load_dataset(
"InstaDeepAI/nucleotide_transformer_downstream_tasks",
dataset_name,
split="train",
streaming= False,
)
test_dataset_enhancers = load_dataset(
"InstaDeepAI/nucleotide_transformer_downstream_tasks",
dataset_name,
split="test",
streaming= False,
)
# Get training data
train_sequences_enhancers = train_dataset_enhancers['sequence']
train_labels_enhancers = train_dataset_enhancers['label']
# Split the dataset into a training and a validation dataset
train_sequences_enhancers, validation_sequences_enhancers, train_labels_enhancers, validation_labels_enhancers = train_test_split(train_sequences_enhancers,
train_labels_enhancers, test_size=0.10, random_state=42)
# Get test data
test_sequences_enhancers = test_dataset_enhancers['sequence']
test_labels_enhancers = test_dataset_enhancers['label']<jupyter_output><empty_output><jupyter_text>**Tokenizing the datasets**<jupyter_code># Enhancer dataset
ds_train_enhancers = Dataset.from_dict({"data": train_sequences_enhancers,'labels':train_labels_enhancers})
ds_validation_enhancers = Dataset.from_dict({"data": validation_sequences_enhancers,'labels':validation_labels_enhancers})
ds_test_enhancers = Dataset.from_dict({"data": test_sequences_enhancers,'labels':test_labels_enhancers})
# Creating tokenized enhancer dataset
tokenized_datasets_train_enhancers = ds_train_enhancers.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)
tokenized_datasets_validation_enhancers = ds_validation_enhancers.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)
tokenized_datasets_test_enhancers = ds_test_enhancers.map(
tokenize_function,
batched=True,
remove_columns=["data"],
)<jupyter_output><empty_output><jupyter_text>**Fine-tuning and evaluation**<jupyter_code># Load the model
num_labels_enhancers = 3
model = AutoModelForSequenceClassification.from_pretrained("InstaDeepAI/nucleotide-transformer-500m-human-ref", num_labels=num_labels_enhancers)
model = model.to(device)
peft_config = LoraConfig(
task_type=TaskType.SEQ_CLS, inference_mode=False, r=1, lora_alpha= 32, lora_dropout=0.1, target_modules= ["query", "value"],
#modules_to_save=["intermediate"] # modules that are not frozen and updated during the training
)
lora_classifier = get_peft_model(model, peft_config) # transform our classifier into a peft model
lora_classifier.print_trainable_parameters()
lora_classifier.to(device) # Put the model on the GPU<jupyter_output>trainable params: 3409926 || all params: 482208487 || trainable%: 0.7071476533344383<jupyter_text>We initialize our `TrainingArguments`. These control the various training hyperparameters, and will be passed to our `Trainer`.We keep the same hyperparameters as for the promoter task, i.e the same as in the paper except for a learning rate of 5.10⁻⁴, which enables us to get close to [**paper's**](https://www.biorxiv.org/content/10.1101/2023.01.11.523679v1.full.pdf) performance.<jupyter_code>batch_size = 8
model_name='nucleotide-transformer'
args_enhancers = TrainingArguments(
f"{model_name}-finetuned-lora-NucleotideTransformer",
remove_unused_columns=False,
evaluation_strategy="steps",
save_strategy="steps",
learning_rate=5e-4,
per_device_train_batch_size=batch_size,
gradient_accumulation_steps= 1,
per_device_eval_batch_size= 64,
num_train_epochs= 2,
logging_steps= 100,
load_best_model_at_end=True, # Keep the best model according to the evaluation
metric_for_best_model="mcc_score", # The mcc_score on the evaluation dataset used to select the best model
label_names=["labels"],
dataloader_drop_last=True,
max_steps= 1000
)<jupyter_output><empty_output><jupyter_text>Here, the metric used to evaluate the model is the Matthews Correlation Coefficient, which is more relevant than the accuracy when the classes in the dataset are unbalanced. We can load a predefined function from the `scikit-learn` library.<jupyter_code># Define the metric for the evaluation
def compute_metrics_mcc(eval_pred):
"""Computes Matthews correlation coefficient (MCC score) for binary classification"""
predictions = np.argmax(eval_pred.predictions, axis=-1)
references = eval_pred.label_ids
r={'mcc_score': matthews_corrcoef(references, predictions)}
return r
trainer = Trainer(
lora_classifier,
args_enhancers,
train_dataset= tokenized_datasets_train_enhancers,
eval_dataset= tokenized_datasets_validation_enhancers,
tokenizer=tokenizer,
compute_metrics=compute_metrics_mcc,
)<jupyter_output><empty_output><jupyter_text>We can now finetune our model by just calling the `train` method:<jupyter_code>train_results = trainer.train()<jupyter_output>/usr/local/lib/python3.10/dist-packages/transformers/optimization.py:411: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning
warnings.warn(<jupyter_text>As with the first task, the time can be greatly reduced by increasing the batch size. **Validation MCC score**<jupyter_code>curve_evaluation_mcc_score=[[a['step'],a['eval_mcc_score']] for a in trainer.state.log_history if 'eval_mcc_score' in a.keys()]
eval_mcc_score = [c[1] for c in curve_evaluation_mcc_score]
steps = [c[0] for c in curve_evaluation_mcc_score]
plt.plot(steps, eval_mcc_score, 'b', label='Validation MCC score')
plt.title('Validation MCC score for enhancer prediction')
plt.xlabel('Number of training steps performed')
plt.ylabel('Validation MCC score')
plt.legend()
plt.show()<jupyter_output><empty_output><jupyter_text>**MCC on the test dataset**<jupyter_code># Compute the MCC score on the test dataset :
print(f"MCC score on the test dataset: {trainer.predict(tokenized_datasets_test_enhancers).metrics['test_mcc_score']}")<jupyter_output><empty_output> | notebooks/examples/nucleotide_transformer_dna_sequence_modelling_with_peft.ipynb/0 | {
"file_path": "notebooks/examples/nucleotide_transformer_dna_sequence_modelling_with_peft.ipynb",
"repo_id": "notebooks",
"token_count": 8292
} | 144 |
<jupyter_start><jupyter_text>How to fine-tune a T5 model with ONNX RuntimeThis notebook is largely inspired by the summarization [notebook of Transformers](https://github.com/huggingface/notebooks/blob/main/examples/summarization.ipynb) which takes PyTorch as backend for fine tuning.Here you will use the `ORTSeq2SeqTrainer` class in [Optimum](https://github.com/huggingface/optimum) library and take [ONNX Runtime](https://microsoft.github.io/onnxruntime/) as backend to accelerate the training. In this notebook, we will walk through the fine-tuning of [T5-small](https://huggingface.co/docs/transformers/model_doc/t5) model in the 🤗 Transformers for a summarization task. We will use the [XSum](https://arxiv.org/pdf/1808.08745.pdf) dataset (for extreme summarization) which contains BBC articles accompanied with single-sentence summaries, and the training as well as inference will be done by leveraging `ORTSeq2SeqTrainer` in Optimum! Let's speed the training up! __Dependencies__To use ONNX Runtime for training, you need a machine with at least one NVIDIA GPU.__ONNX Runtime training module need to be properly installed before launching the notebook! Please follow the instruction in [Optimum's documentation](https://huggingface.co/docs/optimum/onnxruntime/trainer) to set up your environment.__Check your GPU:<jupyter_code>!nvidia-smi<jupyter_output>Fri Sep 16 19:04:38 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01 Driver Version: 440.33.01 CUDA Version: 11.3 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla T4 On | 00000000:00:1E.0 Off | 0 |
| N/A 43C P0 25W / 70W | 3420MiB / 15109MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU [...]<jupyter_text>If you're opening this Notebook on colab, you will probably need to install 🤗 Optimum, 🤗 Transformers, 🤗 Datasets and 🤗 evaluate. Uncomment the following cell and run it.<jupyter_code>!pip install optimum transformers datasets evaluate rouge-score nltk tokenizers>=0.11.0
import nltk
nltk.download("punkt")<jupyter_output>[nltk_data] Downloading package punkt to /root/nltk_data...
[nltk_data] Unzipping tokenizers/punkt.zip.<jupyter_text>__[Optional]__ If you want to share your model with the community and generate an inference API, there are a few more steps to follow.First you have to store your authentication token from the Hugging Face website (sign up [here](https://huggingface.co/welcome) if you haven't already!) then execute the following cell and input your username and password:<jupyter_code>from huggingface_hub import notebook_login
notebook_login()<jupyter_output><empty_output><jupyter_text>Then you need to install Git-LFS. Uncomment the following instructions:<jupyter_code>!apt install git-lfs<jupyter_output><empty_output><jupyter_text>Make sure your version of Transformers is at least 4.15.0:<jupyter_code>import transformers
print(transformers.__version__)<jupyter_output>4.23.0.dev0<jupyter_text>__Setup__<jupyter_code>model_checkpoint = "t5-small"
task = "xsum"
metric_name = "rouge"
batch_size = 8
learning_rate=2e-5
weight_decay = 0.01
num_train_epochs = 1<jupyter_output><empty_output><jupyter_text>We also quickly upload some telemetry - this tells us which examples and software versions are getting used so we know where to prioritize our maintenance efforts. We don't collect (or care about) any personally identifiable information, but if you'd prefer not to be counted, feel free to skip this step or delete this cell entirely.<jupyter_code>from transformers.utils import send_example_telemetry
send_example_telemetry("summarization_notebook_ort", framework="none")<jupyter_output><empty_output><jupyter_text>Loading the datasetWe will use the [🤗 Datasets](https://github.com/huggingface/datasets) library to download the data and get the metric we need to use for evaluation (to compare our model to the benchmark). This can be easily done with the functions `load_dataset` and `load_metric`.<jupyter_code>from datasets import load_dataset, load_metric
raw_datasets = load_dataset(task)
metric = load_metric(metric_name)<jupyter_output><empty_output><jupyter_text>__[Optional]__ To get a sense of what the data looks like, the following function will show some examples picked randomly in the dataset.<jupyter_code>import datasets
import random
import pandas as pd
from IPython.display import display, HTML
def show_random_elements(dataset, num_examples=1):
assert num_examples <= len(dataset), "Can't pick more elements than there are in the dataset."
picks = []
for _ in range(num_examples):
pick = random.randint(0, len(dataset)-1)
while pick in picks:
pick = random.randint(0, len(dataset)-1)
picks.append(pick)
df = pd.DataFrame(dataset[picks])
for column, typ in dataset.features.items():
if isinstance(typ, datasets.ClassLabel):
df[column] = df[column].transform(lambda i: typ.names[i])
display(HTML(df.to_html()))
show_random_elements(raw_datasets["train"])<jupyter_output><empty_output><jupyter_text>The metric is an instance of [`datasets.Metric`](https://huggingface.co/docs/datasets/package_reference/main_classes.htmldatasets.Metric):<jupyter_code>metric
fake_preds = ["hello there", "general kenobi"]
fake_labels = ["hello there", "general kenobi"]
metric.compute(predictions=fake_preds, references=fake_labels)<jupyter_output><empty_output><jupyter_text>Preprocessing the dataBefore we can feed those texts to our model, we need to preprocess them. This is done by a 🤗 Transformers `Tokenizer` which will (as the name indicates) tokenize the inputs (including converting the tokens to their corresponding IDs in the pretrained vocabulary) and put it in a format the model expects, as well as generate the other inputs that the model requires.To do all of this, we instantiate our tokenizer with the `AutoTokenizer.from_pretrained` method, which will ensure:* we get a tokenizer that corresponds to the model architecture we want to use,* we download the vocabulary used when pretraining this specific checkpoint.That vocabulary will be cached, so it's not downloaded again the next time we run the cell.<jupyter_code>from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)<jupyter_output>/usr/local/lib/python3.8/dist-packages/transformers/models/t5/tokenization_t5_fast.py:156: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.
For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.
- Be aware that you SHOULD NOT rely on t5-small automatically truncating your input to 512 when padding/encoding.
- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.
- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.
warnings.warn(<jupyter_text>To prepare the targets for our model, we need to tokenize them inside the as_target_tokenizer context manager. This will make sure the tokenizer uses the special tokens corresponding to the targets:<jupyter_code>with tokenizer.as_target_tokenizer():
print(tokenizer(["Hello, this one sentence!", "This is another sentence."]))<jupyter_output>{'input_ids': [[8774, 6, 48, 80, 7142, 55, 1], [100, 19, 430, 7142, 5, 1]], 'attention_mask': [[1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1]]}<jupyter_text>If you are using one of the five T5 checkpoints we have to prefix the inputs with "summarize:" (the model can also translate and it needs the prefix to know which task it has to perform).<jupyter_code>if model_checkpoint in ["t5-small", "t5-base", "t5-large", "t5-3b", "t5-11b"]:
prefix = "summarize: "
else:
prefix = ""<jupyter_output><empty_output><jupyter_text>We can then write the function that will preprocess our samples. We just feed them to the `tokenizer` with the argument `truncation=True`. This will ensure that an input longer that what the model selected can handle will be truncated to the maximum length accepted by the model. The padding will be dealt with later on (in a data collator) so we pad examples to the longest length in the batch and not the whole dataset.<jupyter_code>max_input_length = 1024
max_target_length = 128
def preprocess_function(examples):
inputs = [prefix + doc for doc in examples["document"]]
model_inputs = tokenizer(inputs, max_length=max_input_length, truncation=True)
# Setup the tokenizer for targets
with tokenizer.as_target_tokenizer():
labels = tokenizer(examples["summary"], max_length=max_target_length, truncation=True)
model_inputs["labels"] = labels["input_ids"]
return model_inputs<jupyter_output><empty_output><jupyter_text>To apply this function on all the pairs of sentences in our dataset, we just use the `map` method of our `dataset` object we created earlier. This will apply the function on all the elements of all the splits in `dataset`, so our training, validation and testing data will be preprocessed in one single command.<jupyter_code>tokenized_datasets = raw_datasets.map(preprocess_function, batched=True)<jupyter_output><empty_output><jupyter_text>Even better, the results are automatically cached by the 🤗 Datasets library to avoid spending time on this step the next time you run your notebook. The 🤗 Datasets library is normally smart enough to detect when the function you pass to map has changed (and thus requires to not use the cache data). For instance, it will properly detect if you change the task in the first cell and rerun the notebook. 🤗 Datasets warns you when it uses cached files, you can pass `load_from_cache_file=False` in the call to `map` to not use the cached files and force the preprocessing to be applied again.Note that we passed `batched=True` to encode the texts by batches together. This is to leverage the full benefit of the fast tokenizer we loaded earlier, which will use multi-threading to treat the texts in a batch concurrently. Fine-tuning the modelNow that our data is ready, we can download the pretrained model and fine-tune it. Since our task is of the sequence-to-sequence kind, we use the `AutoModelForSeq2SeqLM` class to fist load the PyTorch model. Like with the tokenizer, the `from_pretrained` method will download and cache the model for us.<jupyter_code>from transformers import AutoModelForSeq2SeqLM, DataCollatorForSeq2Seq
from optimum.onnxruntime import ORTSeq2SeqTrainer, ORTSeq2SeqTrainingArguments
model = AutoModelForSeq2SeqLM.from_pretrained(model_checkpoint)<jupyter_output>huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)<jupyter_text>Note that we don't get a warning like in our classification example. This means we used all the weights of the pretrained model and there is no randomly initialized head in this case.To instantiate a `ORTSeq2SeqTrainer`, we will need to define three more things. The most important is the [`ORTSeq2SeqTrainingArguments`](https://huggingface.co/docs/optimum/onnxruntime/traineroptimum.onnxruntime.ORTSeq2SeqTrainingArguments), which is a class that contains all the attributes to customize the training. You can also use `Seq2SeqTrainingArguments` in Transformers, but `ORTSeq2SeqTrainingArguments` enables more optimized features of ONNX Runtime. It requires one folder name, which will be used to save the checkpoints of the model, and all other arguments are optional:<jupyter_code>model_name = model_checkpoint.split("/")[-1]
args = ORTSeq2SeqTrainingArguments(
f"{model_name}-finetuned-xsum",
evaluation_strategy = "epoch",
learning_rate=learning_rate,
per_device_train_batch_size=batch_size,
per_device_eval_batch_size=batch_size,
weight_decay=weight_decay,
save_total_limit=3,
num_train_epochs=num_train_epochs,
predict_with_generate=True,
optim="adamw_ort_fused",
# push_to_hub=True,
)<jupyter_output><empty_output><jupyter_text>Here we set the evaluation to be done at the end of each epoch, tweak the learning rate, use the `batch_size` defined at the top of the cell and customize the weight decay. Since the `ORTSeq2SeqTrainer` will save the model regularly and our dataset is quite large, we tell it to make three saves maximum. Lastly, we use the `predict_with_generate` option (to properly generate summaries) and activate mixed precision training (to go a bit faster).The last argument to setup everything so we can push the model to the [Hub](https://huggingface.co/models) regularly during training. Remove it if you didn't follow the installation steps at the top of the notebook. If you want to save your model locally in a name that is different than the name of the repository it will be pushed, or if you want to push your model under an organization and not your name space, use the hub_model_id argument to set the repo name (it needs to be the full name, including your namespace: for instance `"optimum/t5-large-finetuned-xsum"`).Then, we need a special kind of data collator, which will not only pad the inputs to the maximum length in the batch, but also the labels:<jupyter_code>data_collator = DataCollatorForSeq2Seq(
tokenizer,
model=model,
label_pad_token_id=tokenizer.pad_token_id,
pad_to_multiple_of=8 if args.fp16 else None,
)<jupyter_output><empty_output><jupyter_text>The last thing to define for our `ORTSeq2SeqTrainer` is how to compute the metrics from the predictions. We need to define a function for this, which will just use the `metric` we loaded earlier, and we have to do a bit of pre-processing to decode the predictions into texts:<jupyter_code>import nltk
import numpy as np
def compute_metrics(eval_pred):
predictions, labels = eval_pred
decoded_preds = tokenizer.batch_decode(predictions, skip_special_tokens=True)
# Replace -100 in the labels as we can't decode them.
labels = np.where(labels != -100, labels, tokenizer.pad_token_id)
decoded_labels = tokenizer.batch_decode(labels, skip_special_tokens=True)
# Rouge expects a newline after each sentence
decoded_preds = ["\n".join(nltk.sent_tokenize(pred.strip())) for pred in decoded_preds]
decoded_labels = ["\n".join(nltk.sent_tokenize(label.strip())) for label in decoded_labels]
result = metric.compute(predictions=decoded_preds, references=decoded_labels, use_stemmer=True)
# Extract a few results
result = {key: value.mid.fmeasure * 100 for key, value in result.items()}
# Add mean generated length
prediction_lens = [np.count_nonzero(pred != tokenizer.pad_token_id) for pred in predictions]
result["gen_len"] = np.mean(prediction_lens)
return {k: round(v, 4) for k, v in result.items()}<jupyter_output><empty_output><jupyter_text>Then we just need to pass all of this along with our datasets to the `ORTSeq2SeqTrainer`:<jupyter_code>trainer = ORTSeq2SeqTrainer(
model=model,
args=args,
train_dataset=tokenized_datasets["train"],
eval_dataset=tokenized_datasets["validation"],
tokenizer=tokenizer,
data_collator=data_collator,
compute_metrics=compute_metrics if args.predict_with_generate else None,
feature="seq2seq-lm",
)<jupyter_output><empty_output><jupyter_text>We can now finetune our model by just calling the `train` method:<jupyter_code>trainer.train()<jupyter_output>The following columns in the training set don't have a corresponding argument in `T5ForConditionalGeneration.forward` and have been ignored: summary, id, document. If summary, id, document are not expected by `T5ForConditionalGeneration.forward`, you can safely ignore this message.
You're using a T5TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.
/usr/local/lib/python3.8/dist-packages/onnxruntime/training/ortmodule/_training_manager.py:191: UserWarning: Fast path enabled - skipping checks. Rebuild graph: True, Execution agent: True, Device check: True
warnings.warn(
WARNING: The shape inference of org.pytorch.aten::ATen type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of org.pytorch.aten::ATen type is missing, so it ma[...]<jupyter_text>You can now upload the result of the training to the Hub, just execute this instruction:<jupyter_code>trainer.push_to_hub()<jupyter_output><empty_output><jupyter_text>You will also be able to save your fine-tuned model as PyTorch or ONNX model in the `output_dir` that you set in `ORTSeq2SeqTrainer`:<jupyter_code>trainer.save_model()<jupyter_output><empty_output><jupyter_text>Evaluating your model Evaluate the performance of the model that you just fine-tuned with the validation dataset that you've passed to `ORTSeq2SeqTrainer` by just calling the `evaluate` method. If you set `inference_with_ort=True`, the inference will be done with ONNX Runtime backend. Otherwise, the inference will take PyTorch as backend.<jupyter_code>trainer.evaluate(inference_with_ort=True)<jupyter_output>The following columns in the evaluation set don't have a corresponding argument in `T5ForConditionalGeneration.forward` and have been ignored: summary, id, document. If summary, id, document are not expected by `T5ForConditionalGeneration.forward`, you can safely ignore this message.
Using framework PyTorch: 1.11.0+cu113
Overriding 1 configuration item(s)
- use_cache -> False
WARNING: The shape inference of org.pytorch.aten::ATen type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of org.pytorch.aten::ATen type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of org.pytorch.aten::ATen type is missing, so it may result in wrong shape inference for the exported graph. Please consider adding it in symbolic function.
WARNING: The shape inference of org.pytorch.aten::ATen type i[...] | notebooks/examples/summarization_ort.ipynb/0 | {
"file_path": "notebooks/examples/summarization_ort.ipynb",
"repo_id": "notebooks",
"token_count": 6048
} | 145 |
<jupyter_start><jupyter_text>Getting started with Owl-ViTIn this notebook, we are going to run the [OWL-ViT](https://arxiv.org/abs/2205.06230) model (an open-vocabulary object detection model) by Google Research on scikit-image samples images. OWL-ViT: A Quick IntroOWL-ViT is an open-vocabulary object detector. Given an image and one or multiple free-text queries, it finds objects matching the queries in the image. Unlike traditional object detection models, OWL-ViT is not trained on labeled object datasets and leverages multi-modal representations to perform open-vocabulary detection. OWL-ViT uses CLIP with a ViT-like Transformer as its backbone to get multi-modal visual and text features. To use CLIP for object detection, OWL-ViT removes the final token pooling layer of the vision model and attaches a lightweight classification and box head to each transformer output token. Open-vocabulary classification is enabled by replacing the fixed classification layer weights with the class-name embeddings obtained from the text model. The authors first train CLIP from scratch and fine-tune it end-to-end with the classification and box heads on standard detection datasets using a bipartite matching loss. One or multiple text queries per image can be used to perform zero-shot text-conditioned object detection. Set-up environmentFirst, we install the HuggingFace Transformers library (from source for now, as the model was recently added to the library and is under active development). This might take a few minutes.<jupyter_code>!pip install -q git+https://github.com/huggingface/transformers.git<jupyter_output><empty_output><jupyter_text>**Optional:** Install Pillow, matplotlib and OpenCV if you are running this notebook locally.<jupyter_code>!pip install Pillow
!pip install matplotlib
!pip install opencv-python<jupyter_output><empty_output><jupyter_text>We also quickly upload some telemetry - this tells us which examples and software versions are getting used so we know where to prioritize our maintenance efforts. We don't collect (or care about) any personally identifiable information, but if you'd prefer not to be counted, feel free to skip this step or delete this cell entirely.<jupyter_code>from transformers.utils import send_example_telemetry
send_example_telemetry("zeroshot_object_detection_with_owlvit_notebook", framework="pytorch")<jupyter_output><empty_output><jupyter_text>Load pre-trained model and processorLet's first apply the image preprocessing and tokenize the text queries using `OwlViTProcessor`. The processor will resize the image(s), scale it between [0-1] range and normalize it across the channels using the mean and standard deviation specified in the original codebase.Text queries are tokenized using a CLIP tokenizer and stacked to output tensors of shape [batch_size * num_max_text_queries, sequence_length]. If you are inputting more than one set of (image, text prompt/s), num_max_text_queries is the maximum number of text queries per image across the batch. Input samples with fewer text queries are padded.<jupyter_code>from transformers import OwlViTProcessor, OwlViTForObjectDetection
model = OwlViTForObjectDetection.from_pretrained("google/owlvit-base-patch32")
processor = OwlViTProcessor.from_pretrained("google/owlvit-base-patch32")<jupyter_output><empty_output><jupyter_text>Preprocess input image and text queriesLet's use the image of astronaut Eileen Collins to test OWL-ViT. It's part of the [NASA](https://www.nasa.gov/multimedia/imagegallery/index.html) Great Images dataset.You can use one or multiple text prompts per image to search for the target object(s). Let's start with a simple example where we search for multiple objects in a single image.<jupyter_code>import cv2
import skimage
import numpy as np
from PIL import Image
# Download sample image
image = skimage.data.astronaut()
image = Image.fromarray(np.uint8(image)).convert("RGB")
# Text queries to search the image for
text_queries = ["human face", "rocket", "nasa badge", "star-spangled banner"]
image
import torch
# Use GPU if available
if torch.cuda.is_available():
device = torch.device("cuda")
else:
device = torch.device("cpu")
# Process image and text inputs
inputs = processor(text=text_queries, images=image, return_tensors="pt").to(device)
# Print input names and shapes
for key, val in inputs.items():
print(f"{key}: {val.shape}")<jupyter_output>input_ids: torch.Size([4, 16])
attention_mask: torch.Size([4, 16])
pixel_values: torch.Size([1, 3, 768, 768])<jupyter_text>Forward passNow we can pass the inputs to our OWL-ViT model to get object detection predictions. `OwlViTForObjectDetection` model outputs the prediction logits, boundary boxes and class embeddings, along with the image and text embeddings outputted by the `OwlViTModel`, which is the CLIP backbone.<jupyter_code># Set model in evaluation mode
model = model.to(device)
model.eval()
# Get predictions
with torch.no_grad():
outputs = model(**inputs)
for k, val in outputs.items():
if k not in {"text_model_output", "vision_model_output"}:
print(f"{k}: shape of {val.shape}")
print("\nText model outputs")
for k, val in outputs.text_model_output.items():
print(f"{k}: shape of {val.shape}")
print("\nVision model outputs")
for k, val in outputs.vision_model_output.items():
print(f"{k}: shape of {val.shape}")<jupyter_output>logits: shape of torch.Size([1, 576, 4])
pred_boxes: shape of torch.Size([1, 576, 4])
text_embeds: shape of torch.Size([1, 4, 512])
image_embeds: shape of torch.Size([1, 24, 24, 768])
class_embeds: shape of torch.Size([1, 576, 512])
Text model outputs
last_hidden_state: shape of torch.Size([4, 16, 512])
pooler_output: shape of torch.Size([4, 512])
Vision model outputs
last_hidden_state: shape of torch.Size([1, 577, 768])
pooler_output: shape of torch.Size([1, 768])<jupyter_text>Draw predictions on imageLet's draw the predictions / found objects on the input image. Remember the found objects correspond to the input text queries.<jupyter_code>import matplotlib.pyplot as plt
from transformers.image_utils import ImageFeatureExtractionMixin
mixin = ImageFeatureExtractionMixin()
# Load example image
image_size = model.config.vision_config.image_size
image = mixin.resize(image, image_size)
input_image = np.asarray(image).astype(np.float32) / 255.0
# Threshold to eliminate low probability predictions
score_threshold = 0.1
# Get prediction logits
logits = torch.max(outputs["logits"][0], dim=-1)
scores = torch.sigmoid(logits.values).cpu().detach().numpy()
# Get prediction labels and boundary boxes
labels = logits.indices.cpu().detach().numpy()
boxes = outputs["pred_boxes"][0].cpu().detach().numpy()
def plot_predictions(input_image, text_queries, scores, boxes, labels):
fig, ax = plt.subplots(1, 1, figsize=(8, 8))
ax.imshow(input_image, extent=(0, 1, 1, 0))
ax.set_axis_off()
for score, box, label in zip(scores, boxes, labels):
if score < score_threshold:
continue
cx, cy, w, h = box
ax.plot([cx-w/2, cx+w/2, cx+w/2, cx-w/2, cx-w/2],
[cy-h/2, cy-h/2, cy+h/2, cy+h/2, cy-h/2], "r")
ax.text(
cx - w / 2,
cy + h / 2 + 0.015,
f"{text_queries[label]}: {score:1.2f}",
ha="left",
va="top",
color="red",
bbox={
"facecolor": "white",
"edgecolor": "red",
"boxstyle": "square,pad=.3"
})
plot_predictions(input_image, text_queries, scores, boxes, labels)<jupyter_output><empty_output><jupyter_text>Batch processingWe can also pass in multiple sets of images and text queries to search for different (or same) objects in different images. Let's download an image of a coffee mug to process alongside the astronaut image.For batch processing, we need to input text queries as a nested list to `OwlViTProcessor` and images as lists of (PIL images or PyTorch tensors or NumPy arrays).<jupyter_code># Download the coffee mug image
image = skimage.data.coffee()
image = Image.fromarray(np.uint8(image)).convert("RGB")
image
# Preprocessing
images = [skimage.data.astronaut(), skimage.data.coffee()]
images = [Image.fromarray(np.uint8(img)).convert("RGB") for img in images]
# Nexted list of text queries to search each image for
text_queries = [["human face", "rocket", "nasa badge", "star-spangled banner"], ["coffee mug", "spoon", "plate"]]
# Process image and text inputs
inputs = processor(text=text_queries, images=images, return_tensors="pt").to(device)
# Print input names and shapes
for key, val in inputs.items():
print(f"{key}: {val.shape}")<jupyter_output>input_ids: torch.Size([8, 16])
attention_mask: torch.Size([8, 16])
pixel_values: torch.Size([2, 3, 768, 768])<jupyter_text>**Note:** Notice the size of the `input_ids `and `attention_mask` is `[batch_size * num_max_text_queries, max_length]`. Max_length is set to 16 for all OWL-ViT models.<jupyter_code># Get predictions
with torch.no_grad():
outputs = model(**inputs)
for k, val in outputs.items():
if k not in {"text_model_output", "vision_model_output"}:
print(f"{k}: shape of {val.shape}")
print("\nText model outputs")
for k, val in outputs.text_model_output.items():
print(f"{k}: shape of {val.shape}")
print("\nVision model outputs")
for k, val in outputs.vision_model_output.items():
print(f"{k}: shape of {val.shape}")
# Let's plot the predictions for the second image
image_idx = 1
image_size = model.config.vision_config.image_size
image = mixin.resize(images[image_idx], image_size)
input_image = np.asarray(image).astype(np.float32) / 255.0
# Threshold to eliminate low probability predictions
score_threshold = 0.1
# Get prediction logits
logits = torch.max(outputs["logits"][image_idx], dim=-1)
scores = torch.sigmoid(logits.values).cpu().detach().numpy()
# Get prediction labels and boundary boxes
labels = logits.indices.cpu().detach().numpy()
boxes = outputs["pred_boxes"][image_idx].cpu().detach().numpy()
plot_predictions(input_image, text_queries[image_idx], scores, boxes, labels)<jupyter_output><empty_output><jupyter_text>Post-processing model predictionsNotice how we printed the output predictions on the resized input image. This is because OWL-ViT outputs normalized box coordinates in `[cx, cy, w, h]` format assuming a fixed input image size. We can use the `OwlViTProcessor`'s convenient post_process() method to convert the model outputs to a COCO API format and retrieve rescaled coordinates (with respect to the original image sizes) in `[x0, y0, x1, y1]` format.<jupyter_code># Target image sizes (height, width) to rescale box predictions [batch_size, 2]
target_sizes = torch.Tensor([img.size[::-1] for img in images]).to(device)
# Convert outputs (bounding boxes and class logits) to COCO API
results = processor.post_process(outputs=outputs, target_sizes=target_sizes)
# Loop over predictions for each image in the batch
for i in range(len(images)):
print(f"\nProcessing image {i}")
text = text_queries[i]
boxes, scores, labels = results[i]["boxes"], results[i]["scores"], results[i]["labels"]
score_threshold = 0.1
for box, score, label in zip(boxes, scores, labels):
box = [round(i, 2) for i in box.tolist()]
if score >= score_threshold:
print(f"Detected {text[label]} with confidence {round(score.item(), 3)} at location {box}")<jupyter_output>Processing image 0
Detected human face with confidence 0.357 at location [180.23, 71.53, 271.25, 178.76]
Detected rocket with confidence 0.106 at location [358.81, 64.85, 424.18, 280.84]
Detected star-spangled banner with confidence 0.138 at location [1.43, 1.26, 105.38, 509.68]
Detected rocket with confidence 0.211 at location [350.98, -1.17, 468.6, 288.51]
Detected nasa badge with confidence 0.281 at location [129.58, 348.54, 206.46, 427.98]
Detected nasa badge with confidence 0.12 at location [277.15, 338.86, 327.42, 380.85]
Processing image 1
Detected coffee mug with confidence 0.175 at location [175.23, 24.72, 420.98, 246.83]
Detected spoon with confidence 0.132 at location [248.54, 63.72, 418.45, 327.98]
Detected plate with confidence 0.115 at location [78.9, 74.04, 481.06, 391.81]<jupyter_text>Bonus: one-shot / image-guided object detectionInstead of performing zero-shot detection with text inputs, we can use the `OwlViTForObjectDetection.image_guided_detection()` method to query an input image with a query / example image and detect similar looking objects. To do this, we simply pass in `query_images` instead of text to the processor to get the `query_pixel_values`. Note that, unlike text input, `OwlViTProcessor` expects one query image per target image we'd like to query for similar objects. We will also see that the output and post-processing of one-shot object detection is very similar to the zero-shot / text-guided detection.Let's try this out by querying an image with cats with another random cat image. For this part of the demo, we will perform image-guided object detection, post-process the results and display the predicted boundary boxes on the original input image using OpenCV.<jupyter_code>import cv2
import requests
from matplotlib import rcParams
# Set figure size
%matplotlib inline
rcParams['figure.figsize'] = 11 ,8
# Input image
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
target_sizes = torch.Tensor([image.size[::-1]])
# Query image
query_url = "http://images.cocodataset.org/val2017/000000058111.jpg"
query_image = Image.open(requests.get(query_url, stream=True).raw)
# Display input image and query image
fig, ax = plt.subplots(1,2)
ax[0].imshow(image)
ax[1].imshow(query_image)
# Process input and query image
inputs = processor(images=image, query_images=query_image, return_tensors="pt").to(device)
# Print input names and shapes
for key, val in inputs.items():
print(f"{key}: {val.shape}")
# Get predictions
with torch.no_grad():
outputs = model.image_guided_detection(**inputs)
for k, val in outputs.items():
if k not in {"text_model_output", "vision_model_output"}:
print(f"{k}: shape of {val.shape}")
print("\nVision model outputs")
for k, val in outputs.vision_model_output.items():
print(f"{k}: shape of {val.shape}")
img = cv2.cvtColor(np.array(image), cv2.COLOR_BGR2RGB)
outputs.logits = outputs.logits.cpu()
outputs.target_pred_boxes = outputs.target_pred_boxes.cpu()
results = processor.post_process_image_guided_detection(outputs=outputs, threshold=0.6, nms_threshold=0.3, target_sizes=target_sizes)
boxes, scores = results[0]["boxes"], results[0]["scores"]
# Draw predicted bounding boxes
for box, score in zip(boxes, scores):
box = [int(i) for i in box.tolist()]
img = cv2.rectangle(img, box[:2], box[2:], (255,0,0), 5)
if box[3] + 25 > 768:
y = box[3] - 10
else:
y = box[3] + 25
plt.imshow(img[:,:,::-1])<jupyter_output><empty_output> | notebooks/examples/zeroshot_object_detection_with_owlvit.ipynb/0 | {
"file_path": "notebooks/examples/zeroshot_object_detection_with_owlvit.ipynb",
"repo_id": "notebooks",
"token_count": 4929
} | 146 |
<jupyter_start><jupyter_text>Huggingface Sagemaker-sdk - Distributed Training Demo for `TensorFlow` Distributed Data Parallelism with `transformers` and `tensorflow` 1. [Introduction](Introduction) 2. [Development Environment and Permissions](Development-Environment-and-Permissions) 1. [Installation](Installation) 2. [Development environment](Development-environment) 3. [Permissions](Permissions)3. [Processing](Preprocessing) 1. [Tokenization](Tokenization) 2. [Uploading data to sagemaker_session_bucket](Uploading-data-to-sagemaker_session_bucket) 4. [Fine-tuning & starting Sagemaker Training Job](Fine-tuning-\&-starting-Sagemaker-Training-Job) 1. [Creating an Estimator and start a training job](Creating-an-Estimator-and-start-a-training-job) 2. [Estimator Parameters](Estimator-Parameters) 3. [Download fine-tuned model from s3](Download-fine-tuned-model-from-s3) 3. [Attach to old training job to an estimator ](Attach-to-old-training-job-to-an-estimator) 5. [_Coming soon_:Push model to the Hugging Face hub](Push-model-to-the-Hugging-Face-hub) IntroductionWelcome to our distributed end-to-end binary Text-Classification example. In this demo, we will use the Hugging Faces `transformers` and `datasets` library together with a custom Amazon sagemaker-sdk extension to fine-tune a pre-trained transformer on binary text classification. In particular, the pre-trained model will be fine-tuned using the `imdb` dataset. To speed upload Training we are going to use SageMaker distributed Data Parallel library to run our training distributed across multiple gpus. To get started, we need to set up the environment with a few prerequisite steps, for permissions, configurations, and so on. _**NOTE: You can run this demo in Sagemaker Studio, your local machine or Sagemaker Notebook Instances**_ Development Environment and Permissions Installation_*Note:* we only install the required libraries from Hugging Face and AWS. You also need PyTorch or Tensorflow, if you haven´t it installed_<jupyter_code>!pip install "sagemaker>=2.48.0" --upgrade<jupyter_output><empty_output><jupyter_text>Development environment **upgrade ipywidgets for `datasets` library and restart kernel, only needed when prerpocessing is done in the notebook**<jupyter_code>%%capture
import IPython
!conda install -c conda-forge ipywidgets -y
IPython.Application.instance().kernel.do_shutdown(True) # has to restart kernel so changes are used
import sagemaker.huggingface<jupyter_output><empty_output><jupyter_text>Permissions _If you are going to use Sagemaker in a local environment. You need access to an IAM Role with the required permissions for Sagemaker. You can find [here](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-roles.html) more about it._<jupyter_code>import sagemaker
sess = sagemaker.Session()
# sagemaker session bucket -> used for uploading data, models and logs
# sagemaker will automatically create this bucket if it not exists
sagemaker_session_bucket=None
if sagemaker_session_bucket is None and sess is not None:
# set to default bucket if a bucket name is not given
sagemaker_session_bucket = sess.default_bucket()
role = sagemaker.get_execution_role()
sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)
print(f"sagemaker role arn: {role}")
print(f"sagemaker bucket: {sess.default_bucket()}")
print(f"sagemaker session region: {sess.boto_region_name}")<jupyter_output><empty_output><jupyter_text>PreprocessingIn this example the preproccsing will be done in the `train.py` when executing the script. You could also move the `preprocessing` outside of the script and upload the data to s3 and pass it into it. Fine-tuning & starting Sagemaker Training JobIn order to create a sagemaker training job we need an `HuggingFace` Estimator. The Estimator handles end-to-end Amazon SageMaker training and deployment tasks. In a Estimator we define, which fine-tuning script should be used as `entry_point`, which `instance_type` should be used, which `hyperparameters` are passed in .....```pythonhuggingface_estimator = HuggingFace(entry_point='train.py', source_dir='./scripts', base_job_name='huggingface-sdk-extension', instance_type='ml.p3.2xlarge', instance_count=1, transformers_version='4.4', pytorch_version='1.6', py_version='py37', role=role, hyperparameters = {'epochs': 1, 'train_batch_size': 32, 'model_name':'distilbert-base-uncased' })```When we create a SageMaker training job, SageMaker takes care of starting and managing all the required ec2 instances for us with the `huggingface` container, uploads the provided fine-tuning script `train.py` and downloads the data from our `sagemaker_session_bucket` into the container at `/opt/ml/input/data`. Then, it starts the training job by running. ```python/opt/conda/bin/python train.py --epochs 1 --model_name distilbert-base-uncased --train_batch_size 32```The `hyperparameters` you define in the `HuggingFace` estimator are passed in as named arguments. Sagemaker is providing useful properties about the training environment through various environment variables, including the following:* `SM_MODEL_DIR`: A string that represents the path where the training job writes the model artifacts to. After training, artifacts in this directory are uploaded to S3 for model hosting.* `SM_NUM_GPUS`: An integer representing the number of GPUs available to the host.* `SM_CHANNEL_XXXX:` A string that represents the path to the directory that contains the input data for the specified channel. For example, if you specify two input channels in the HuggingFace estimator’s fit call, named `train` and `test`, the environment variables `SM_CHANNEL_TRAIN` and `SM_CHANNEL_TEST` are set.To run your training job locally you can define `instance_type='local'` or `instance_type='local_gpu'` for gpu usage. _Note: this does not working within SageMaker Studio_<jupyter_code>!pygmentize ./scripts/train.py<jupyter_output>[34mimport[39;49;00m [04m[36margparse[39;49;00m
[34mimport[39;49;00m [04m[36mlogging[39;49;00m
[34mimport[39;49;00m [04m[36mos[39;49;00m
[34mimport[39;49;00m [04m[36msys[39;49;00m
[34mimport[39;49;00m [04m[36mtensorflow[39;49;00m [34mas[39;49;00m [04m[36mtf[39;49;00m
[34mfrom[39;49;00m [04m[36mdatasets[39;49;00m [34mimport[39;49;00m load_dataset
[34mfrom[39;49;00m [04m[36mtqdm[39;49;00m [34mimport[39;49;00m tqdm
[34mfrom[39;49;00m [04m[36mtransformers[39;49;00m [34mimport[39;49;00m AutoTokenizer, TFAutoModelForSequenceClassification
[34mfrom[39;49;00m [04m[36mtransformers[39;49;00m[04m[36m.[39;49;00m[04m[36mfile_utils[39;49;00m [34mimport[39;49;00m is_sagemaker_distributed_available
[34mif[39;49;00m os.environ.get([33m"[39;49;00m[33mSDP_ENABLED[39;49;00m[33m"[39;49;00m) [35mor[39;49;00m is_sagemaker_distributed_available():
SDP_ENABLED = [34mTrue[39;49;00m
os.environ[[33m"[39;49;00m[33[...]<jupyter_text>Creating an Estimator and start a training job<jupyter_code>from sagemaker.huggingface import HuggingFace
# hyperparameters, which are passed into the training job
hyperparameters={
'epochs': 1,
'train_batch_size': 16,
'model_name':'distilbert-base-uncased',
}
# configuration for running training on smdistributed Data Parallel
distribution = {'smdistributed':{'dataparallel':{ 'enabled': True }}}
# instance configurations
instance_type='ml.p3dn.24xlarge'
instance_count=2
volume_size=200
huggingface_estimator = HuggingFace(
entry_point='train.py',
source_dir='./scripts',
instance_type=instance_type,
instance_count=instance_count,
role=role,
transformers_version='4.6',
tensorflow_version='2.4',
py_version='py37',
distribution=distribution,
hyperparameters=hyperparameters,
debugger_hook_config=False, # currently needed
)
huggingface_estimator.fit()<jupyter_output><empty_output><jupyter_text>Deploying the endpointTo deploy our endpoint, we call `deploy()` on our HuggingFace estimator object, passing in our desired number of instances and instance type.<jupyter_code>predictor = huggingface_estimator.deploy(1,"ml.g4dn.xlarge")<jupyter_output><empty_output><jupyter_text>Then, we use the returned predictor object to call the endpoint.<jupyter_code>sentiment_input= {"inputs":"I love using the new Inference DLC."}
predictor.predict(sentiment_input)<jupyter_output><empty_output><jupyter_text>Finally, we delete the endpoint again.<jupyter_code>predictor.delete_endpoint()<jupyter_output><empty_output><jupyter_text>Extras Estimator Parameters<jupyter_code># container image used for training job
print(f"container image used for training job: \n{huggingface_estimator.image_uri}\n")
# s3 uri where the trained model is located
print(f"s3 uri where the trained model is located: \n{huggingface_estimator.model_data}\n")
# latest training job name for this estimator
print(f"latest training job name for this estimator: \n{huggingface_estimator.latest_training_job.name}\n")
# access the logs of the training job
huggingface_estimator.sagemaker_session.logs_for_job(huggingface_estimator.latest_training_job.name)<jupyter_output><empty_output><jupyter_text>Attach to old training job to an estimator In Sagemaker you can attach an old training job to an estimator to continue training, get results etc..<jupyter_code>from sagemaker.estimator import Estimator
# job which is going to be attached to the estimator
old_training_job_name=''
# attach old training job
huggingface_estimator_loaded = Estimator.attach(old_training_job_name)
# get model output s3 from training job
huggingface_estimator_loaded.model_data<jupyter_output><empty_output> | notebooks/sagemaker/07_tensorflow_distributed_training_data_parallelism/sagemaker-notebook.ipynb/0 | {
"file_path": "notebooks/sagemaker/07_tensorflow_distributed_training_data_parallelism/sagemaker-notebook.ipynb",
"repo_id": "notebooks",
"token_count": 3614
} | 147 |
<jupyter_start><jupyter_text>Accelerate BERT Inference with Hugging Face Transformers and AWS inferentia In this end-to-end tutorial, you will learn how to speed up BERT inference for text classification with Hugging Face Transformers, Amazon SageMaker, and AWS Inferentia. You will learn how to: 1. Convert your Hugging Face Transformer to AWS Neuron (Inferentia)2. Create a custom `inference.py` script for `text-classification`3. Create and upload the neuron model and inference script to Amazon S34. Deploy a Real-time Inference Endpoint on Amazon SageMaker5. Run and evaluate Inference performance of BERT on InferentiaLet's get started! 🚀---*If you are going to use Sagemaker in a local environment (not SageMaker Studio or Notebook Instances). You need access to an IAM Role with the required permissions for Sagemaker. You can find [here](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-roles.html) more about it.* 1. Convert your Hugging Face Transformer to AWS NeuronWe are going to use the [AWS Neuron SDK for AWS Inferentia](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/index.html). The Neuron SDK includes a deep learning compiler, runtime, and tools for converting and compiling PyTorch and TensorFlow models to neuron compatible models, which can be run on [EC2 Inf1 instances](https://aws.amazon.com/ec2/instance-types/inf1/). As a first step, we need to install the [Neuron SDK](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-intro/neuron-install-guide.html) and the required packages.*Tip: If you are using Amazon SageMaker Notebook Instances or Studio you can go with the `conda_python3` conda kernel.*<jupyter_code># Set Pip repository to point to the Neuron repository
!pip config set global.extra-index-url https://pip.repos.neuron.amazonaws.com
# Install Neuron PyTorch
!pip install torch-neuron==1.9.1.* neuron-cc[tensorflow] sagemaker>=2.79.0 transformers==4.12.3 --upgrade<jupyter_output><empty_output><jupyter_text>After we have installed the Neuron SDK we can convert load and convert our model. Neuron models are converted using `torch_neuron` with its `trace` method similar to `torchscript`. You can find more information in our [documentation](https://huggingface.co/docs/transformers/serializationtorchscript).To be able to convert our model we first need to select the model we want to use for our text classification pipeline from [hf.co/models](http://hf.co/models). For this example lets go with [distilbert-base-uncased-finetuned-sst-2-english](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english) but this can be easily adjusted with other BERT-like models.<jupyter_code>model_id = "distilbert-base-uncased-finetuned-sst-2-english"<jupyter_output><empty_output><jupyter_text>At the time of writing, the [AWS Neuron SDK does not support dynamic shapes](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-guide/models/models-inferentia.htmldynamic-shapes), which means that the input size needs to be static for compiling and inference. In simpler terms, this means when the model is compiled with an input of batch size 1 and sequence length of 16. The model can only run inference on inputs with the same shape._When using a `t2.medium` instance the compiling takes around 2-3 minutes_<jupyter_code>import os
import tensorflow # to workaround a protobuf version conflict issue
import torch
import torch.neuron
from transformers import AutoTokenizer, AutoModelForSequenceClassification
# load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSequenceClassification.from_pretrained(model_id, torchscript=True)
# create dummy input for max length 128
dummy_input = "dummy input which will be padded later"
max_length = 128
embeddings = tokenizer(dummy_input, max_length=max_length, padding="max_length",return_tensors="pt")
neuron_inputs = tuple(embeddings.values())
# compile model with torch.neuron.trace and update config
model_neuron = torch.neuron.trace(model, neuron_inputs)
model.config.update({"traced_sequence_length": max_length})
# save tokenizer, neuron model and config for later use
save_dir="tmp"
os.makedirs("tmp",exist_ok=True)
model_neuron.save(os.path.join(save_dir,"neuron_model.pt"))
tokenizer.save_pretrained(save_dir)
model.config.save_pretrained(save_dir)<jupyter_output>Couldn't call 'get_role' to get Role ARN from role name philippschmid to get Role path.<jupyter_text>2. Create a custom inference.py script for text-classificationThe [Hugging Face Inference Toolkit](https://github.com/aws/sagemaker-huggingface-inference-toolkit) supports zero-code deployments on top of the [pipeline feature](https://huggingface.co/transformers/main_classes/pipelines.html) from 🤗 Transformers. This allows users to deploy Hugging Face transformers without an inference script [[Example](https://github.com/huggingface/notebooks/blob/master/sagemaker/11_deploy_model_from_hf_hub/deploy_transformer_model_from_hf_hub.ipynb)]. Currently is this feature not supported with AWS Inferentia, which means we need to provide an `inference.py` for running inference. *If you would be interested in support for zero-code deployments for inferentia let us know on the [forum](https://discuss.huggingface.co/c/sagemaker/17).*---To use the inference script, we need to create an `inference.py` script. In our example, we are going to overwrite the `model_fn` to load our neuron model and the `predict_fn` to create a text-classification pipeline. If you want to know more about the `inference.py` script check out this [example](https://github.com/huggingface/notebooks/blob/master/sagemaker/17_custom_inference_script/sagemaker-notebook.ipynb). It explains amongst other things what the `model_fn` and `predict_fn` are.<jupyter_code>!mkdir code<jupyter_output><empty_output><jupyter_text>We are using the `NEURON_RT_NUM_CORES=1` to make sure that each HTTP worker uses 1 Neuron core to maximize throughput.<jupyter_code>%%writefile code/inference.py
import os
from transformers import AutoConfig, AutoTokenizer
import torch
import torch.neuron
# To use one neuron core per worker
os.environ["NEURON_RT_NUM_CORES"] = "1"
# saved weights name
AWS_NEURON_TRACED_WEIGHTS_NAME = "neuron_model.pt"
def model_fn(model_dir):
# load tokenizer and neuron model from model_dir
tokenizer = AutoTokenizer.from_pretrained(model_dir)
model = torch.jit.load(os.path.join(model_dir, AWS_NEURON_TRACED_WEIGHTS_NAME))
model_config = AutoConfig.from_pretrained(model_dir)
return model, tokenizer, model_config
def predict_fn(data, model_tokenizer_model_config):
# destruct model, tokenizer and model config
model, tokenizer, model_config = model_tokenizer_model_config
# create embeddings for inputs
inputs = data.pop("inputs", data)
embeddings = tokenizer(
inputs,
return_tensors="pt",
max_length=model_config.traced_sequence_length,
padding="max_length",
truncation=True,
)
# convert to tuple for neuron model
neuron_inputs = tuple(embeddings.values())
# run prediciton
with torch.no_grad():
predictions = model(*neuron_inputs)[0]
scores = torch.nn.Softmax(dim=1)(predictions)
# return dictonary, which will be json serializable
return [{"label": model_config.id2label[item.argmax().item()], "score": item.max().item()} for item in scores]<jupyter_output>Overwriting code/inference.py<jupyter_text>3. Create and upload the neuron model and inference script to Amazon S3Before we can deploy our neuron model to Amazon SageMaker we need to create a `model.tar.gz` archive with all our model artifacts saved into `tmp/`, e.g. `neuron_model.pt` and upload this to Amazon S3.To do this we need to set up our permissions.<jupyter_code>import sagemaker
import boto3
sess = sagemaker.Session()
# sagemaker session bucket -> used for uploading data, models and logs
# sagemaker will automatically create this bucket if it not exists
sagemaker_session_bucket=None
if sagemaker_session_bucket is None and sess is not None:
# set to default bucket if a bucket name is not given
sagemaker_session_bucket = sess.default_bucket()
try:
role = sagemaker.get_execution_role()
except ValueError:
iam = boto3.client('iam')
role = iam.get_role(RoleName='sagemaker_execution_role')['Role']['Arn']
sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)
print(f"sagemaker role arn: {role}")
print(f"sagemaker bucket: {sess.default_bucket()}")
print(f"sagemaker session region: {sess.boto_region_name}")<jupyter_output><empty_output><jupyter_text>Next, we create our `model.tar.gz`.The `inference.py` script will be placed into a `code/` folder.<jupyter_code># copy inference.py into the code/ directory of the model directory.
!cp -r code/ tmp/code/
# create a model.tar.gz archive with all the model artifacts and the inference.py script.
%cd tmp
!tar zcvf model.tar.gz *
%cd ..<jupyter_output><empty_output><jupyter_text>Now we can upload our `model.tar.gz` to our session S3 bucket with `sagemaker`.<jupyter_code>from sagemaker.s3 import S3Uploader
# create s3 uri
s3_model_path = f"s3://{sess.default_bucket()}/{model_id}"
# upload model.tar.gz
s3_model_uri = S3Uploader.upload(local_path="tmp/model.tar.gz",desired_s3_uri=s3_model_path)
print(f"model artifcats uploaded to {s3_model_uri}")<jupyter_output><empty_output><jupyter_text>4. Deploy a Real-time Inference Endpoint on Amazon SageMakerAfter we have uploaded our `model.tar.gz` to Amazon S3 can we create a custom `HuggingfaceModel`. This class will be used to create and deploy our real-time inference endpoint on Amazon SageMaker.<jupyter_code>from sagemaker.huggingface.model import HuggingFaceModel
# create Hugging Face Model Class
huggingface_model = HuggingFaceModel(
model_data=s3_model_uri, # path to your model and script
role=role, # iam role with permissions to create an Endpoint
transformers_version="4.12", # transformers version used
pytorch_version="1.9", # pytorch version used
py_version='py37', # python version used
)
# Let SageMaker know that we've already compiled the model via neuron-cc
huggingface_model._is_compiled_model = True
# deploy the endpoint endpoint
predictor = huggingface_model.deploy(
initial_instance_count=1, # number of instances
instance_type="ml.inf1.xlarge" # AWS Inferentia Instance
)<jupyter_output><empty_output><jupyter_text>5. Run and evaluate Inference performance of BERT on InferentiaThe `.deploy()` returns an `HuggingFacePredictor` object which can be used to request inference.<jupyter_code>data = {
"inputs": "the mesmerizing performances of the leads keep the film grounded and keep the audience riveted .",
}
res = predictor.predict(data=data)
res<jupyter_output><empty_output><jupyter_text>We managed to deploy our neuron compiled BERT to AWS Inferentia on Amazon SageMaker. Now, let's test its performance of it. As a dummy load test will we loop and send 10000 synchronous requests to our endpoint.<jupyter_code># send 10000 requests
for i in range(10000):
resp = predictor.predict(
data={"inputs": "it 's a charming and often affecting journey ."}
)<jupyter_output><empty_output><jupyter_text>Let's inspect the performance in cloudwatch.<jupyter_code>print(f"https://console.aws.amazon.com/cloudwatch/home?region={sess.boto_region_name}#metricsV2:graph=~(metrics~(~(~'AWS*2fSageMaker~'ModelLatency~'EndpointName~'{predictor.endpoint_name}~'VariantName~'AllTraffic))~view~'timeSeries~stacked~false~region~'{sess.boto_region_name}~start~'-PT5M~end~'P0D~stat~'Average~period~30);query=~'*7bAWS*2fSageMaker*2cEndpointName*2cVariantName*7d*20{predictor.endpoint_name}")<jupyter_output><empty_output><jupyter_text>The average latency for our BERT model is `5-6ms` for a sequence length of 128. **Delete model and endpoint**To clean up, we can delete the model and endpoint.<jupyter_code>predictor.delete_model()
predictor.delete_endpoint()<jupyter_output><empty_output> | notebooks/sagemaker/18_inferentia_inference/sagemaker-notebook.ipynb/0 | {
"file_path": "notebooks/sagemaker/18_inferentia_inference/sagemaker-notebook.ipynb",
"repo_id": "notebooks",
"token_count": 3902
} | 148 |
<jupyter_start><jupyter_text>Document AI: Fine-tuning Donut for document-parsing using Hugging Face Transformers on Amazon SageMakerIn this tutorial, you will learn how to fine-tune and deploy [Donut-base](https://huggingface.co/naver-clova-ix/donut-base) for document-understand/document-parsing using Hugging Face Transformers and Amazon SageMaker. Donut is a new document-understanding model achieving state-of-art performance with an MIT-license, which allows it to be used for commercial purposes compared to other models like LayoutLMv2/LayoutLMv3. You will learn how to:1. [Setup Development Environment](1-setup-development-environment)2. [Load SROIE dataset](2-load-sroie-dataset)3. [Preprocess and upload dataset for Donut](3-preprocess-and-upload-dataset-for-donut)4. [Fine-tune Donut model on Amazon SageMaker](4-fine-tune-and-evaluate-donut-model)5. [Deploy Donut model on Amazon SageMaker](5-deploy-donut-model) Quick intro: Document Understanding Transformer (Donut) by ClovaAIDocument Understanding Transformer (Donut) is a new Transformer model for OCR-free document understanding. It doesn't require an OCR engine to process scanned documents but is achieving state-of-the-art performances on various visual document understanding tasks, such as visual document classification or information extraction (a.k.a. document parsing). Donut is a multimodal sequence-to-sequence model with a vision encoder ([Swin Transformer](https://huggingface.co/docs/transformers/v4.21.2/en/model_doc/swinoverview)) and text decoder ([BART](https://huggingface.co/docs/transformers/v4.21.2/en/model_doc/bart)). The encoder receives the images and computes it into an embedding, which is then passed to the decoder, which generates a sequence of tokens.* Paper: https://arxiv.org/abs/2111.15664* Official repo: https://github.com/clovaai/donut--- Now we know how Donut works, so let's get started. 🚀 1. Setup Development EnvironmentThe first step is to install the required libraries and setup the environment. We will use the [Amazon SageMaker Python SDK](https://sagemaker.readthedocs.io/en/stable/) to interact with SageMaker. We will also use [Hugging Face Transformers & Datasets](https://huggingface.co/transformers/) to preprocess the data.<jupyter_code>!pip install "transformers[sentencepiece]==4.26.0" "datasets[s3]==2.9.0" sagemaker --upgrade --quiet<jupyter_output><empty_output><jupyter_text>If you are going to use Sagemaker in a local environment. You need access to an IAM Role with the required permissions for Sagemaker. You can find [here](https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-roles.html) more about it.<jupyter_code>import sagemaker
import boto3
sess = sagemaker.Session()
# sagemaker session bucket -> used for uploading data, models and logs
# sagemaker will automatically create this bucket if it not exists
sagemaker_session_bucket=None
if sagemaker_session_bucket is None and sess is not None:
# set to default bucket if a bucket name is not given
sagemaker_session_bucket = sess.default_bucket()
try:
role = sagemaker.get_execution_role()
except ValueError:
iam = boto3.client('iam')
role = iam.get_role(RoleName='sagemaker_execution_role')['Role']['Arn']
sess = sagemaker.Session(default_bucket=sagemaker_session_bucket)
print(f"sagemaker role arn: {role}")
print(f"sagemaker bucket: {sess.default_bucket()}")
print(f"sagemaker session region: {sess.boto_region_name}")<jupyter_output><empty_output><jupyter_text>2. Load SROIE datasetWe will use the [SROIE](https://github.com/zzzDavid/ICDAR-2019-SROIE) dataset a collection of 1000 scanned receipts including their OCR, more specifically we will use the dataset from task 2 "Scanned Receipt OCR". The available dataset on Hugging Face ([darentang/sroie](https://huggingface.co/datasets/darentang/sroie)) is not compatible with Donut. Thats why we will use the original dataset together with the `imagefolder` feature of `datasets` to load our dataset. Learn more about loading image data [here](https://huggingface.co/docs/datasets/v2.4.0/en/image_loadload-image-data)._Note: The test data for task2 is sadly not available. Meaning that we end up only with 626 images._First, we will clone the repository, extract the dataset into a separate folder and remove the unnecessary files.<jupyter_code>%%bash
# clone repository
git clone https://github.com/zzzDavid/ICDAR-2019-SROIE.git
# copy data
cp -r ICDAR-2019-SROIE/data ./
# clean up
rm -rf ICDAR-2019-SROIE
rm -rf data/box<jupyter_output><empty_output><jupyter_text>Now we have two folders inside the `data/` directory. One contains the images of the receipts and the other contains the OCR text. The nex step is to create a `metadata.json` file that contains the information about the images including the OCR-text. This is necessary for the `imagefolder` feature of `datasets`.The `metadata.json` should look at the end similar to the example below.```json{"file_name": "0001.png", "text": "This is a golden retriever playing with a ball"}{"file_name": "0002.png", "text": "A german shepherd"}```In our example will `"text"` column contain the OCR text of the image, which will later be used for creating the Donut specific format.<jupyter_code>import json
from pathlib import Path
import shutil
# define paths
base_path = Path("data")
metadata_path = base_path.joinpath("key")
image_path = base_path.joinpath("img")
# define metadata list
metadata_list = []
# parse metadata
for file_name in metadata_path.glob("*.json"):
with open(file_name, "r") as json_file:
# load json file
data = json.load(json_file)
# create "text" column with json string
text = json.dumps(data)
# add to metadata list if image exists
if image_path.joinpath(f"{file_name.stem}.jpg").is_file():
metadata_list.append({"text":text,"file_name":f"{file_name.stem}.jpg"})
# delete json file
# write jsonline file
with open(image_path.joinpath('metadata.jsonl'), 'w') as outfile:
for entry in metadata_list:
json.dump(entry, outfile)
outfile.write('\n')
# remove old meta data
shutil.rmtree(metadata_path)<jupyter_output><empty_output><jupyter_text>Good Job! Now we can load the dataset using the `imagefolder` feature of `datasets`.<jupyter_code>from datasets import load_dataset
dataset = load_dataset("imagefolder", data_dir=image_path, split="train")
print(f"Dataset has {len(dataset)} images")
print(f"Dataset features are: {dataset.features.keys()}")<jupyter_output><empty_output><jupyter_text>Now, lets take a closer look at our dataset<jupyter_code>import random
random_sample = random.randint(0, len(dataset))
print(f"Random sample is {random_sample}")
print(f"OCR text is {dataset[random_sample]['text']}")
dataset[random_sample]['image'].resize((250,400))<jupyter_output>Random sample is 623
OCR text is {"company": "ONE ONE THREE SEAFOOD RESTAURANT SDN BHD", "date": "23-06-2018", "address": "NO.1, TAMAN SRI DENGKIL, JALAN AIR HITAM 43800 DENGKIL, SELANGOR.", "total": "179.50"}<jupyter_text>3. Preprocess and upload dataset for DonutAs we learned in the introduction, Donut is a sequence-to-sequence model with a vision encoder and text decoder. When fine-tuning the model we want it to generate the `"text"` based on the image we pass it. Similar to NLP tasks, we have to tokenize and preprocess the text. Before we can tokenize the text, we need to transform the JSON string into a Donut compatible document. **current JSON string**```json{"company": "ADVANCO COMPANY", "date": "17/01/2018", "address": "NO 1&3, JALAN WANGSA DELIMA 12, WANGSA LINK, WANGSA MAJU, 53300 KUALA LUMPUR", "total": "7.00"}```**Donut document**```jsonADVANCO COMPANY17/01/2018NO 1&3, JALAN WANGSA DELIMA 12, WANGSA LINK, WANGSA MAJU, 53300 KUALA LUMPUR7.00```To easily create those documents the ClovaAI team has created a [json2token](https://github.com/clovaai/donut/blob/master/donut/model.pyL497) method, which we extract and then apply.<jupyter_code>new_special_tokens = [] # new tokens which will be added to the tokenizer
task_start_token = "<s>" # start of task token
eos_token = "</s>" # eos token of tokenizer
def json2token(obj, update_special_tokens_for_json_key: bool = True, sort_json_key: bool = True):
"""
Convert an ordered JSON object into a token sequence
"""
if type(obj) == dict:
if len(obj) == 1 and "text_sequence" in obj:
return obj["text_sequence"]
else:
output = ""
if sort_json_key:
keys = sorted(obj.keys(), reverse=True)
else:
keys = obj.keys()
for k in keys:
if update_special_tokens_for_json_key:
new_special_tokens.append(fr"<s_{k}>") if fr"<s_{k}>" not in new_special_tokens else None
new_special_tokens.append(fr"</s_{k}>") if fr"</s_{k}>" not in new_special_tokens else None
output += (
fr"<s_{k}>"
+ json2token(obj[k], update_special_tokens_for_json_key, sort_json_key)
+ fr"</s_{k}>"
)
return output
elif type(obj) == list:
return r"<sep/>".join(
[json2token(item, update_special_tokens_for_json_key, sort_json_key) for item in obj]
)
else:
# excluded special tokens for now
obj = str(obj)
if f"<{obj}/>" in new_special_tokens:
obj = f"<{obj}/>" # for categorical special tokens
return obj
def preprocess_documents_for_donut(sample):
# create Donut-style input
text = json.loads(sample["text"])
d_doc = task_start_token + json2token(text) + eos_token
# convert all images to RGB
image = sample["image"].convert('RGB')
return {"image": image, "text": d_doc}
proc_dataset = dataset.map(preprocess_documents_for_donut)
print(f"Sample: {proc_dataset[45]['text']}")
print(f"New special tokens: {new_special_tokens + [task_start_token] + [eos_token]}")<jupyter_output><empty_output><jupyter_text>The next step is to tokenize our text and encode the images into tensors. Therefore we need to load `DonutProcessor`, add our new special tokens and adjust the size of the images when processing from `[1920, 2560]` to `[720, 960]` to need less memory and have faster training.<jupyter_code>from transformers import DonutProcessor
# Load processor
model_id = "naver-clova-ix/donut-base"
processor = DonutProcessor.from_pretrained(model_id)
# add new special tokens to tokenizer
processor.tokenizer.add_special_tokens({"additional_special_tokens": new_special_tokens + [task_start_token] + [eos_token]})
# we update some settings which differ from pretraining; namely the size of the images + no rotation required
# resizing the image to smaller sizes from [1920, 2560] to [960,1280]
processor.feature_extractor.size = [720,960] # should be (width, height)
processor.feature_extractor.do_align_long_axis = False<jupyter_output><empty_output><jupyter_text>Now, we can prepare our dataset, which we will use for the training later.<jupyter_code>def transform_and_tokenize(sample, processor=processor, split="train", max_length=512, ignore_id=-100):
# create tensor from image
try:
pixel_values = processor(
sample["image"], random_padding=split == "train", return_tensors="pt"
).pixel_values.squeeze()
except Exception as e:
print(sample)
print(f"Error: {e}")
return {}
# tokenize document
input_ids = processor.tokenizer(
sample["text"],
add_special_tokens=False,
max_length=max_length,
padding="max_length",
truncation=True,
return_tensors="pt",
)["input_ids"].squeeze(0)
labels = input_ids.clone()
labels[labels == processor.tokenizer.pad_token_id] = ignore_id # model doesn't need to predict pad token
return {"pixel_values": pixel_values, "labels": labels, "target_sequence": sample["text"]}
# need at least 32-64GB of RAM to run this
processed_dataset = proc_dataset.map(transform_and_tokenize,remove_columns=["image","text"])<jupyter_output><empty_output><jupyter_text>Before we can upload our dataset to S3 for training we want to split the dataset into train and test sets.<jupyter_code>processed_dataset = processed_dataset.train_test_split(test_size=0.1)
print(processed_dataset)<jupyter_output><empty_output><jupyter_text>After that is done we use the new [FileSystem integration](https://huggingface.co/docs/datasets/filesystems) to upload our dataset to S3. We are using the `sess.default_bucket()`, adjust this if you want to store the dataset in a different S3 bucket. We will use the S3 path later in our training script.<jupyter_code># save train_dataset to s3
training_input_path = f's3://{sess.default_bucket()}/processed/donut-sagemaker/train'
processed_dataset["train"].save_to_disk(training_input_path)
# save train_dataset to s3
test_input_path = f's3://{sess.default_bucket()}/processed/donut-sagemaker/test'
processed_dataset["test"].save_to_disk(test_input_path)
print("uploaded data to:")
print(f"training dataset to: {training_input_path}")
print(f"test dataset to: {test_input_path}")<jupyter_output><empty_output><jupyter_text>4. Fine-tune and evaluate Donut model on Amazon SageMakerAfter we have processed our dataset, we can start training our model using a Amazon SageMaker training job using the `HuggingFace` Estimator. The Estimator handles end-to-end Amazon SageMaker training and deployment tasks. The Estimator manages the infrastructure use. SagMaker takes care of starting and managing all the required ec2 instances for us, provides the correct huggingface container, uploads the provided scripts and downloads the data from our S3 bucket into the container at `/opt/ml/input/data`. Then, it starts the training job by running._Important steps we need to think of is that we extended the `DonutProcessor` earlier and added special tokens, which we need to pass through to our training script. We also need to pass the `image_size` and `max_length` to our training script._ As pretrained model we will use [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base). The `donut-base` includes only the pre-trained weights and was introduced in the paper [OCR-free Document Understanding Transformer](https://arxiv.org/abs/2111.15664) by Geewok et al. and first released in [this repository](https://github.com/clovaai/donut).In addition to loading our model, we are resizing the `embedding` layer to match newly added tokens and adjusting the `image_size` of our encoder to match our dataset. We are also adding tokens for inference later.<jupyter_code>import time
import json
from sagemaker.huggingface import HuggingFace
# define Training Job Name
job_name = f'huggingface-donut-{time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime())}'
# stingify special tokens
special_tokens = ",".join(processor.tokenizer.special_tokens_map_extended["additional_special_tokens"])
# hyperparameters, which are passed into the training job
hyperparameters = {
'model_id': model_id, # pre-trained model
'special_tokens': json.dumps(special_tokens), # special tokens which will be added to the tokenizer
'dataset_path': '/opt/ml/input/data/training', # path where sagemaker will save training dataset
'epochs': 3, # number of training epochs
'per_device_train_batch_size': 8, # batch size for training
'gradient_checkpointing': True, # batch size for training
'lr': 4e-5, # learning rate used during training
}
# create the Estimator
huggingface_estimator = HuggingFace(
entry_point = 'train.py', # train script
source_dir = 'scripts', # directory which includes all the files needed for training
instance_type = 'ml.g5.2xlarge', # instances type used for the training job
instance_count = 1, # the number of instances used for training
base_job_name = job_name, # the name of the training job
role = role, # Iam role used in training job to access AWS ressources, e.g. S3
volume_size = 100, # the size of the EBS volume in GB
transformers_version = '4.26', # the transformers version used in the training job
pytorch_version = '1.13', # the pytorch_version version used in the training job
py_version = 'py39', # the python version used in the training job
hyperparameters = hyperparameters
)<jupyter_output><empty_output><jupyter_text>Lets start the training job and wait until it is finished. This will take around 30 minutes.<jupyter_code># define a data input dictonary with our uploaded s3 uris
data = {'training': training_input_path}
# starting the train job with our uploaded datasets as input
huggingface_estimator.fit(data, wait=True)<jupyter_output><empty_output><jupyter_text>5. Deploy Donut model on Amazon SageMakerDuring the training we copied a `infernece.py` into out `model.tar.gz` which allows us now to easily deploy our model to SageMaker for inference. The [inference.py](./scripts/inference.py) implements a custom `model_fn` and `predict_fn` for our Donut model. The `model_fn` loads the model and processor and the `predict_fn` tokenizes the input and returns the prediction.<jupyter_code>from sagemaker.huggingface import HuggingFaceModel
# create Hugging Face Model Class
huggingface_model = HuggingFaceModel(
model_data=huggingface_estimator.model_data,
# model_data="s3://sagemaker-us-east-1-558105141721/huggingface-donut-2023-05-18-15-15-20-2023-05-18-15-15-20-285/output/model.tar.gz"
role=role,
transformers_version="4.26",
pytorch_version="1.13",
py_version="py39",
model_server_workers=1
)<jupyter_output><empty_output><jupyter_text>Before we can deploy model with the `HuggingFaceModel` class we need to create a new serializer, which supports our image data. The Serializer are used in Predictor and in the `predict` method to serializer our data to a specific `mime-type`. The default serialzier for the `HuggingFacePredcitor` is a `JSON` serializer, but since we are not going to send text data to the endpoint we will use the DataSerializer.<jupyter_code>from sagemaker.serializers import DataSerializer
# create a serializer for the data
image_serializer = DataSerializer(content_type='image/x-image') # using x-image to support multiple image formats
# deploy model to SageMaker Inference
predictor = huggingface_model.deploy(
initial_instance_count=1,
instance_type= "ml.g5.2xlarge",
serializer=image_serializer, # serializer for our image files.
)<jupyter_output><empty_output><jupyter_text>SageMaker starts the deployment process by creating a SageMaker Endpoint Configuration and a SageMaker Endpoint. The Endpoint Configuration defines the model and the instance type.Lets test by using a example from the `test` split.<jupyter_code>from PIL import Image
import io
from random import randrange
from transformers.image_transforms import to_pil_image
import numpy as np
test_sample = processed_dataset["test"][randrange(0,len(processed_dataset["test"]))]
image = to_pil_image(np.array(test_sample["pixel_values"]))
def image_to_byte_array(image: Image) -> bytes:
format = image.format if image.format else 'JPEG'
# BytesIO is a file-like buffer stored in memory
img_byte_arr = io.BytesIO()
# image.save expects a file-like as a argument
image.save(img_byte_arr, format=format)
# Turn the BytesIO object back into a bytes object
return img_byte_arr.getvalue()
res = predictor.predict(data=image_to_byte_array(image))
target = processor.token2json(test_sample["target_sequence"])
print(f"Reference:\n {target}")
print(f"Prediction:\n {res}")
image.resize((350,600))<jupyter_output>Reference:
{'total': '41.87', 'date': '24/10/2017', 'company': 'GARDENIA BAKERIES (KL) SDN BHD', 'address': 'LOT 3, JALAN PELABUR 23/1, 40300 SHAH ALAM, SELANGOR.'}
Prediction:
{'total': '41.87', 'date': '24/10/2017', 'company': 'GARDENIA BAKERIES (KL) SDN BHD', 'address': 'LOT 3, JALAN PELABUR 23/1, 40300 SHAH ALAM, SELANGOR.'}<jupyter_text>Awesome!! Our fine-tuned model parsed the document correctly and extracted the right values. The next step is to evalute our model on the test set. Since the model itself is a seq2seq is not that straightforward to evaluate. To keep things simple we will use [rogue](https://huggingface.co/spaces/evaluate-metric/rouge) short for Recall-Oriented Understudy for Gisting Evaluation. This metric does not behave like the standard accuracy: it will compare a generated text against a set of reference text. The rogue score is mostly used for summarization or machine translation tasks. The higher the score the closer the generated text is to the reference text.<jupyter_code>!pip install rouge-score py7zr evaluate
import evaluate
from tqdm import tqdm
import os
os.environ["TOKENIZERS_PARALLELISM"] = "false" # disable parallelism in tokenizers library
# Metric
rogue = evaluate.load("rouge")
predictions, references = [], []
# iterate over dataset
for sample in tqdm(processed_dataset["test"],total=len(processed_dataset["test"])):
image = to_pil_image(np.array(sample["pixel_values"]))
prediction = predictor.predict(data=image_to_byte_array(image))
reference = processor.token2json(sample["target_sequence"])
predictions.append(json.dumps(prediction))
references.append(json.dumps(reference))
# compute scores
results = rogue.compute(predictions=predictions,references=references)
print(results)
# {'rouge1': 0.8173891303548311, 'rouge2': 0.7266157117328251, 'rougeL': 0.8167726537736875, 'rougeLsum': 0.8144718562842747}<jupyter_output><empty_output><jupyter_text>Our model achieves an rogue 1 score of `81.7%` on the test set. The rogue1 refers to the overlap of unigrams (each word) between the prediction and reference._Note: The evaluation we did was very simple._In an inference test the model predicted for the `address` the value `NO. 31G&33G, JALAN SETIA INDAH X ,U13/X 40170 SETIA ALAM` and the ground truth was `'NO. 31G&33G, JALAN SETIA INDAH X,U13/X 40170 SETIA ALAM'`, where the only difference is the ` ` whitespace in between `X` and `,U13/X`.<jupyter_code>predictor.delete_model()
predictor.delete_endpoint()<jupyter_output><empty_output> | notebooks/sagemaker/26_document_ai_donut/sagemaker-notebook.ipynb/0 | {
"file_path": "notebooks/sagemaker/26_document_ai_donut/sagemaker-notebook.ipynb",
"repo_id": "notebooks",
"token_count": 7780
} | 149 |
<!--⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Fully Sharded Data Parallel
[Fully sharded data parallel](https://pytorch.org/docs/stable/fsdp.html) (FSDP) is developed for distributed training of large pretrained models up to 1T parameters. FSDP achieves this by sharding the model parameters, gradients, and optimizer states across data parallel processes and it can also offload sharded model parameters to a CPU. The memory efficiency afforded by FSDP allows you to scale training to larger batch or model sizes.
<Tip warning={true}>
Currently, FSDP does not confer any reduction in GPU memory usage and FSDP with CPU offload actually consumes 1.65x more GPU memory during training. You can track this PyTorch [issue](https://github.com/pytorch/pytorch/issues/91165) for any updates.
</Tip>
FSDP is supported in 🤗 Accelerate, and you can use it with 🤗 PEFT. This guide will help you learn how to use our FSDP [training script](https://github.com/huggingface/peft/blob/main/examples/conditional_generation/peft_lora_seq2seq_accelerate_fsdp.py). You'll configure the script to train a large model for conditional generation.
## Configuration
Begin by running the following command to [create a FSDP configuration file](https://huggingface.co/docs/accelerate/main/en/usage_guides/fsdp) with 🤗 Accelerate. Use the `--config_file` flag to save the configuration file to a specific location, otherwise it is saved as a `default_config.yaml` file in the 🤗 Accelerate cache.
The configuration file is used to set the default options when you launch the training script.
```bash
accelerate config --config_file fsdp_config.yaml
```
You'll be asked a few questions about your setup, and configure the following arguments. For this example, make sure you fully shard the model parameters, gradients, optimizer states, leverage the CPU for offloading, and wrap model layers based on the Transformer layer class name.
```bash
`Sharding Strategy`: [1] FULL_SHARD (shards optimizer states, gradients and parameters), [2] SHARD_GRAD_OP (shards optimizer states and gradients), [3] NO_SHARD
`Offload Params`: Decides Whether to offload parameters and gradients to CPU
`Auto Wrap Policy`: [1] TRANSFORMER_BASED_WRAP, [2] SIZE_BASED_WRAP, [3] NO_WRAP
`Transformer Layer Class to Wrap`: When using `TRANSFORMER_BASED_WRAP`, user specifies comma-separated string of transformer layer class names (case-sensitive) to wrap ,e.g,
`BertLayer`, `GPTJBlock`, `T5Block`, `BertLayer,BertEmbeddings,BertSelfOutput`...
`Min Num Params`: minimum number of parameters when using `SIZE_BASED_WRAP`
`Backward Prefetch`: [1] BACKWARD_PRE, [2] BACKWARD_POST, [3] NO_PREFETCH
`State Dict Type`: [1] FULL_STATE_DICT, [2] LOCAL_STATE_DICT, [3] SHARDED_STATE_DICT
```
For example, your FSDP configuration file may look like the following:
```yaml
command_file: null
commands: null
compute_environment: LOCAL_MACHINE
deepspeed_config: {}
distributed_type: FSDP
downcast_bf16: 'no'
dynamo_backend: 'NO'
fsdp_config:
fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
fsdp_backward_prefetch_policy: BACKWARD_PRE
fsdp_offload_params: true
fsdp_sharding_strategy: 1
fsdp_state_dict_type: FULL_STATE_DICT
fsdp_transformer_layer_cls_to_wrap: T5Block
gpu_ids: null
machine_rank: 0
main_process_ip: null
main_process_port: null
main_training_function: main
megatron_lm_config: {}
mixed_precision: 'no'
num_machines: 1
num_processes: 2
rdzv_backend: static
same_network: true
tpu_name: null
tpu_zone: null
use_cpu: false
```
## The important parts
Let's dig a bit deeper into the training script to understand how it works.
The [`main()`](https://github.com/huggingface/peft/blob/2822398fbe896f25d4dac5e468624dc5fd65a51b/examples/conditional_generation/peft_lora_seq2seq_accelerate_fsdp.py#L14) function begins with initializing an [`~accelerate.Accelerator`] class which handles everything for distributed training, such as automatically detecting your training environment.
<Tip>
💡 Feel free to change the model and dataset inside the `main` function. If your dataset format is different from the one in the script, you may also need to write your own preprocessing function.
</Tip>
The script also creates a configuration corresponding to the 🤗 PEFT method you're using. For LoRA, you'll use [`LoraConfig`] to specify the task type, and several other important parameters such as the dimension of the low-rank matrices, the matrices scaling factor, and the dropout probability of the LoRA layers. If you want to use a different 🤗 PEFT method, replace `LoraConfig` with the appropriate [class](../package_reference/tuners).
Next, the script wraps the base model and `peft_config` with the [`get_peft_model`] function to create a [`PeftModel`].
```diff
def main():
+ accelerator = Accelerator()
model_name_or_path = "t5-base"
base_path = "temp/data/FinancialPhraseBank-v1.0"
+ peft_config = LoraConfig(
task_type=TaskType.SEQ_2_SEQ_LM, inference_mode=False, r=8, lora_alpha=32, lora_dropout=0.1
)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name_or_path)
+ model = get_peft_model(model, peft_config)
```
Throughout the script, you'll see the [`~accelerate.Accelerator.main_process_first`] and [`~accelerate.Accelerator.wait_for_everyone`] functions which help control and synchronize when processes are executed.
After your dataset is prepared, and all the necessary training components are loaded, the script checks if you're using the `fsdp_plugin`. PyTorch offers two ways for wrapping model layers in FSDP, automatically or manually. The simplest method is to allow FSDP to automatically recursively wrap model layers without changing any other code. You can choose to wrap the model layers based on the layer name or on the size (number of parameters). In the FSDP configuration file, it uses the `TRANSFORMER_BASED_WRAP` option to wrap the [`T5Block`] layer.
```py
if getattr(accelerator.state, "fsdp_plugin", None) is not None:
accelerator.state.fsdp_plugin.auto_wrap_policy = fsdp_auto_wrap_policy(model)
```
Next, use 🤗 Accelerate's [`~accelerate.Accelerator.prepare`] function to prepare the model, datasets, optimizer, and scheduler for training.
```py
model, train_dataloader, eval_dataloader, optimizer, lr_scheduler = accelerator.prepare(
model, train_dataloader, eval_dataloader, optimizer, lr_scheduler
)
```
From here, the remainder of the script handles the training loop, evaluation, and sharing your model to the Hub.
## Train
Run the following command to launch the training script. Earlier, you saved the configuration file to `fsdp_config.yaml`, so you'll need to pass the path to the launcher with the `--config_file` argument like this:
```bash
accelerate launch --config_file fsdp_config.yaml examples/peft_lora_seq2seq_accelerate_fsdp.py
```
Once training is complete, the script returns the accuracy and compares the predictions to the labels. | peft/docs/source/accelerate/fsdp.md/0 | {
"file_path": "peft/docs/source/accelerate/fsdp.md",
"repo_id": "peft",
"token_count": 2180
} | 150 |
<!--⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Configuration
[`PeftConfigMixin`] is the base configuration class for storing the adapter configuration of a [`PeftModel`], and [`PromptLearningConfig`] is the base configuration class for soft prompt methods (p-tuning, prefix tuning, and prompt tuning). These base classes contain methods for saving and loading model configurations from the Hub, specifying the PEFT method to use, type of task to perform, and model configurations like number of layers and number of attention heads.
## PeftConfigMixin
[[autodoc]] config.PeftConfigMixin
- all
## PeftConfig
[[autodoc]] PeftConfig
- all
## PromptLearningConfig
[[autodoc]] PromptLearningConfig
- all
| peft/docs/source/package_reference/config.md/0 | {
"file_path": "peft/docs/source/package_reference/config.md",
"repo_id": "peft",
"token_count": 224
} | 151 |
<!--Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# DreamBooth fine-tuning with LoRA
This guide demonstrates how to use LoRA, a low-rank approximation technique, to fine-tune DreamBooth with the
`CompVis/stable-diffusion-v1-4` model.
Although LoRA was initially designed as a technique for reducing the number of trainable parameters in
large-language models, the technique can also be applied to diffusion models. Performing a complete model fine-tuning
of diffusion models is a time-consuming task, which is why lightweight techniques like DreamBooth or Textual Inversion
gained popularity. With the introduction of LoRA, customizing and fine-tuning a model on a specific dataset has become
even faster.
In this guide we'll be using a DreamBooth fine-tuning script that is available in
[PEFT's GitHub repo](https://github.com/huggingface/peft/tree/main/examples/lora_dreambooth). Feel free to explore it and
learn how things work.
## Set up your environment
Start by cloning the PEFT repository:
```bash
git clone https://github.com/huggingface/peft
```
Navigate to the directory containing the training scripts for fine-tuning Dreambooth with LoRA:
```bash
cd peft/examples/lora_dreambooth
```
Set up your environment: install PEFT, and all the required libraries. At the time of writing this guide we recommend
installing PEFT from source.
```bash
pip install -r requirements.txt
pip install git+https://github.com/huggingface/peft
```
## Fine-tuning DreamBooth
Prepare the images that you will use for fine-tuning the model. Set up a few environment variables:
```bash
export MODEL_NAME="CompVis/stable-diffusion-v1-4"
export INSTANCE_DIR="path-to-instance-images"
export CLASS_DIR="path-to-class-images"
export OUTPUT_DIR="path-to-save-model"
```
Here:
- `INSTANCE_DIR`: The directory containing the images that you intend to use for training your model.
- `CLASS_DIR`: The directory containing class-specific images. In this example, we use prior preservation to avoid overfitting and language-drift. For prior preservation, you need other images of the same class as part of the training process. However, these images can be generated and the training script will save them to a local path you specify here.
- `OUTPUT_DIR`: The destination folder for storing the trained model's weights.
To learn more about DreamBooth fine-tuning with prior-preserving loss, check out the [Diffusers documentation](https://huggingface.co/docs/diffusers/training/dreambooth#finetuning-with-priorpreserving-loss).
Launch the training script with `accelerate` and pass hyperparameters, as well as LoRa-specific arguments to it such as:
- `use_lora`: Enables LoRa in the training script.
- `lora_r`: The dimension used by the LoRA update matrices.
- `lora_alpha`: Scaling factor.
- `lora_text_encoder_r`: LoRA rank for text encoder.
- `lora_text_encoder_alpha`: LoRA alpha (scaling factor) for text encoder.
Here's what the full set of script arguments may look like:
```bash
accelerate launch train_dreambooth.py \
--pretrained_model_name_or_path=$MODEL_NAME \
--instance_data_dir=$INSTANCE_DIR \
--class_data_dir=$CLASS_DIR \
--output_dir=$OUTPUT_DIR \
--train_text_encoder \
--with_prior_preservation --prior_loss_weight=1.0 \
--num_dataloader_workers=1 \
--instance_prompt="a photo of sks dog" \
--class_prompt="a photo of dog" \
--resolution=512 \
--train_batch_size=1 \
--lr_scheduler="constant" \
--lr_warmup_steps=0 \
--num_class_images=200 \
--use_lora \
--lora_r 16 \
--lora_alpha 27 \
--lora_text_encoder_r 16 \
--lora_text_encoder_alpha 17 \
--learning_rate=1e-4 \
--gradient_accumulation_steps=1 \
--gradient_checkpointing \
--max_train_steps=800
```
If you are running this script on Windows, you may need to set the `--num_dataloader_workers` to 0.
## Inference with a single adapter
To run inference with the fine-tuned model, first specify the base model with which the fine-tuned LoRA weights will be combined:
```python
import os
import torch
from diffusers import StableDiffusionPipeline
from peft import PeftModel, LoraConfig
MODEL_NAME = "CompVis/stable-diffusion-v1-4"
```
Next, add a function that will create a Stable Diffusion pipeline for image generation. It will combine the weights of
the base model with the fine-tuned LoRA weights using `LoraConfig`.
```python
def get_lora_sd_pipeline(
ckpt_dir, base_model_name_or_path=None, dtype=torch.float16, device="cuda", adapter_name="default"
):
unet_sub_dir = os.path.join(ckpt_dir, "unet")
text_encoder_sub_dir = os.path.join(ckpt_dir, "text_encoder")
if os.path.exists(text_encoder_sub_dir) and base_model_name_or_path is None:
config = LoraConfig.from_pretrained(text_encoder_sub_dir)
base_model_name_or_path = config.base_model_name_or_path
if base_model_name_or_path is None:
raise ValueError("Please specify the base model name or path")
pipe = StableDiffusionPipeline.from_pretrained(base_model_name_or_path, torch_dtype=dtype).to(device)
pipe.unet = PeftModel.from_pretrained(pipe.unet, unet_sub_dir, adapter_name=adapter_name)
if os.path.exists(text_encoder_sub_dir):
pipe.text_encoder = PeftModel.from_pretrained(
pipe.text_encoder, text_encoder_sub_dir, adapter_name=adapter_name
)
if dtype in (torch.float16, torch.bfloat16):
pipe.unet.half()
pipe.text_encoder.half()
pipe.to(device)
return pipe
```
Now you can use the function above to create a Stable Diffusion pipeline using the LoRA weights that you have created during the fine-tuning step.
Note, if you're running inference on the same machine, the path you specify here will be the same as `OUTPUT_DIR`.
```python
pipe = get_lora_sd_pipeline(Path("path-to-saved-model"), adapter_name="dog")
```
Once you have the pipeline with your fine-tuned model, you can use it to generate images:
```python
prompt = "sks dog playing fetch in the park"
negative_prompt = "low quality, blurry, unfinished"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7, negative_prompt=negative_prompt).images[0]
image.save("DESTINATION_PATH_FOR_THE_IMAGE")
```
<div class="flex justify-center">
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/peft/lora_dreambooth_dog_park.png" alt="Generated image of a dog in a park"/>
</div>
## Multi-adapter inference
With PEFT you can combine multiple adapters for inference. In the previous example you have fine-tuned Stable Diffusion on
some dog images. The pipeline created based on these weights got a name - `adapter_name="dog"`. Now, suppose you also fine-tuned
this base model on images of a crochet toy. Let's see how we can use both adapters.
First, you'll need to perform all the steps as in the single adapter inference example:
1. Specify the base model.
2. Add a function that creates a Stable Diffusion pipeline for image generation uses LoRA weights.
3. Create a `pipe` with `adapter_name="dog"` based on the model fine-tuned on dog images.
Next, you're going to need a few more helper functions.
To load another adapter, create a `load_adapter()` function that leverages `load_adapter()` method of `PeftModel` (e.g. `pipe.unet.load_adapter(peft_model_path, adapter_name)`):
```python
def load_adapter(pipe, ckpt_dir, adapter_name):
unet_sub_dir = os.path.join(ckpt_dir, "unet")
text_encoder_sub_dir = os.path.join(ckpt_dir, "text_encoder")
pipe.unet.load_adapter(unet_sub_dir, adapter_name=adapter_name)
if os.path.exists(text_encoder_sub_dir):
pipe.text_encoder.load_adapter(text_encoder_sub_dir, adapter_name=adapter_name)
```
To switch between adapters, write a function that uses `set_adapter()` method of `PeftModel` (see `pipe.unet.set_adapter(adapter_name)`)
```python
def set_adapter(pipe, adapter_name):
pipe.unet.set_adapter(adapter_name)
if isinstance(pipe.text_encoder, PeftModel):
pipe.text_encoder.set_adapter(adapter_name)
```
Finally, add a function to create weighted LoRA adapter.
```python
def create_weighted_lora_adapter(pipe, adapters, weights, adapter_name="default"):
pipe.unet.add_weighted_adapter(adapters, weights, adapter_name)
if isinstance(pipe.text_encoder, PeftModel):
pipe.text_encoder.add_weighted_adapter(adapters, weights, adapter_name)
return pipe
```
Let's load the second adapter from the model fine-tuned on images of a crochet toy, and give it a unique name:
```python
load_adapter(pipe, Path("path-to-the-second-saved-model"), adapter_name="crochet")
```
Create a pipeline using weighted adapters:
```python
pipe = create_weighted_lora_adapter(pipe, ["crochet", "dog"], [1.0, 1.05], adapter_name="crochet_dog")
```
Now you can switch between adapters. If you'd like to generate more dog images, set the adapter to `"dog"`:
```python
set_adapter(pipe, adapter_name="dog")
prompt = "sks dog in a supermarket isle"
negative_prompt = "low quality, blurry, unfinished"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7, negative_prompt=negative_prompt).images[0]
image
```
<div class="flex justify-center">
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/peft/lora_dreambooth_dog_supermarket.png" alt="Generated image of a dog in a supermarket"/>
</div>
In the same way, you can switch to the second adapter:
```python
set_adapter(pipe, adapter_name="crochet")
prompt = "a fish rendered in the style of <1>"
negative_prompt = "low quality, blurry, unfinished"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7, negative_prompt=negative_prompt).images[0]
image
```
<div class="flex justify-center">
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/peft/lora_dreambooth_fish.png" alt="Generated image of a crochet fish"/>
</div>
Finally, you can use combined weighted adapters:
```python
set_adapter(pipe, adapter_name="crochet_dog")
prompt = "sks dog rendered in the style of <1>, close up portrait, 4K HD"
negative_prompt = "low quality, blurry, unfinished"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7, negative_prompt=negative_prompt).images[0]
image
```
<div class="flex justify-center">
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/peft/lora_dreambooth_crochet_dog.png" alt="Generated image of a crochet dog"/>
</div>
| peft/docs/source/task_guides/dreambooth_lora.md/0 | {
"file_path": "peft/docs/source/task_guides/dreambooth_lora.md",
"repo_id": "peft",
"token_count": 3693
} | 152 |
compute_environment: LOCAL_MACHINE
deepspeed_config:
gradient_accumulation_steps: 1
gradient_clipping: 1.0
offload_optimizer_device: none
offload_param_device: none
zero3_init_flag: true
zero3_save_16bit_model: true
zero_stage: 3
distributed_type: DEEPSPEED
downcast_bf16: 'no'
dynamo_backend: 'NO'
fsdp_config: {}
machine_rank: 0
main_training_function: main
megatron_lm_config: {}
mixed_precision: 'no'
num_machines: 1
num_processes: 1
rdzv_backend: static
same_network: true
use_cpu: false | peft/examples/conditional_generation/accelerate_ds_zero3_cpu_offload_config.yaml/0 | {
"file_path": "peft/examples/conditional_generation/accelerate_ds_zero3_cpu_offload_config.yaml",
"repo_id": "peft",
"token_count": 198
} | 153 |
# Fine-tuning for image classification using LoRA and 🤗 PEFT
## Vision Transformer model from transformers
[](https://colab.research.google.com/github/huggingface/peft/blob/main/examples/image_classification/image_classification_peft_lora.ipynb)
We provide a notebook (`image_classification_peft_lora.ipynb`) where we learn how to use [LoRA](https://arxiv.org/abs/2106.09685) from 🤗 PEFT to fine-tune an image classification model by ONLY using **0.7%** of the original trainable parameters of the model.
LoRA adds low-rank "update matrices" to certain blocks in the underlying model (in this case the attention blocks) and ONLY trains those matrices during fine-tuning. During inference, these update matrices are _merged_ with the original model parameters. For more details, check out the [original LoRA paper](https://arxiv.org/abs/2106.09685).
## PoolFormer model from timm
[](https://colab.research.google.com/github/huggingface/peft/blob/main/examples/image_classification/image_classification_timm_peft_lora.ipynb)
The notebook `image_classification_timm_peft_lora.ipynb` showcases fine-tuning an image classification model using from the [timm](https://huggingface.co/docs/timm/index) library. Again, LoRA is used to reduce the numberof trainable parameters to a fraction of the total.
| peft/examples/image_classification/README.md/0 | {
"file_path": "peft/examples/image_classification/README.md",
"repo_id": "peft",
"token_count": 457
} | 154 |
# coding=utf-8
# Copyright 2023-present the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from __future__ import annotations
from typing import TYPE_CHECKING, Any, Dict
import torch
from .config import PeftConfig
from .mixed_model import PeftMixedModel
from .peft_model import (
PeftModel,
PeftModelForCausalLM,
PeftModelForFeatureExtraction,
PeftModelForQuestionAnswering,
PeftModelForSeq2SeqLM,
PeftModelForSequenceClassification,
PeftModelForTokenClassification,
)
from .tuners import (
AdaLoraConfig,
AdaLoraModel,
AdaptionPromptConfig,
IA3Config,
IA3Model,
LoHaConfig,
LoHaModel,
LoKrConfig,
LoKrModel,
LoraConfig,
LoraModel,
MultitaskPromptTuningConfig,
OFTConfig,
OFTModel,
PolyConfig,
PolyModel,
PrefixTuningConfig,
PromptEncoderConfig,
PromptTuningConfig,
)
from .utils import _prepare_prompt_learning_config
if TYPE_CHECKING:
from transformers import PreTrainedModel
MODEL_TYPE_TO_PEFT_MODEL_MAPPING: Dict[str, PeftModel] = {
"SEQ_CLS": PeftModelForSequenceClassification,
"SEQ_2_SEQ_LM": PeftModelForSeq2SeqLM,
"CAUSAL_LM": PeftModelForCausalLM,
"TOKEN_CLS": PeftModelForTokenClassification,
"QUESTION_ANS": PeftModelForQuestionAnswering,
"FEATURE_EXTRACTION": PeftModelForFeatureExtraction,
}
PEFT_TYPE_TO_CONFIG_MAPPING: Dict[str, PeftConfig] = {
"ADAPTION_PROMPT": AdaptionPromptConfig,
"PROMPT_TUNING": PromptTuningConfig,
"PREFIX_TUNING": PrefixTuningConfig,
"P_TUNING": PromptEncoderConfig,
"LORA": LoraConfig,
"LOHA": LoHaConfig,
"LOKR": LoKrConfig,
"ADALORA": AdaLoraConfig,
"IA3": IA3Config,
"MULTITASK_PROMPT_TUNING": MultitaskPromptTuningConfig,
"OFT": OFTConfig,
"POLY": PolyConfig,
}
PEFT_TYPE_TO_TUNER_MAPPING = {
"LORA": LoraModel,
"LOHA": LoHaModel,
"LOKR": LoKrModel,
"ADALORA": AdaLoraModel,
"IA3": IA3Model,
"OFT": OFTModel,
"POLY": PolyModel,
}
def get_peft_config(config_dict: Dict[str, Any]) -> PeftConfig:
"""
Returns a Peft config object from a dictionary.
Args:
config_dict (`Dict[str, Any]`): Dictionary containing the configuration parameters.
"""
return PEFT_TYPE_TO_CONFIG_MAPPING[config_dict["peft_type"]](**config_dict)
def get_peft_model(
model: PreTrainedModel, peft_config: PeftConfig, adapter_name: str = "default", mixed: bool = False
) -> PeftModel | PeftMixedModel:
"""
Returns a Peft model object from a model and a config.
Args:
model ([`transformers.PreTrainedModel`]):
Model to be wrapped.
peft_config ([`PeftConfig`]):
Configuration object containing the parameters of the Peft model.
adapter_name (`str`, `optional`, defaults to `"default"`):
The name of the adapter to be injected, if not provided, the default adapter name is used ("default").
mixed (`bool`, `optional`, defaults to `False`):
Whether to allow mixing different (compatible) adapter types.
"""
model_config = getattr(model, "config", {"model_type": "custom"})
if hasattr(model_config, "to_dict"):
model_config = model_config.to_dict()
peft_config.base_model_name_or_path = model.__dict__.get("name_or_path", None)
if mixed:
return PeftMixedModel(model, peft_config, adapter_name=adapter_name)
if peft_config.task_type not in MODEL_TYPE_TO_PEFT_MODEL_MAPPING.keys() and not peft_config.is_prompt_learning:
return PeftModel(model, peft_config, adapter_name=adapter_name)
if peft_config.is_prompt_learning:
peft_config = _prepare_prompt_learning_config(peft_config, model_config)
return MODEL_TYPE_TO_PEFT_MODEL_MAPPING[peft_config.task_type](model, peft_config, adapter_name=adapter_name)
def inject_adapter_in_model(
peft_config: PeftConfig, model: torch.nn.Module, adapter_name: str = "default"
) -> torch.nn.Module:
r"""
A simple API to create and inject adapter in-place into a model. Currently the API does not support prompt learning
methods and adaption prompt. Make sure to have the correct `target_names` set in the `peft_config` object. The API
calls `get_peft_model` under the hood but would be restricted only to non-prompt learning methods.
Args:
peft_config (`PeftConfig`):
Configuration object containing the parameters of the Peft model.
model (`torch.nn.Module`):
The input model where the adapter will be injected.
adapter_name (`str`, `optional`, defaults to `"default"`):
The name of the adapter to be injected, if not provided, the default adapter name is used ("default").
"""
if peft_config.is_prompt_learning or peft_config.is_adaption_prompt:
raise ValueError("`create_and_replace` does not support prompt learning and adaption prompt yet.")
if peft_config.peft_type not in PEFT_TYPE_TO_TUNER_MAPPING.keys():
raise ValueError(
f"`inject_adapter_in_model` does not support {peft_config.peft_type} yet. Please use `get_peft_model`."
)
tuner_cls = PEFT_TYPE_TO_TUNER_MAPPING[peft_config.peft_type]
# By instantiating a peft model we are injecting randomly initialized LoRA layers into the model's modules.
peft_model = tuner_cls(model, peft_config, adapter_name=adapter_name)
return peft_model.model
| peft/src/peft/mapping.py/0 | {
"file_path": "peft/src/peft/mapping.py",
"repo_id": "peft",
"token_count": 2278
} | 155 |
# coding=utf-8
# Copyright 2023-present the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import Any, Optional
import torch
from peft.tuners.lora.layer import LoraLayer
from peft.tuners.tuners_utils import BaseTunerLayer
from peft.utils import get_auto_gptq_quant_linear
class QuantLinear(torch.nn.Module, LoraLayer):
def __init__(
self,
base_layer,
adapter_name: str,
r: int = 0,
lora_alpha: int = 1,
lora_dropout: float = 0.0,
init_lora_weights: bool = True,
use_rslora: bool = False,
**kwargs,
):
super().__init__()
LoraLayer.__init__(self, base_layer)
# self.base_layer and self.quant_linear_module are the same; we need the former for consistency and the latter
# for backwards compatibility
self.quant_linear_module = base_layer
self._active_adapter = adapter_name
self.update_layer(adapter_name, r, lora_alpha, lora_dropout, init_lora_weights, use_rslora)
def forward(self, x: torch.Tensor):
# note: logic differs from default Linear because merging is not supported
result = self.quant_linear_module(x)
if self.disable_adapters:
return result
for active_adapter in self.active_adapters:
if active_adapter not in self.lora_A.keys():
continue
lora_A = self.lora_A[active_adapter]
lora_B = self.lora_B[active_adapter]
dropout = self.lora_dropout[active_adapter]
scaling = self.scaling[active_adapter]
requires_conversion = not torch.is_autocast_enabled()
if requires_conversion:
expected_dtype = result.dtype
x = x.to(lora_A.weight.dtype)
output = lora_B(lora_A(dropout(x)))
if requires_conversion:
output = output.to(expected_dtype)
output = output * scaling
result += output
return result
def __repr__(self) -> str:
rep = super().__repr__()
return "lora." + rep
# TODO: Check if it is better as suggested by users https://github.com/PanQiWei/AutoGPTQ/pull/102
# def reset_lora_parameters(self, adapter_name):
# if adapter_name in self.lora_A.keys():
# torch.nn.init.xavier_uniform_(self.lora_A[adapter_name].weight)
# torch.nn.init.zeros_(self.lora_B[adapter_name].weight)
def dispatch_gptq(
target: torch.nn.Module,
adapter_name: str,
**kwargs: Any,
) -> Optional[torch.nn.Module]:
new_module = None
if isinstance(target, BaseTunerLayer):
target_base_layer = target.get_base_layer()
else:
target_base_layer = target
gptq_quantization_config = kwargs.get("gptq_quantization_config", None)
AutoGPTQQuantLinear = get_auto_gptq_quant_linear(gptq_quantization_config)
if AutoGPTQQuantLinear is not None and isinstance(target_base_layer, AutoGPTQQuantLinear):
new_module = QuantLinear(target, adapter_name, **kwargs)
target.qweight = target_base_layer.qweight
return new_module
| peft/src/peft/tuners/lora/gptq.py/0 | {
"file_path": "peft/src/peft/tuners/lora/gptq.py",
"repo_id": "peft",
"token_count": 1516
} | 156 |
# coding=utf-8
# Copyright 2023-present the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Based on https://github.com/NVIDIA/NeMo/blob/main/nemo/collections/nlp/modules/common/prompt_encoder.py
# with some refactor
import warnings
import torch
from .config import PromptEncoderConfig, PromptEncoderReparameterizationType
class PromptEncoder(torch.nn.Module):
"""
The prompt encoder network that is used to generate the virtual token embeddings for p-tuning.
Args:
config ([`PromptEncoderConfig`]): The configuration of the prompt encoder.
Example:
```py
>>> from peft import PromptEncoder, PromptEncoderConfig
>>> config = PromptEncoderConfig(
... peft_type="P_TUNING",
... task_type="SEQ_2_SEQ_LM",
... num_virtual_tokens=20,
... token_dim=768,
... num_transformer_submodules=1,
... num_attention_heads=12,
... num_layers=12,
... encoder_reparameterization_type="MLP",
... encoder_hidden_size=768,
... )
>>> prompt_encoder = PromptEncoder(config)
```
**Attributes**:
- **embedding** (`torch.nn.Embedding`) -- The embedding layer of the prompt encoder.
- **mlp_head** (`torch.nn.Sequential`) -- The MLP head of the prompt encoder if `inference_mode=False`.
- **lstm_head** (`torch.nn.LSTM`) -- The LSTM head of the prompt encoder if `inference_mode=False` and
`encoder_reparameterization_type="LSTM"`.
- **token_dim** (`int`) -- The hidden embedding dimension of the base transformer model.
- **input_size** (`int`) -- The input size of the prompt encoder.
- **output_size** (`int`) -- The output size of the prompt encoder.
- **hidden_size** (`int`) -- The hidden size of the prompt encoder.
- **total_virtual_tokens** (`int`): The total number of virtual tokens of the
prompt encoder.
- **encoder_type** (Union[[`PromptEncoderReparameterizationType`], `str`]): The encoder type of the prompt
encoder.
Input shape: (`batch_size`, `total_virtual_tokens`)
Output shape: (`batch_size`, `total_virtual_tokens`, `token_dim`)
"""
def __init__(self, config):
super().__init__()
self.token_dim = config.token_dim
self.input_size = self.token_dim
self.output_size = self.token_dim
self.hidden_size = config.encoder_hidden_size
self.total_virtual_tokens = config.num_virtual_tokens * config.num_transformer_submodules
self.encoder_type = config.encoder_reparameterization_type
# embedding
self.embedding = torch.nn.Embedding(self.total_virtual_tokens, self.token_dim)
if not config.inference_mode:
if self.encoder_type == PromptEncoderReparameterizationType.LSTM:
lstm_dropout = config.encoder_dropout
num_layers = config.encoder_num_layers
# LSTM
self.lstm_head = torch.nn.LSTM(
input_size=self.input_size,
hidden_size=self.hidden_size,
num_layers=num_layers,
dropout=lstm_dropout,
bidirectional=True,
batch_first=True,
)
self.mlp_head = torch.nn.Sequential(
torch.nn.Linear(self.hidden_size * 2, self.hidden_size * 2),
torch.nn.ReLU(),
torch.nn.Linear(self.hidden_size * 2, self.output_size),
)
elif self.encoder_type == PromptEncoderReparameterizationType.MLP:
encoder_num_layers_default = PromptEncoderConfig.encoder_num_layers
if config.encoder_num_layers != encoder_num_layers_default:
warnings.warn(
f"for {self.encoder_type.value}, the argument `encoder_num_layers` is ignored. "
f"Exactly {encoder_num_layers_default} MLP layers are used."
)
layers = [
torch.nn.Linear(self.input_size, self.hidden_size),
torch.nn.ReLU(),
torch.nn.Linear(self.hidden_size, self.hidden_size),
torch.nn.ReLU(),
torch.nn.Linear(self.hidden_size, self.output_size),
]
self.mlp_head = torch.nn.Sequential(*layers)
else:
raise ValueError("Prompt encoder type not recognized. Please use one of MLP (recommended) or LSTM.")
def forward(self, indices):
input_embeds = self.embedding(indices)
if self.encoder_type == PromptEncoderReparameterizationType.LSTM:
output_embeds = self.mlp_head(self.lstm_head(input_embeds)[0])
elif self.encoder_type == PromptEncoderReparameterizationType.MLP:
output_embeds = self.mlp_head(input_embeds)
else:
raise ValueError("Prompt encoder type not recognized. Please use one of MLP (recommended) or LSTM.")
return output_embeds
| peft/src/peft/tuners/p_tuning/model.py/0 | {
"file_path": "peft/src/peft/tuners/p_tuning/model.py",
"repo_id": "peft",
"token_count": 2483
} | 157 |
# coding=utf-8
# Copyright 2023-present the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import copy
import inspect
import warnings
from typing import Optional, Tuple
import accelerate
import torch
from accelerate.hooks import add_hook_to_module, remove_hook_from_module
from accelerate.utils import is_npu_available, is_xpu_available
from safetensors.torch import storage_ptr, storage_size
from ..import_utils import is_auto_gptq_available, is_torch_tpu_available
from .constants import (
CONFIG_NAME,
EMBEDDING_LAYER_NAMES,
INCLUDE_LINEAR_LAYERS_SHORTHAND,
SAFETENSORS_WEIGHTS_NAME,
TRANSFORMERS_MODELS_TO_ADALORA_TARGET_MODULES_MAPPING,
TRANSFORMERS_MODELS_TO_IA3_FEEDFORWARD_MODULES_MAPPING,
TRANSFORMERS_MODELS_TO_IA3_TARGET_MODULES_MAPPING,
TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING,
TRANSFORMERS_MODELS_TO_PREFIX_TUNING_POSTPROCESS_MAPPING,
WEIGHTS_NAME,
bloom_model_postprocess_past_key_value,
starcoder_model_postprocess_past_key_value,
)
__all__ = [
"CONFIG_NAME",
"EMBEDDING_LAYER_NAMES",
"SAFETENSORS_WEIGHTS_NAME",
"TRANSFORMERS_MODELS_TO_ADALORA_TARGET_MODULES_MAPPING",
"TRANSFORMERS_MODELS_TO_IA3_FEEDFORWARD_MODULES_MAPPING",
"TRANSFORMERS_MODELS_TO_IA3_TARGET_MODULES_MAPPING",
"TRANSFORMERS_MODELS_TO_LORA_TARGET_MODULES_MAPPING",
"TRANSFORMERS_MODELS_TO_PREFIX_TUNING_POSTPROCESS_MAPPING",
"WEIGHTS_NAME",
"INCLUDE_LINEAR_LAYERS_SHORTHAND",
"bloom_model_postprocess_past_key_value",
"starcoder_model_postprocess_past_key_value",
]
# Get current device name based on available devices
def infer_device():
if torch.cuda.is_available():
torch_device = "cuda"
elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available():
torch_device = torch.device("mps")
elif is_xpu_available():
torch_device = "xpu"
elif is_npu_available():
torch_device = "npu"
else:
torch_device = "cpu"
return torch_device
def prepare_model_for_kbit_training(model, use_gradient_checkpointing=True, gradient_checkpointing_kwargs=None):
r"""
Note this method only works for `transformers` models.
This method wraps the entire protocol for preparing a model before running a training. This includes:
1- Cast the layernorm in fp32 2- making output embedding layer require grads 3- Add the upcasting of the lm
head to fp32
Args:
model (`transformers.PreTrainedModel`):
The loaded model from `transformers`
use_gradient_checkpointing (`bool`, *optional*, defaults to `True`):
If True, use gradient checkpointing to save memory at the expense of slower backward pass.
gradient_checkpointing_kwargs (`dict`, *optional*, defaults to `None`):
Keyword arguments to pass to the gradient checkpointing function, please refer to the documentation of
`torch.utils.checkpoint.checkpoint` for more details about the arguments that you can pass to that method.
Note this is only available in the latest transformers versions (> 4.34.1).
"""
loaded_in_kbit = getattr(model, "is_loaded_in_8bit", False) or getattr(model, "is_loaded_in_4bit", False)
is_gptq_quantized = getattr(model, "quantization_method", None) == "gptq"
if gradient_checkpointing_kwargs is None:
gradient_checkpointing_kwargs = {}
for name, param in model.named_parameters():
# freeze base model's layers
param.requires_grad = False
if not is_gptq_quantized:
# cast all non INT8 parameters to fp32
for param in model.parameters():
if (param.dtype == torch.float16) or (param.dtype == torch.bfloat16):
param.data = param.data.to(torch.float32)
if (loaded_in_kbit or is_gptq_quantized) and use_gradient_checkpointing:
# When having `use_reentrant=False` + gradient_checkpointing, there is no need for this hack
if "use_reentrant" not in gradient_checkpointing_kwargs or gradient_checkpointing_kwargs["use_reentrant"]:
# For backward compatibility
if hasattr(model, "enable_input_require_grads"):
model.enable_input_require_grads()
else:
def make_inputs_require_grad(module, input, output):
output.requires_grad_(True)
model.get_input_embeddings().register_forward_hook(make_inputs_require_grad)
# To support older transformers versions, check if the model supports gradient_checkpointing_kwargs
_supports_gc_kwargs = "gradient_checkpointing_kwargs" in list(
inspect.signature(model.gradient_checkpointing_enable).parameters
)
if not _supports_gc_kwargs and len(gradient_checkpointing_kwargs) > 0:
warnings.warn(
"gradient_checkpointing_kwargs is not supported in this version of transformers. The passed kwargs will be ignored."
" if you want to use that feature, please upgrade to the latest version of transformers.",
FutureWarning,
)
gc_enable_kwargs = (
{} if not _supports_gc_kwargs else {"gradient_checkpointing_kwargs": gradient_checkpointing_kwargs}
)
# enable gradient checkpointing for memory efficiency
model.gradient_checkpointing_enable(**gc_enable_kwargs)
return model
# For backward compatibility
def prepare_model_for_int8_training(*args, **kwargs):
warnings.warn(
"prepare_model_for_int8_training is deprecated and will be removed in a future version. Use prepare_model_for_kbit_training instead.",
FutureWarning,
)
return prepare_model_for_kbit_training(*args, **kwargs)
# copied from transformers.models.bart.modeling_bart
def shift_tokens_right(input_ids: torch.Tensor, pad_token_id: int, decoder_start_token_id: int):
"""
Shift input ids one token to the right.
Args:
input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): input ids
pad_token_id (`int`): The id of the `padding` token.
decoder_start_token_id (`int`): The id of the `start` token.
"""
shifted_input_ids = input_ids.new_zeros(input_ids.shape)
shifted_input_ids[:, 1:] = input_ids[:, :-1].clone()
shifted_input_ids[:, 0] = decoder_start_token_id
if pad_token_id is None:
raise ValueError("self.model.config.pad_token_id has to be defined.")
# replace possible -100 values in labels by `pad_token_id`
shifted_input_ids.masked_fill_(shifted_input_ids == -100, pad_token_id)
return shifted_input_ids
class ModulesToSaveWrapper(torch.nn.Module):
def __init__(self, module_to_save, adapter_name):
super().__init__()
self.original_module = module_to_save
self.modules_to_save = torch.nn.ModuleDict({})
self._active_adapter = adapter_name
self._disable_adapters = False
self.update(adapter_name)
@property
def disable_adapters(self) -> bool:
# use a property to ensure that disable_adapters is not set directly, instead use the enable_adapters method
return self._disable_adapters
@property
def active_adapter(self) -> str:
# use a property to ensure that active_adapter is not set directly, instead use the set_adapter method
return self._active_adapter
@property
def weight(self):
if self.active_adapter not in self.modules_to_save:
return self.original_module.weight
return self.modules_to_save[self.active_adapter].weight
def update(self, adapter_name):
self.modules_to_save.update(torch.nn.ModuleDict({adapter_name: copy.deepcopy(self.original_module)}))
if hasattr(self.modules_to_save[adapter_name], "_hf_hook"):
old_hook = self.modules_to_save[adapter_name]._hf_hook
new_hook = self._create_new_hook(old_hook)
remove_hook_from_module(self.modules_to_save[adapter_name])
add_hook_to_module(self.modules_to_save[adapter_name], new_hook)
self.original_module.requires_grad_(False)
if adapter_name == self.active_adapter:
self.modules_to_save[adapter_name].requires_grad_(True)
def _create_new_hook(self, old_hook):
r"""
Creates a new hook based on the old hook. Use it only if you know what you are doing !
"""
old_hook_cls = getattr(accelerate.hooks, old_hook.__class__.__name__)
old_hook_attr = old_hook.__dict__
filtered_old_hook_attr = {}
old_hook_init_signature = inspect.signature(old_hook_cls.__init__)
for k in old_hook_attr.keys():
if k in old_hook_init_signature.parameters:
filtered_old_hook_attr[k] = old_hook_attr[k]
new_hook = old_hook_cls(**filtered_old_hook_attr)
return new_hook
def forward(self, *args, **kwargs):
if self.disable_adapters or (self.active_adapter not in self.modules_to_save):
return self.original_module(*args, **kwargs)
return self.modules_to_save[self.active_adapter](*args, **kwargs)
def enable_adapters(self, enabled: bool):
"""Toggle the enabling and disabling of adapters
Takes care of setting the requires_grad flag for the adapter weights.
Args:
enabled (bool): True to enable adapters, False to disable adapters
"""
if self._disable_adapters is not enabled:
# already in the desired state, do nothing
return
if enabled:
self.original_module.requires_grad_(False)
self.modules_to_save[self.active_adapter].requires_grad_(True)
self._disable_adapters = False
else:
self.original_module.requires_grad_(True)
self.modules_to_save.requires_grad_(False)
self._disable_adapters = True
def set_adapter(self, adapter_name: str):
"""Set the active adapter
Args:
adapter_name (str): The name of the adapter to set as active
"""
if adapter_name not in self.modules_to_save:
raise ValueError(f"Adapter {adapter_name} not found in {self.modules_to_save.keys()}")
self.modules_to_save[self.active_adapter].requires_grad_(False)
self.modules_to_save[adapter_name].requires_grad_(True)
self._active_adapter = adapter_name
def _get_submodules(model, key):
parent = model.get_submodule(".".join(key.split(".")[:-1]))
target_name = key.split(".")[-1]
target = model.get_submodule(key)
return parent, target, target_name
def _freeze_adapter(model, adapter_name):
for n, p in model.named_parameters():
if adapter_name in n:
p.requires_grad = False
def _set_trainable(model, adapter_name):
key_list = [key for key, _ in model.named_modules()]
for key in key_list:
target_module_found = any(key.endswith(target_key) for target_key in model.modules_to_save)
if target_module_found:
parent, target, target_name = _get_submodules(model, key)
if isinstance(target, ModulesToSaveWrapper):
target.update(adapter_name)
target.set_adapter(target.active_adapter)
else:
new_module = ModulesToSaveWrapper(target, adapter_name)
new_module.set_adapter(adapter_name)
setattr(parent, target_name, new_module)
def _set_adapter(model, adapter_name):
def check_adapter_name(adapter_name):
if isinstance(adapter_name, str):
return adapter_name
# adapter_name is a list of str
if len(adapter_name) > 1:
raise ValueError("Only one adapter can be set at a time for modules_to_save")
elif len(adapter_name) == 0:
raise ValueError("Please specify at least one adapter to set")
adapter_name = adapter_name[0]
return adapter_name
for module in model.modules():
if isinstance(module, ModulesToSaveWrapper):
# only check the adapter_name if we actually encounter a ModulesToSaveWrapper, otherwise we don't care
adapter_name = check_adapter_name(adapter_name)
module.set_adapter(adapter_name)
def _prepare_prompt_learning_config(peft_config, model_config):
if peft_config.num_layers is None:
if "num_hidden_layers" in model_config:
num_layers = model_config["num_hidden_layers"]
elif "num_layers" in model_config:
num_layers = model_config["num_layers"]
elif "n_layer" in model_config:
num_layers = model_config["n_layer"]
else:
raise ValueError("Please specify `num_layers` in `peft_config`")
peft_config.num_layers = num_layers
if peft_config.token_dim is None:
if "hidden_size" in model_config:
token_dim = model_config["hidden_size"]
elif "n_embd" in model_config:
token_dim = model_config["n_embd"]
elif "d_model" in model_config:
token_dim = model_config["d_model"]
else:
raise ValueError("Please specify `token_dim` in `peft_config`")
peft_config.token_dim = token_dim
if peft_config.num_attention_heads is None:
if "num_attention_heads" in model_config:
num_attention_heads = model_config["num_attention_heads"]
elif "n_head" in model_config:
num_attention_heads = model_config["n_head"]
elif "num_heads" in model_config:
num_attention_heads = model_config["num_heads"]
elif "encoder_attention_heads" in model_config:
num_attention_heads = model_config["encoder_attention_heads"]
else:
raise ValueError("Please specify `num_attention_heads` in `peft_config`")
peft_config.num_attention_heads = num_attention_heads
if getattr(peft_config, "encoder_hidden_size", None) is None:
setattr(peft_config, "encoder_hidden_size", peft_config.token_dim)
return peft_config
def fsdp_auto_wrap_policy(model):
import functools
import os
from accelerate import FullyShardedDataParallelPlugin
from torch.distributed.fsdp.wrap import _or_policy, lambda_auto_wrap_policy, transformer_auto_wrap_policy
from ..tuners import PrefixEncoder, PromptEmbedding, PromptEncoder
default_transformer_cls_names_to_wrap = (
",".join(model._no_split_modules) if getattr(model, "_no_split_modules", None) is not None else ""
)
transformer_cls_names_to_wrap = os.environ.get(
"FSDP_TRANSFORMER_CLS_TO_WRAP", default_transformer_cls_names_to_wrap
).split(",")
transformer_cls_to_wrap = {PrefixEncoder, PromptEncoder, PromptEmbedding}
for layer_class in transformer_cls_names_to_wrap:
transformer_cls = FullyShardedDataParallelPlugin.get_module_class_from_name(model, layer_class)
if transformer_cls is None:
raise Exception("Could not find the transformer layer class to wrap in the model.")
else:
transformer_cls_to_wrap.add(transformer_cls)
def lambda_policy_fn(module):
if (
len(list(module.named_children())) == 0
and getattr(module, "weight", None) is not None
and module.weight.requires_grad
):
return True
return False
lambda_policy = functools.partial(lambda_auto_wrap_policy, lambda_fn=lambda_policy_fn)
transformer_wrap_policy = functools.partial(
transformer_auto_wrap_policy,
transformer_layer_cls=transformer_cls_to_wrap,
)
auto_wrap_policy = functools.partial(_or_policy, policies=[lambda_policy, transformer_wrap_policy])
return auto_wrap_policy
def transpose(weight, fan_in_fan_out):
if not fan_in_fan_out:
return weight
if isinstance(weight, torch.nn.Parameter):
return torch.nn.Parameter(weight.T)
return weight.T
def _is_valid_match(key: str, target_key: str):
"""
Helper function to match module names target_key and key. Makes sure that either the key is exactly the target_key
or the target_key is a submodule of key
"""
if key.endswith(target_key):
if len(key) > len(target_key):
return key.endswith("." + target_key) # must be a sub module
return True
return False
def _get_batch_size(input_ids: Optional[torch.Tensor], inputs_embeds: Optional[torch.Tensor]) -> int:
"""Get the batch size based on either input_ids or input_embeds
Raises an ValueError if both are None.
"""
if (input_ids is None) and (inputs_embeds is None):
raise ValueError("You have to provide either input_ids or inputs_embeds")
if input_ids is not None:
batch_size = input_ids.shape[0]
else:
batch_size = inputs_embeds.shape[0]
return batch_size
def get_quantization_config(model: torch.nn.Module, method: str):
"""
Get the quantization config of the related quantization method
"""
if (
hasattr(model, "config")
and hasattr(model.config, "quantization_config")
and (getattr(model, "quantization_method", None) == method)
):
return model.config.quantization_config
return None
def get_auto_gptq_quant_linear(gptq_quantization_config):
"""
Get the right AutoGPTQQuantLinear class based on the quantization config file
"""
if gptq_quantization_config is not None and is_auto_gptq_available():
from auto_gptq.utils.import_utils import dynamically_import_QuantLinear
desc_act = gptq_quantization_config.desc_act
group_size = gptq_quantization_config.group_size
bits = gptq_quantization_config.bits
if hasattr(gptq_quantization_config, "use_exllama"):
use_exllama = gptq_quantization_config.use_exllama
else:
use_exllama = not gptq_quantization_config.disable_exllama
if hasattr(gptq_quantization_config, "exllama_config"):
exllama_version = gptq_quantization_config.exllama_config["version"]
else:
exllama_version = 1
AutoGPTQQuantLinear = dynamically_import_QuantLinear(
use_triton=False,
desc_act=desc_act,
group_size=group_size,
bits=bits,
disable_exllama=not (use_exllama and exllama_version == 1),
disable_exllamav2=not (use_exllama and exllama_version == 2),
)
return AutoGPTQQuantLinear
return None
def id_tensor_storage(tensor: torch.Tensor) -> Tuple[torch.device, int, int]:
"""
Unique identifier to a tensor storage. Multiple different tensors can share the same underlying storage. For
example, "meta" tensors all share the same storage, and thus their identifier will all be equal. This identifier is
guaranteed to be unique and constant for this tensor's storage during its lifetime. Two tensor storages with
non-overlapping lifetimes may have the same id.
This method is the exact same copy of
https://github.com/huggingface/transformers/blob/main/src/transformers/pytorch_utils.py#L282C1-L300C58 but we added
it here manually to avoid import issue with old versions of transformers.
"""
if tensor.device.type == "xla" and is_torch_tpu_available():
# NOTE: xla tensors dont have storage
# use some other unique id to distinguish.
# this is a XLA tensor, it must be created using torch_xla's
# device. So the following import is safe:
import torch_xla
unique_id = torch_xla._XLAC._xla_get_tensor_id(tensor)
else:
unique_id = storage_ptr(tensor)
return tensor.device, unique_id, storage_size(tensor)
def cast_mixed_precision_params(model, dtype):
"""
Cast all non-trainable parameters of the model to the given `dtype`. The `dtype` can be `torch.float16` or
`torch.bfloat16` as per the mixed-precision training you are performing. The trainable parameters are cast to full
precision. This is meant to reduce the GPU memory usage when using PEFT methods by using half-precision dtype for
non-trainable parameters. Having the trainable parameters in full-precision preserves training stability when using
automatic mixed-precision training.
Args:
model (`torch.nn.Module`):
The model to cast the non-trainable parameters of.
dtype (`torch.dtype`):
The dtype to cast the non-trainable parameters to. The `dtype` can be `torch.float16` or
`torch.bfloat16` as per the mixed-precision training you are performing.
"""
for p in model.parameters():
if not p.requires_grad:
p.data = p.to(dtype)
else:
p.data = p.to(torch.float32)
| peft/src/peft/utils/other.py/0 | {
"file_path": "peft/src/peft/utils/other.py",
"repo_id": "peft",
"token_count": 8600
} | 158 |
# coding=utf-8
# Copyright 2023-present the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import unittest
import torch
from scipy import stats
from torch import nn
from peft import LoraConfig, get_peft_model
from peft.utils import infer_device
class InitializationTest(unittest.TestCase):
"""Test class to check the initialization of adapters."""
torch_device = infer_device()
def get_uniform(self, amin, amax, size=(10000,)):
unif = torch.distributions.uniform.Uniform(amin, amax)
samples = unif.sample(size)
return samples
def get_normal(self, mean, std, size=(10000,)):
normal = torch.distributions.normal.Normal(mean, std)
samples = normal.sample(size)
return samples
def get_model(self):
class MyModule(nn.Module):
def __init__(self):
super().__init__()
# choose a large weight so that averages are close to expected values
self.linear = nn.Linear(1000, 1000)
self.embed = nn.Embedding(1000, 1000)
self.conv2d = nn.Conv2d(100, 100, 3)
def forward(self, x):
return self.linear(x)
return MyModule().eval().to(self.torch_device)
def test_lora_linear_init_default(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["linear"])
model = get_peft_model(model, config)
weight_A = model.linear.lora_A["default"].weight
weight_B = model.linear.lora_B["default"].weight
# use statistical test to check if weight A is from a uniform distribution
unif = self.get_uniform(weight_A.min().item(), weight_A.max().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertGreater(p_value, 0.5)
# check that weight A is *not* from a normal distribution
normal = self.get_normal(weight_A.mean().item(), weight_A.std().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight B is zero
self.assertTrue((weight_B == 0.0).all())
def test_lora_linear_init_gaussian(self):
# use gaussian init
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["linear"], init_lora_weights="gaussian")
model = get_peft_model(model, config)
weight_A = model.linear.lora_A["default"].weight
weight_B = model.linear.lora_B["default"].weight
# use statistical test to check if weight A is from a normal distribution
normal = self.get_normal(0.0, 1 / config.r)
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
# import matplotlib.pyplot as plt
# x = weight_A.detach().flatten().cpu().numpy()
# breakpoint()
self.assertGreater(p_value, 0.5)
# check that weight A is *not* from a uniform distribution
unif = self.get_uniform(weight_A.min().item(), weight_A.max().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight B is zero
self.assertTrue((weight_B == 0.0).all())
def test_lora_linear_false(self):
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["linear"], init_lora_weights=False)
model = get_peft_model(model, config)
weight_B = model.linear.lora_B["default"].weight
# with init_lora_weights=False, weight B should *not* be zero. We don't care so much about the actual values
# as long as they are not zero, in order to avoid identity transformation.
self.assertFalse(torch.allclose(weight_B, torch.zeros_like(weight_B)))
def test_lora_embedding_default(self):
# embedding is initialized as a normal distribution, not kaiming uniform
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["embed"])
model = get_peft_model(model, config)
weight_A = model.embed.lora_embedding_A["default"]
weight_B = model.embed.lora_embedding_B["default"]
# use statistical test to check if weight B is from a normal distribution
normal = self.get_normal(0.0, 1.0)
_, p_value = stats.kstest(weight_B.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
self.assertGreater(p_value, 0.5)
# check that weight B is *not* from a uniform distribution
unif = self.get_uniform(weight_B.min().item(), weight_B.max().item())
_, p_value = stats.kstest(weight_B.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight A is zero
self.assertTrue((weight_A == 0.0).all())
def test_lora_embedding_gaussian(self):
# embedding does not change with init_lora_weights="gaussian" vs True
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["embed"], init_lora_weights="gaussian")
model = get_peft_model(model, config)
weight_A = model.embed.lora_embedding_A["default"]
weight_B = model.embed.lora_embedding_B["default"]
# use statistical test to check if weight B is from a normal distribution
normal = self.get_normal(0.0, 1.0)
_, p_value = stats.kstest(weight_B.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
self.assertGreater(p_value, 0.5)
# check that weight B is *not* from a uniform distribution
unif = self.get_uniform(weight_B.min().item(), weight_B.max().item())
_, p_value = stats.kstest(weight_B.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight A is zero
self.assertTrue((weight_A == 0.0).all())
def test_lora_embedding_false(self):
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["embed"], init_lora_weights=False)
model = get_peft_model(model, config)
weight_A = model.embed.lora_embedding_B["default"]
# with init_lora_weights=False, weight A should *not* be zero. We don't care so much about the actual values
# as long as they are not zero, in order to avoid identity transformation.
self.assertFalse(torch.allclose(weight_A, torch.zeros_like(weight_A)))
def test_lora_conv2d_default(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["conv2d"])
model = get_peft_model(model, config)
weight_A = model.conv2d.lora_A["default"].weight
weight_B = model.conv2d.lora_B["default"].weight
# use statistical test to check if weight A is from a uniform distribution
unif = self.get_uniform(weight_A.min().item(), weight_A.max().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertGreater(p_value, 0.5)
# check that weight A is *not* from a normal distribution
normal = self.get_normal(weight_A.mean().item(), weight_A.std().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight B is zero
self.assertTrue((weight_B == 0.0).all())
def test_lora_conv2d_init_gaussian(self):
# use gaussian init
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["conv2d"], init_lora_weights="gaussian")
model = get_peft_model(model, config)
weight_A = model.conv2d.lora_A["default"].weight
weight_B = model.conv2d.lora_B["default"].weight
# use statistical test to check if weight A is from a normal distribution
normal = self.get_normal(0.0, 1 / config.r)
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), normal.flatten().cpu().numpy())
self.assertGreater(p_value, 0.5)
# check that weight A is *not* from a uniform distribution
unif = self.get_uniform(weight_A.min().item(), weight_A.max().item())
_, p_value = stats.kstest(weight_A.detach().flatten().cpu().numpy(), unif.flatten().cpu().numpy())
self.assertLess(p_value, 0.05)
# check that weight B is zero
self.assertTrue((weight_B == 0.0).all())
def test_lora_conv2d_false(self):
torch.manual_seed(0)
model = self.get_model()
config = LoraConfig(target_modules=["conv2d"], init_lora_weights=False)
model = get_peft_model(model, config)
weight_B = model.conv2d.lora_B["default"].weight
# with init_lora_weights=False, weight B should *not* be zero. We don't care so much about the actual values
# as long as they are not zero, in order to avoid identity transformation.
self.assertFalse(torch.allclose(weight_B, torch.zeros_like(weight_B)))
def test_lora_scaling_default(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
# check scaling factor use_rslora=False
config = LoraConfig(target_modules=["linear", "embed", "conv2d"], lora_alpha=3, r=16, use_rslora=False)
model = get_peft_model(model, config)
expected_scaling = config.lora_alpha / config.r
self.assertTrue(model.linear.scaling["default"] == expected_scaling)
self.assertTrue(model.embed.scaling["default"] == expected_scaling)
self.assertTrue(model.conv2d.scaling["default"] == expected_scaling)
def test_rslora_scaling(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
# check scaling factor use_rslora=True
config = LoraConfig(target_modules=["linear", "embed", "conv2d"], lora_alpha=3, r=16, use_rslora=True)
model = get_peft_model(model, config)
expected_scaling = config.lora_alpha / (config.r**0.5)
self.assertTrue(model.linear.scaling["default"] == expected_scaling)
self.assertTrue(model.embed.scaling["default"] == expected_scaling)
self.assertTrue(model.conv2d.scaling["default"] == expected_scaling)
def test_lora_default_scaling_pattern(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
# check scaling factor use_rslora=False with rank and alpha pattern
config = LoraConfig(
target_modules=["linear", "embed", "conv2d"],
rank_pattern={"embed": 9, "conv2d": 16},
alpha_pattern={"linear": 11, "conv2d": 13},
lora_alpha=17,
r=25,
use_rslora=False,
)
model = get_peft_model(model, config)
expected_scaling = {
"linear": config.alpha_pattern["linear"] / config.r,
"embed": config.lora_alpha / config.rank_pattern["embed"],
"conv2d": config.alpha_pattern["conv2d"] / config.rank_pattern["conv2d"],
}
self.assertTrue(model.linear.scaling["default"] == expected_scaling["linear"])
self.assertTrue(model.embed.scaling["default"] == expected_scaling["embed"])
self.assertTrue(model.conv2d.scaling["default"] == expected_scaling["conv2d"])
def test_rslora_scaling_pattern(self):
# default is True
torch.manual_seed(0)
model = self.get_model()
# check scaling factor use_rslora=True with rank and alpha pattern
config = LoraConfig(
target_modules=["linear", "embed", "conv2d"],
rank_pattern={"embed": 9, "conv2d": 16},
alpha_pattern={"linear": 11, "conv2d": 13},
lora_alpha=17,
r=25,
use_rslora=True,
)
model = get_peft_model(model, config)
expected_scaling = {
"linear": config.alpha_pattern["linear"] / (config.r**0.5),
"embed": config.lora_alpha / (config.rank_pattern["embed"] ** 0.5),
"conv2d": config.alpha_pattern["conv2d"] / (config.rank_pattern["conv2d"] ** 0.5),
}
self.assertTrue(model.linear.scaling["default"] == expected_scaling["linear"])
self.assertTrue(model.embed.scaling["default"] == expected_scaling["embed"])
self.assertTrue(model.conv2d.scaling["default"] == expected_scaling["conv2d"])
| peft/tests/test_initialization.py/0 | {
"file_path": "peft/tests/test_initialization.py",
"repo_id": "peft",
"token_count": 5568
} | 159 |
# PyTorch Image Models
- [What's New](#whats-new)
- [Introduction](#introduction)
- [Models](#models)
- [Features](#features)
- [Results](#results)
- [Getting Started (Documentation)](#getting-started-documentation)
- [Train, Validation, Inference Scripts](#train-validation-inference-scripts)
- [Awesome PyTorch Resources](#awesome-pytorch-resources)
- [Licenses](#licenses)
- [Citing](#citing)
## What's New
❗Updates after Oct 10, 2022 are available in version >= 0.9❗
* Many changes since the last 0.6.x stable releases. They were previewed in 0.8.x dev releases but not everyone transitioned.
* `timm.models.layers` moved to `timm.layers`:
* `from timm.models.layers import name` will still work via deprecation mapping (but please transition to `timm.layers`).
* `import timm.models.layers.module` or `from timm.models.layers.module import name` needs to be changed now.
* Builder, helper, non-model modules in `timm.models` have a `_` prefix added, ie `timm.models.helpers` -> `timm.models._helpers`, there are temporary deprecation mapping files but those will be removed.
* All models now support `architecture.pretrained_tag` naming (ex `resnet50.rsb_a1`).
* The pretrained_tag is the specific weight variant (different head) for the architecture.
* Using only `architecture` defaults to the first weights in the default_cfgs for that model architecture.
* In adding pretrained tags, many model names that existed to differentiate were renamed to use the tag (ex: `vit_base_patch16_224_in21k` -> `vit_base_patch16_224.augreg_in21k`). There are deprecation mappings for these.
* A number of models had their checkpoints remaped to match architecture changes needed to better support `features_only=True`, there are `checkpoint_filter_fn` methods in any model module that was remapped. These can be passed to `timm.models.load_checkpoint(..., filter_fn=timm.models.swin_transformer_v2.checkpoint_filter_fn)` to remap your existing checkpoint.
* The Hugging Face Hub (https://huggingface.co/timm) is now the primary source for `timm` weights. Model cards include link to papers, original source, license.
* Previous 0.6.x can be cloned from [0.6.x](https://github.com/rwightman/pytorch-image-models/tree/0.6.x) branch or installed via pip with version.
### Jan 8, 2024
Datasets & transform refactoring
* HuggingFace streaming (iterable) dataset support (`--dataset hfids:org/dataset`)
* Webdataset wrapper tweaks for improved split info fetching, can auto fetch splits from supported HF hub webdataset
* Tested HF `datasets` and webdataset wrapper streaming from HF hub with recent `timm` ImageNet uploads to https://huggingface.co/timm
* Make input & target column/field keys consistent across datasets and pass via args
* Full monochrome support when using e:g: `--input-size 1 224 224` or `--in-chans 1`, sets PIL image conversion appropriately in dataset
* Improved several alternate crop & resize transforms (ResizeKeepRatio, RandomCropOrPad, etc) for use in PixParse document AI project
* Add SimCLR style color jitter prob along with grayscale and gaussian blur options to augmentations and args
* Allow train without validation set (`--val-split ''`) in train script
* Add `--bce-sum` (sum over class dim) and `--bce-pos-weight` (positive weighting) args for training as they're common BCE loss tweaks I was often hard coding
### Nov 23, 2023
* Added EfficientViT-Large models, thanks [SeeFun](https://github.com/seefun)
* Fix Python 3.7 compat, will be dropping support for it soon
* Other misc fixes
* Release 0.9.12
### Nov 20, 2023
* Added significant flexibility for Hugging Face Hub based timm models via `model_args` config entry. `model_args` will be passed as kwargs through to models on creation.
* See example at https://huggingface.co/gaunernst/vit_base_patch16_1024_128.audiomae_as2m_ft_as20k/blob/main/config.json
* Usage: https://github.com/huggingface/pytorch-image-models/discussions/2035
* Updated imagenet eval and test set csv files with latest models
* `vision_transformer.py` typing and doc cleanup by [Laureηt](https://github.com/Laurent2916)
* 0.9.11 release
### Nov 3, 2023
* [DFN (Data Filtering Networks)](https://huggingface.co/papers/2309.17425) and [MetaCLIP](https://huggingface.co/papers/2309.16671) ViT weights added
* DINOv2 'register' ViT model weights added (https://huggingface.co/papers/2309.16588, https://huggingface.co/papers/2304.07193)
* Add `quickgelu` ViT variants for OpenAI, DFN, MetaCLIP weights that use it (less efficient)
* Improved typing added to ResNet, MobileNet-v3 thanks to [Aryan](https://github.com/a-r-r-o-w)
* ImageNet-12k fine-tuned (from LAION-2B CLIP) `convnext_xxlarge`
* 0.9.9 release
### Oct 20, 2023
* [SigLIP](https://huggingface.co/papers/2303.15343) image tower weights supported in `vision_transformer.py`.
* Great potential for fine-tune and downstream feature use.
* Experimental 'register' support in vit models as per [Vision Transformers Need Registers](https://huggingface.co/papers/2309.16588)
* Updated RepViT with new weight release. Thanks [wangao](https://github.com/jameslahm)
* Add patch resizing support (on pretrained weight load) to Swin models
* 0.9.8 release pending
### Sep 1, 2023
* TinyViT added by [SeeFun](https://github.com/seefun)
* Fix EfficientViT (MIT) to use torch.autocast so it works back to PT 1.10
* 0.9.7 release
### Aug 28, 2023
* Add dynamic img size support to models in `vision_transformer.py`, `vision_transformer_hybrid.py`, `deit.py`, and `eva.py` w/o breaking backward compat.
* Add `dynamic_img_size=True` to args at model creation time to allow changing the grid size (interpolate abs and/or ROPE pos embed each forward pass).
* Add `dynamic_img_pad=True` to allow image sizes that aren't divisible by patch size (pad bottom right to patch size each forward pass).
* Enabling either dynamic mode will break FX tracing unless PatchEmbed module added as leaf.
* Existing method of resizing position embedding by passing different `img_size` (interpolate pretrained embed weights once) on creation still works.
* Existing method of changing `patch_size` (resize pretrained patch_embed weights once) on creation still works.
* Example validation cmd `python validate.py /imagenet --model vit_base_patch16_224 --amp --amp-dtype bfloat16 --img-size 255 --crop-pct 1.0 --model-kwargs dynamic_img_size=True dyamic_img_pad=True`
### Aug 25, 2023
* Many new models since last release
* FastViT - https://arxiv.org/abs/2303.14189
* MobileOne - https://arxiv.org/abs/2206.04040
* InceptionNeXt - https://arxiv.org/abs/2303.16900
* RepGhostNet - https://arxiv.org/abs/2211.06088 (thanks https://github.com/ChengpengChen)
* GhostNetV2 - https://arxiv.org/abs/2211.12905 (thanks https://github.com/yehuitang)
* EfficientViT (MSRA) - https://arxiv.org/abs/2305.07027 (thanks https://github.com/seefun)
* EfficientViT (MIT) - https://arxiv.org/abs/2205.14756 (thanks https://github.com/seefun)
* Add `--reparam` arg to `benchmark.py`, `onnx_export.py`, and `validate.py` to trigger layer reparameterization / fusion for models with any one of `reparameterize()`, `switch_to_deploy()` or `fuse()`
* Including FastViT, MobileOne, RepGhostNet, EfficientViT (MSRA), RepViT, RepVGG, and LeViT
* Preparing 0.9.6 'back to school' release
### Aug 11, 2023
* Swin, MaxViT, CoAtNet, and BEiT models support resizing of image/window size on creation with adaptation of pretrained weights
* Example validation cmd to test w/ non-square resize `python validate.py /imagenet --model swin_base_patch4_window7_224.ms_in22k_ft_in1k --amp --amp-dtype bfloat16 --input-size 3 256 320 --model-kwargs window_size=8,10 img_size=256,320`
### Aug 3, 2023
* Add GluonCV weights for HRNet w18_small and w18_small_v2. Converted by [SeeFun](https://github.com/seefun)
* Fix `selecsls*` model naming regression
* Patch and position embedding for ViT/EVA works for bfloat16/float16 weights on load (or activations for on-the-fly resize)
* v0.9.5 release prep
### July 27, 2023
* Added timm trained `seresnextaa201d_32x8d.sw_in12k_ft_in1k_384` weights (and `.sw_in12k` pretrain) with 87.3% top-1 on ImageNet-1k, best ImageNet ResNet family model I'm aware of.
* RepViT model and weights (https://arxiv.org/abs/2307.09283) added by [wangao](https://github.com/jameslahm)
* I-JEPA ViT feature weights (no classifier) added by [SeeFun](https://github.com/seefun)
* SAM-ViT (segment anything) feature weights (no classifier) added by [SeeFun](https://github.com/seefun)
* Add support for alternative feat extraction methods and -ve indices to EfficientNet
* Add NAdamW optimizer
* Misc fixes
### May 11, 2023
* `timm` 0.9 released, transition from 0.8.xdev releases
### May 10, 2023
* Hugging Face Hub downloading is now default, 1132 models on https://huggingface.co/timm, 1163 weights in `timm`
* DINOv2 vit feature backbone weights added thanks to [Leng Yue](https://github.com/leng-yue)
* FB MAE vit feature backbone weights added
* OpenCLIP DataComp-XL L/14 feat backbone weights added
* MetaFormer (poolformer-v2, caformer, convformer, updated poolformer (v1)) w/ weights added by [Fredo Guan](https://github.com/fffffgggg54)
* Experimental `get_intermediate_layers` function on vit/deit models for grabbing hidden states (inspired by DINO impl). This is WIP and may change significantly... feedback welcome.
* Model creation throws error if `pretrained=True` and no weights exist (instead of continuing with random initialization)
* Fix regression with inception / nasnet TF sourced weights with 1001 classes in original classifiers
* bitsandbytes (https://github.com/TimDettmers/bitsandbytes) optimizers added to factory, use `bnb` prefix, ie `bnbadam8bit`
* Misc cleanup and fixes
* Final testing before switching to a 0.9 and bringing `timm` out of pre-release state
### April 27, 2023
* 97% of `timm` models uploaded to HF Hub and almost all updated to support multi-weight pretrained configs
* Minor cleanup and refactoring of another batch of models as multi-weight added. More fused_attn (F.sdpa) and features_only support, and torchscript fixes.
### April 21, 2023
* Gradient accumulation support added to train script and tested (`--grad-accum-steps`), thanks [Taeksang Kim](https://github.com/voidbag)
* More weights on HF Hub (cspnet, cait, volo, xcit, tresnet, hardcorenas, densenet, dpn, vovnet, xception_aligned)
* Added `--head-init-scale` and `--head-init-bias` to train.py to scale classiifer head and set fixed bias for fine-tune
* Remove all InplaceABN (`inplace_abn`) use, replaced use in tresnet with standard BatchNorm (modified weights accordingly).
### April 12, 2023
* Add ONNX export script, validate script, helpers that I've had kicking around for along time. Tweak 'same' padding for better export w/ recent ONNX + pytorch.
* Refactor dropout args for vit and vit-like models, separate drop_rate into `drop_rate` (classifier dropout), `proj_drop_rate` (block mlp / out projections), `pos_drop_rate` (position embedding drop), `attn_drop_rate` (attention dropout). Also add patch dropout (FLIP) to vit and eva models.
* fused F.scaled_dot_product_attention support to more vit models, add env var (TIMM_FUSED_ATTN) to control, and config interface to enable/disable
* Add EVA-CLIP backbones w/ image tower weights, all the way up to 4B param 'enormous' model, and 336x336 OpenAI ViT mode that was missed.
### April 5, 2023
* ALL ResNet models pushed to Hugging Face Hub with multi-weight support
* All past `timm` trained weights added with recipe based tags to differentiate
* All ResNet strikes back A1/A2/A3 (seed 0) and R50 example B/C1/C2/D weights available
* Add torchvision v2 recipe weights to existing torchvision originals
* See comparison table in https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k_288#model-comparison
* New ImageNet-12k + ImageNet-1k fine-tunes available for a few anti-aliased ResNet models
* `resnetaa50d.sw_in12k_ft_in1k` - 81.7 @ 224, 82.6 @ 288
* `resnetaa101d.sw_in12k_ft_in1k` - 83.5 @ 224, 84.1 @ 288
* `seresnextaa101d_32x8d.sw_in12k_ft_in1k` - 86.0 @ 224, 86.5 @ 288
* `seresnextaa101d_32x8d.sw_in12k_ft_in1k_288` - 86.5 @ 288, 86.7 @ 320
### March 31, 2023
* Add first ConvNext-XXLarge CLIP -> IN-1k fine-tune and IN-12k intermediate fine-tunes for convnext-base/large CLIP models.
| model |top1 |top5 |img_size|param_count|gmacs |macts |
|----------------------------------------------------------------------------------------------------------------------|------|------|--------|-----------|------|------|
| [convnext_xxlarge.clip_laion2b_soup_ft_in1k](https://huggingface.co/timm/convnext_xxlarge.clip_laion2b_soup_ft_in1k) |88.612|98.704|256 |846.47 |198.09|124.45|
| convnext_large_mlp.clip_laion2b_soup_ft_in12k_in1k_384 |88.312|98.578|384 |200.13 |101.11|126.74|
| convnext_large_mlp.clip_laion2b_soup_ft_in12k_in1k_320 |87.968|98.47 |320 |200.13 |70.21 |88.02 |
| convnext_base.clip_laion2b_augreg_ft_in12k_in1k_384 |87.138|98.212|384 |88.59 |45.21 |84.49 |
| convnext_base.clip_laion2b_augreg_ft_in12k_in1k |86.344|97.97 |256 |88.59 |20.09 |37.55 |
* Add EVA-02 MIM pretrained and fine-tuned weights, push to HF hub and update model cards for all EVA models. First model over 90% top-1 (99% top-5)! Check out the original code & weights at https://github.com/baaivision/EVA for more details on their work blending MIM, CLIP w/ many model, dataset, and train recipe tweaks.
| model |top1 |top5 |param_count|img_size|
|----------------------------------------------------|------|------|-----------|--------|
| [eva02_large_patch14_448.mim_m38m_ft_in22k_in1k](https://huggingface.co/timm/eva02_large_patch14_448.mim_m38m_ft_in1k) |90.054|99.042|305.08 |448 |
| eva02_large_patch14_448.mim_in22k_ft_in22k_in1k |89.946|99.01 |305.08 |448 |
| eva_giant_patch14_560.m30m_ft_in22k_in1k |89.792|98.992|1014.45 |560 |
| eva02_large_patch14_448.mim_in22k_ft_in1k |89.626|98.954|305.08 |448 |
| eva02_large_patch14_448.mim_m38m_ft_in1k |89.57 |98.918|305.08 |448 |
| eva_giant_patch14_336.m30m_ft_in22k_in1k |89.56 |98.956|1013.01 |336 |
| eva_giant_patch14_336.clip_ft_in1k |89.466|98.82 |1013.01 |336 |
| eva_large_patch14_336.in22k_ft_in22k_in1k |89.214|98.854|304.53 |336 |
| eva_giant_patch14_224.clip_ft_in1k |88.882|98.678|1012.56 |224 |
| eva02_base_patch14_448.mim_in22k_ft_in22k_in1k |88.692|98.722|87.12 |448 |
| eva_large_patch14_336.in22k_ft_in1k |88.652|98.722|304.53 |336 |
| eva_large_patch14_196.in22k_ft_in22k_in1k |88.592|98.656|304.14 |196 |
| eva02_base_patch14_448.mim_in22k_ft_in1k |88.23 |98.564|87.12 |448 |
| eva_large_patch14_196.in22k_ft_in1k |87.934|98.504|304.14 |196 |
| eva02_small_patch14_336.mim_in22k_ft_in1k |85.74 |97.614|22.13 |336 |
| eva02_tiny_patch14_336.mim_in22k_ft_in1k |80.658|95.524|5.76 |336 |
* Multi-weight and HF hub for DeiT and MLP-Mixer based models
### March 22, 2023
* More weights pushed to HF hub along with multi-weight support, including: `regnet.py`, `rexnet.py`, `byobnet.py`, `resnetv2.py`, `swin_transformer.py`, `swin_transformer_v2.py`, `swin_transformer_v2_cr.py`
* Swin Transformer models support feature extraction (NCHW feat maps for `swinv2_cr_*`, and NHWC for all others) and spatial embedding outputs.
* FocalNet (from https://github.com/microsoft/FocalNet) models and weights added with significant refactoring, feature extraction, no fixed resolution / sizing constraint
* RegNet weights increased with HF hub push, SWAG, SEER, and torchvision v2 weights. SEER is pretty poor wrt to performance for model size, but possibly useful.
* More ImageNet-12k pretrained and 1k fine-tuned `timm` weights:
* `rexnetr_200.sw_in12k_ft_in1k` - 82.6 @ 224, 83.2 @ 288
* `rexnetr_300.sw_in12k_ft_in1k` - 84.0 @ 224, 84.5 @ 288
* `regnety_120.sw_in12k_ft_in1k` - 85.0 @ 224, 85.4 @ 288
* `regnety_160.lion_in12k_ft_in1k` - 85.6 @ 224, 86.0 @ 288
* `regnety_160.sw_in12k_ft_in1k` - 85.6 @ 224, 86.0 @ 288 (compare to SWAG PT + 1k FT this is same BUT much lower res, blows SEER FT away)
* Model name deprecation + remapping functionality added (a milestone for bringing 0.8.x out of pre-release). Mappings being added...
* Minor bug fixes and improvements.
### Feb 26, 2023
* Add ConvNeXt-XXLarge CLIP pretrained image tower weights for fine-tune & features (fine-tuning TBD) -- see [model card](https://huggingface.co/laion/CLIP-convnext_xxlarge-laion2B-s34B-b82K-augreg-soup)
* Update `convnext_xxlarge` default LayerNorm eps to 1e-5 (for CLIP weights, improved stability)
* 0.8.15dev0
### Feb 20, 2023
* Add 320x320 `convnext_large_mlp.clip_laion2b_ft_320` and `convnext_lage_mlp.clip_laion2b_ft_soup_320` CLIP image tower weights for features & fine-tune
* 0.8.13dev0 pypi release for latest changes w/ move to huggingface org
### Feb 16, 2023
* `safetensor` checkpoint support added
* Add ideas from 'Scaling Vision Transformers to 22 B. Params' (https://arxiv.org/abs/2302.05442) -- qk norm, RmsNorm, parallel block
* Add F.scaled_dot_product_attention support (PyTorch 2.0 only) to `vit_*`, `vit_relpos*`, `coatnet` / `maxxvit` (to start)
* Lion optimizer (w/ multi-tensor option) added (https://arxiv.org/abs/2302.06675)
* gradient checkpointing works with `features_only=True`
### Feb 7, 2023
* New inference benchmark numbers added in [results](results/) folder.
* Add convnext LAION CLIP trained weights and initial set of in1k fine-tunes
* `convnext_base.clip_laion2b_augreg_ft_in1k` - 86.2% @ 256x256
* `convnext_base.clip_laiona_augreg_ft_in1k_384` - 86.5% @ 384x384
* `convnext_large_mlp.clip_laion2b_augreg_ft_in1k` - 87.3% @ 256x256
* `convnext_large_mlp.clip_laion2b_augreg_ft_in1k_384` - 87.9% @ 384x384
* Add DaViT models. Supports `features_only=True`. Adapted from https://github.com/dingmyu/davit by [Fredo](https://github.com/fffffgggg54).
* Use a common NormMlpClassifierHead across MaxViT, ConvNeXt, DaViT
* Add EfficientFormer-V2 model, update EfficientFormer, and refactor LeViT (closely related architectures). Weights on HF hub.
* New EfficientFormer-V2 arch, significant refactor from original at (https://github.com/snap-research/EfficientFormer). Supports `features_only=True`.
* Minor updates to EfficientFormer.
* Refactor LeViT models to stages, add `features_only=True` support to new `conv` variants, weight remap required.
* Move ImageNet meta-data (synsets, indices) from `/results` to [`timm/data/_info`](timm/data/_info/).
* Add ImageNetInfo / DatasetInfo classes to provide labelling for various ImageNet classifier layouts in `timm`
* Update `inference.py` to use, try: `python inference.py /folder/to/images --model convnext_small.in12k --label-type detail --topk 5`
* Ready for 0.8.10 pypi pre-release (final testing).
### Jan 20, 2023
* Add two convnext 12k -> 1k fine-tunes at 384x384
* `convnext_tiny.in12k_ft_in1k_384` - 85.1 @ 384
* `convnext_small.in12k_ft_in1k_384` - 86.2 @ 384
* Push all MaxxViT weights to HF hub, and add new ImageNet-12k -> 1k fine-tunes for `rw` base MaxViT and CoAtNet 1/2 models
|model |top1 |top5 |samples / sec |Params (M) |GMAC |Act (M)|
|------------------------------------------------------------------------------------------------------------------------|----:|----:|--------------:|--------------:|-----:|------:|
|[maxvit_xlarge_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_xlarge_tf_512.in21k_ft_in1k) |88.53|98.64| 21.76| 475.77|534.14|1413.22|
|[maxvit_xlarge_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_xlarge_tf_384.in21k_ft_in1k) |88.32|98.54| 42.53| 475.32|292.78| 668.76|
|[maxvit_base_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_base_tf_512.in21k_ft_in1k) |88.20|98.53| 50.87| 119.88|138.02| 703.99|
|[maxvit_large_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_large_tf_512.in21k_ft_in1k) |88.04|98.40| 36.42| 212.33|244.75| 942.15|
|[maxvit_large_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_large_tf_384.in21k_ft_in1k) |87.98|98.56| 71.75| 212.03|132.55| 445.84|
|[maxvit_base_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_base_tf_384.in21k_ft_in1k) |87.92|98.54| 104.71| 119.65| 73.80| 332.90|
|[maxvit_rmlp_base_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/maxvit_rmlp_base_rw_384.sw_in12k_ft_in1k) |87.81|98.37| 106.55| 116.14| 70.97| 318.95|
|[maxxvitv2_rmlp_base_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/maxxvitv2_rmlp_base_rw_384.sw_in12k_ft_in1k) |87.47|98.37| 149.49| 116.09| 72.98| 213.74|
|[coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k) |87.39|98.31| 160.80| 73.88| 47.69| 209.43|
|[maxvit_rmlp_base_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/maxvit_rmlp_base_rw_224.sw_in12k_ft_in1k) |86.89|98.02| 375.86| 116.14| 23.15| 92.64|
|[maxxvitv2_rmlp_base_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/maxxvitv2_rmlp_base_rw_224.sw_in12k_ft_in1k) |86.64|98.02| 501.03| 116.09| 24.20| 62.77|
|[maxvit_base_tf_512.in1k](https://huggingface.co/timm/maxvit_base_tf_512.in1k) |86.60|97.92| 50.75| 119.88|138.02| 703.99|
|[coatnet_2_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_2_rw_224.sw_in12k_ft_in1k) |86.57|97.89| 631.88| 73.87| 15.09| 49.22|
|[maxvit_large_tf_512.in1k](https://huggingface.co/timm/maxvit_large_tf_512.in1k) |86.52|97.88| 36.04| 212.33|244.75| 942.15|
|[coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k) |86.49|97.90| 620.58| 73.88| 15.18| 54.78|
|[maxvit_base_tf_384.in1k](https://huggingface.co/timm/maxvit_base_tf_384.in1k) |86.29|97.80| 101.09| 119.65| 73.80| 332.90|
|[maxvit_large_tf_384.in1k](https://huggingface.co/timm/maxvit_large_tf_384.in1k) |86.23|97.69| 70.56| 212.03|132.55| 445.84|
|[maxvit_small_tf_512.in1k](https://huggingface.co/timm/maxvit_small_tf_512.in1k) |86.10|97.76| 88.63| 69.13| 67.26| 383.77|
|[maxvit_tiny_tf_512.in1k](https://huggingface.co/timm/maxvit_tiny_tf_512.in1k) |85.67|97.58| 144.25| 31.05| 33.49| 257.59|
|[maxvit_small_tf_384.in1k](https://huggingface.co/timm/maxvit_small_tf_384.in1k) |85.54|97.46| 188.35| 69.02| 35.87| 183.65|
|[maxvit_tiny_tf_384.in1k](https://huggingface.co/timm/maxvit_tiny_tf_384.in1k) |85.11|97.38| 293.46| 30.98| 17.53| 123.42|
|[maxvit_large_tf_224.in1k](https://huggingface.co/timm/maxvit_large_tf_224.in1k) |84.93|96.97| 247.71| 211.79| 43.68| 127.35|
|[coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k) |84.90|96.96| 1025.45| 41.72| 8.11| 40.13|
|[maxvit_base_tf_224.in1k](https://huggingface.co/timm/maxvit_base_tf_224.in1k) |84.85|96.99| 358.25| 119.47| 24.04| 95.01|
|[maxxvit_rmlp_small_rw_256.sw_in1k](https://huggingface.co/timm/maxxvit_rmlp_small_rw_256.sw_in1k) |84.63|97.06| 575.53| 66.01| 14.67| 58.38|
|[coatnet_rmlp_2_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_224.sw_in1k) |84.61|96.74| 625.81| 73.88| 15.18| 54.78|
|[maxvit_rmlp_small_rw_224.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_small_rw_224.sw_in1k) |84.49|96.76| 693.82| 64.90| 10.75| 49.30|
|[maxvit_small_tf_224.in1k](https://huggingface.co/timm/maxvit_small_tf_224.in1k) |84.43|96.83| 647.96| 68.93| 11.66| 53.17|
|[maxvit_rmlp_tiny_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_tiny_rw_256.sw_in1k) |84.23|96.78| 807.21| 29.15| 6.77| 46.92|
|[coatnet_1_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_1_rw_224.sw_in1k) |83.62|96.38| 989.59| 41.72| 8.04| 34.60|
|[maxvit_tiny_rw_224.sw_in1k](https://huggingface.co/timm/maxvit_tiny_rw_224.sw_in1k) |83.50|96.50| 1100.53| 29.06| 5.11| 33.11|
|[maxvit_tiny_tf_224.in1k](https://huggingface.co/timm/maxvit_tiny_tf_224.in1k) |83.41|96.59| 1004.94| 30.92| 5.60| 35.78|
|[coatnet_rmlp_1_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_1_rw_224.sw_in1k) |83.36|96.45| 1093.03| 41.69| 7.85| 35.47|
|[maxxvitv2_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxxvitv2_nano_rw_256.sw_in1k) |83.11|96.33| 1276.88| 23.70| 6.26| 23.05|
|[maxxvit_rmlp_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxxvit_rmlp_nano_rw_256.sw_in1k) |83.03|96.34| 1341.24| 16.78| 4.37| 26.05|
|[maxvit_rmlp_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_nano_rw_256.sw_in1k) |82.96|96.26| 1283.24| 15.50| 4.47| 31.92|
|[maxvit_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_nano_rw_256.sw_in1k) |82.93|96.23| 1218.17| 15.45| 4.46| 30.28|
|[coatnet_bn_0_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_bn_0_rw_224.sw_in1k) |82.39|96.19| 1600.14| 27.44| 4.67| 22.04|
|[coatnet_0_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_0_rw_224.sw_in1k) |82.39|95.84| 1831.21| 27.44| 4.43| 18.73|
|[coatnet_rmlp_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_nano_rw_224.sw_in1k) |82.05|95.87| 2109.09| 15.15| 2.62| 20.34|
|[coatnext_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnext_nano_rw_224.sw_in1k) |81.95|95.92| 2525.52| 14.70| 2.47| 12.80|
|[coatnet_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_nano_rw_224.sw_in1k) |81.70|95.64| 2344.52| 15.14| 2.41| 15.41|
|[maxvit_rmlp_pico_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_pico_rw_256.sw_in1k) |80.53|95.21| 1594.71| 7.52| 1.85| 24.86|
### Jan 11, 2023
* Update ConvNeXt ImageNet-12k pretrain series w/ two new fine-tuned weights (and pre FT `.in12k` tags)
* `convnext_nano.in12k_ft_in1k` - 82.3 @ 224, 82.9 @ 288 (previously released)
* `convnext_tiny.in12k_ft_in1k` - 84.2 @ 224, 84.5 @ 288
* `convnext_small.in12k_ft_in1k` - 85.2 @ 224, 85.3 @ 288
### Jan 6, 2023
* Finally got around to adding `--model-kwargs` and `--opt-kwargs` to scripts to pass through rare args directly to model classes from cmd line
* `train.py /imagenet --model resnet50 --amp --model-kwargs output_stride=16 act_layer=silu`
* `train.py /imagenet --model vit_base_patch16_clip_224 --img-size 240 --amp --model-kwargs img_size=240 patch_size=12`
* Cleanup some popular models to better support arg passthrough / merge with model configs, more to go.
### Jan 5, 2023
* ConvNeXt-V2 models and weights added to existing `convnext.py`
* Paper: [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](http://arxiv.org/abs/2301.00808)
* Reference impl: https://github.com/facebookresearch/ConvNeXt-V2 (NOTE: weights currently CC-BY-NC)
## Introduction
Py**T**orch **Im**age **M**odels (`timm`) is a collection of image models, layers, utilities, optimizers, schedulers, data-loaders / augmentations, and reference training / validation scripts that aim to pull together a wide variety of SOTA models with ability to reproduce ImageNet training results.
The work of many others is present here. I've tried to make sure all source material is acknowledged via links to github, arxiv papers, etc in the README, documentation, and code docstrings. Please let me know if I missed anything.
## Features
### Models
All model architecture families include variants with pretrained weights. There are specific model variants without any weights, it is NOT a bug. Help training new or better weights is always appreciated.
* Aggregating Nested Transformers - https://arxiv.org/abs/2105.12723
* BEiT - https://arxiv.org/abs/2106.08254
* Big Transfer ResNetV2 (BiT) - https://arxiv.org/abs/1912.11370
* Bottleneck Transformers - https://arxiv.org/abs/2101.11605
* CaiT (Class-Attention in Image Transformers) - https://arxiv.org/abs/2103.17239
* CoaT (Co-Scale Conv-Attentional Image Transformers) - https://arxiv.org/abs/2104.06399
* CoAtNet (Convolution and Attention) - https://arxiv.org/abs/2106.04803
* ConvNeXt - https://arxiv.org/abs/2201.03545
* ConvNeXt-V2 - http://arxiv.org/abs/2301.00808
* ConViT (Soft Convolutional Inductive Biases Vision Transformers)- https://arxiv.org/abs/2103.10697
* CspNet (Cross-Stage Partial Networks) - https://arxiv.org/abs/1911.11929
* DeiT - https://arxiv.org/abs/2012.12877
* DeiT-III - https://arxiv.org/pdf/2204.07118.pdf
* DenseNet - https://arxiv.org/abs/1608.06993
* DLA - https://arxiv.org/abs/1707.06484
* DPN (Dual-Path Network) - https://arxiv.org/abs/1707.01629
* EdgeNeXt - https://arxiv.org/abs/2206.10589
* EfficientFormer - https://arxiv.org/abs/2206.01191
* EfficientNet (MBConvNet Family)
* EfficientNet NoisyStudent (B0-B7, L2) - https://arxiv.org/abs/1911.04252
* EfficientNet AdvProp (B0-B8) - https://arxiv.org/abs/1911.09665
* EfficientNet (B0-B7) - https://arxiv.org/abs/1905.11946
* EfficientNet-EdgeTPU (S, M, L) - https://ai.googleblog.com/2019/08/efficientnet-edgetpu-creating.html
* EfficientNet V2 - https://arxiv.org/abs/2104.00298
* FBNet-C - https://arxiv.org/abs/1812.03443
* MixNet - https://arxiv.org/abs/1907.09595
* MNASNet B1, A1 (Squeeze-Excite), and Small - https://arxiv.org/abs/1807.11626
* MobileNet-V2 - https://arxiv.org/abs/1801.04381
* Single-Path NAS - https://arxiv.org/abs/1904.02877
* TinyNet - https://arxiv.org/abs/2010.14819
* EfficientViT (MIT) - https://arxiv.org/abs/2205.14756
* EfficientViT (MSRA) - https://arxiv.org/abs/2305.07027
* EVA - https://arxiv.org/abs/2211.07636
* EVA-02 - https://arxiv.org/abs/2303.11331
* FastViT - https://arxiv.org/abs/2303.14189
* FlexiViT - https://arxiv.org/abs/2212.08013
* FocalNet (Focal Modulation Networks) - https://arxiv.org/abs/2203.11926
* GCViT (Global Context Vision Transformer) - https://arxiv.org/abs/2206.09959
* GhostNet - https://arxiv.org/abs/1911.11907
* GhostNet-V2 - https://arxiv.org/abs/2211.12905
* gMLP - https://arxiv.org/abs/2105.08050
* GPU-Efficient Networks - https://arxiv.org/abs/2006.14090
* Halo Nets - https://arxiv.org/abs/2103.12731
* HRNet - https://arxiv.org/abs/1908.07919
* InceptionNeXt - https://arxiv.org/abs/2303.16900
* Inception-V3 - https://arxiv.org/abs/1512.00567
* Inception-ResNet-V2 and Inception-V4 - https://arxiv.org/abs/1602.07261
* Lambda Networks - https://arxiv.org/abs/2102.08602
* LeViT (Vision Transformer in ConvNet's Clothing) - https://arxiv.org/abs/2104.01136
* MaxViT (Multi-Axis Vision Transformer) - https://arxiv.org/abs/2204.01697
* MetaFormer (PoolFormer-v2, ConvFormer, CAFormer) - https://arxiv.org/abs/2210.13452
* MLP-Mixer - https://arxiv.org/abs/2105.01601
* MobileNet-V3 (MBConvNet w/ Efficient Head) - https://arxiv.org/abs/1905.02244
* FBNet-V3 - https://arxiv.org/abs/2006.02049
* HardCoRe-NAS - https://arxiv.org/abs/2102.11646
* LCNet - https://arxiv.org/abs/2109.15099
* MobileOne - https://arxiv.org/abs/2206.04040
* MobileViT - https://arxiv.org/abs/2110.02178
* MobileViT-V2 - https://arxiv.org/abs/2206.02680
* MViT-V2 (Improved Multiscale Vision Transformer) - https://arxiv.org/abs/2112.01526
* NASNet-A - https://arxiv.org/abs/1707.07012
* NesT - https://arxiv.org/abs/2105.12723
* NFNet-F - https://arxiv.org/abs/2102.06171
* NF-RegNet / NF-ResNet - https://arxiv.org/abs/2101.08692
* PNasNet - https://arxiv.org/abs/1712.00559
* PoolFormer (MetaFormer) - https://arxiv.org/abs/2111.11418
* Pooling-based Vision Transformer (PiT) - https://arxiv.org/abs/2103.16302
* PVT-V2 (Improved Pyramid Vision Transformer) - https://arxiv.org/abs/2106.13797
* RegNet - https://arxiv.org/abs/2003.13678
* RegNetZ - https://arxiv.org/abs/2103.06877
* RepVGG - https://arxiv.org/abs/2101.03697
* RepGhostNet - https://arxiv.org/abs/2211.06088
* RepViT - https://arxiv.org/abs/2307.09283
* ResMLP - https://arxiv.org/abs/2105.03404
* ResNet/ResNeXt
* ResNet (v1b/v1.5) - https://arxiv.org/abs/1512.03385
* ResNeXt - https://arxiv.org/abs/1611.05431
* 'Bag of Tricks' / Gluon C, D, E, S variations - https://arxiv.org/abs/1812.01187
* Weakly-supervised (WSL) Instagram pretrained / ImageNet tuned ResNeXt101 - https://arxiv.org/abs/1805.00932
* Semi-supervised (SSL) / Semi-weakly Supervised (SWSL) ResNet/ResNeXts - https://arxiv.org/abs/1905.00546
* ECA-Net (ECAResNet) - https://arxiv.org/abs/1910.03151v4
* Squeeze-and-Excitation Networks (SEResNet) - https://arxiv.org/abs/1709.01507
* ResNet-RS - https://arxiv.org/abs/2103.07579
* Res2Net - https://arxiv.org/abs/1904.01169
* ResNeSt - https://arxiv.org/abs/2004.08955
* ReXNet - https://arxiv.org/abs/2007.00992
* SelecSLS - https://arxiv.org/abs/1907.00837
* Selective Kernel Networks - https://arxiv.org/abs/1903.06586
* Sequencer2D - https://arxiv.org/abs/2205.01972
* Swin S3 (AutoFormerV2) - https://arxiv.org/abs/2111.14725
* Swin Transformer - https://arxiv.org/abs/2103.14030
* Swin Transformer V2 - https://arxiv.org/abs/2111.09883
* Transformer-iN-Transformer (TNT) - https://arxiv.org/abs/2103.00112
* TResNet - https://arxiv.org/abs/2003.13630
* Twins (Spatial Attention in Vision Transformers) - https://arxiv.org/pdf/2104.13840.pdf
* Visformer - https://arxiv.org/abs/2104.12533
* Vision Transformer - https://arxiv.org/abs/2010.11929
* VOLO (Vision Outlooker) - https://arxiv.org/abs/2106.13112
* VovNet V2 and V1 - https://arxiv.org/abs/1911.06667
* Xception - https://arxiv.org/abs/1610.02357
* Xception (Modified Aligned, Gluon) - https://arxiv.org/abs/1802.02611
* Xception (Modified Aligned, TF) - https://arxiv.org/abs/1802.02611
* XCiT (Cross-Covariance Image Transformers) - https://arxiv.org/abs/2106.09681
### Optimizers
Included optimizers available via `create_optimizer` / `create_optimizer_v2` factory methods:
* `adabelief` an implementation of AdaBelief adapted from https://github.com/juntang-zhuang/Adabelief-Optimizer - https://arxiv.org/abs/2010.07468
* `adafactor` adapted from [FAIRSeq impl](https://github.com/pytorch/fairseq/blob/master/fairseq/optim/adafactor.py) - https://arxiv.org/abs/1804.04235
* `adahessian` by [David Samuel](https://github.com/davda54/ada-hessian) - https://arxiv.org/abs/2006.00719
* `adamp` and `sgdp` by [Naver ClovAI](https://github.com/clovaai) - https://arxiv.org/abs/2006.08217
* `adan` an implementation of Adan adapted from https://github.com/sail-sg/Adan - https://arxiv.org/abs/2208.06677
* `lamb` an implementation of Lamb and LambC (w/ trust-clipping) cleaned up and modified to support use with XLA - https://arxiv.org/abs/1904.00962
* `lars` an implementation of LARS and LARC (w/ trust-clipping) - https://arxiv.org/abs/1708.03888
* `lion` and implementation of Lion adapted from https://github.com/google/automl/tree/master/lion - https://arxiv.org/abs/2302.06675
* `lookahead` adapted from impl by [Liam](https://github.com/alphadl/lookahead.pytorch) - https://arxiv.org/abs/1907.08610
* `madgrad` - and implementation of MADGRAD adapted from https://github.com/facebookresearch/madgrad - https://arxiv.org/abs/2101.11075
* `nadam` an implementation of Adam w/ Nesterov momentum
* `nadamw` an impementation of AdamW (Adam w/ decoupled weight-decay) w/ Nesterov momentum. A simplified impl based on https://github.com/mlcommons/algorithmic-efficiency
* `novograd` by [Masashi Kimura](https://github.com/convergence-lab/novograd) - https://arxiv.org/abs/1905.11286
* `radam` by [Liyuan Liu](https://github.com/LiyuanLucasLiu/RAdam) - https://arxiv.org/abs/1908.03265
* `rmsprop_tf` adapted from PyTorch RMSProp by myself. Reproduces much improved Tensorflow RMSProp behaviour
* `sgdw` and implementation of SGD w/ decoupled weight-decay
* `fused<name>` optimizers by name with [NVIDIA Apex](https://github.com/NVIDIA/apex/tree/master/apex/optimizers) installed
* `bits<name>` optimizers by name with [BitsAndBytes](https://github.com/TimDettmers/bitsandbytes) installed
### Augmentations
* Random Erasing from [Zhun Zhong](https://github.com/zhunzhong07/Random-Erasing/blob/master/transforms.py) - https://arxiv.org/abs/1708.04896)
* Mixup - https://arxiv.org/abs/1710.09412
* CutMix - https://arxiv.org/abs/1905.04899
* AutoAugment (https://arxiv.org/abs/1805.09501) and RandAugment (https://arxiv.org/abs/1909.13719) ImageNet configurations modeled after impl for EfficientNet training (https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/autoaugment.py)
* AugMix w/ JSD loss, JSD w/ clean + augmented mixing support works with AutoAugment and RandAugment as well - https://arxiv.org/abs/1912.02781
* SplitBachNorm - allows splitting batch norm layers between clean and augmented (auxiliary batch norm) data
### Regularization
* DropPath aka "Stochastic Depth" - https://arxiv.org/abs/1603.09382
* DropBlock - https://arxiv.org/abs/1810.12890
* Blur Pooling - https://arxiv.org/abs/1904.11486
### Other
Several (less common) features that I often utilize in my projects are included. Many of their additions are the reason why I maintain my own set of models, instead of using others' via PIP:
* All models have a common default configuration interface and API for
* accessing/changing the classifier - `get_classifier` and `reset_classifier`
* doing a forward pass on just the features - `forward_features` (see [documentation](https://huggingface.co/docs/timm/feature_extraction))
* these makes it easy to write consistent network wrappers that work with any of the models
* All models support multi-scale feature map extraction (feature pyramids) via create_model (see [documentation](https://huggingface.co/docs/timm/feature_extraction))
* `create_model(name, features_only=True, out_indices=..., output_stride=...)`
* `out_indices` creation arg specifies which feature maps to return, these indices are 0 based and generally correspond to the `C(i + 1)` feature level.
* `output_stride` creation arg controls output stride of the network by using dilated convolutions. Most networks are stride 32 by default. Not all networks support this.
* feature map channel counts, reduction level (stride) can be queried AFTER model creation via the `.feature_info` member
* All models have a consistent pretrained weight loader that adapts last linear if necessary, and from 3 to 1 channel input if desired
* High performance [reference training, validation, and inference scripts](https://huggingface.co/docs/timm/training_script) that work in several process/GPU modes:
* NVIDIA DDP w/ a single GPU per process, multiple processes with APEX present (AMP mixed-precision optional)
* PyTorch DistributedDataParallel w/ multi-gpu, single process (AMP disabled as it crashes when enabled)
* PyTorch w/ single GPU single process (AMP optional)
* A dynamic global pool implementation that allows selecting from average pooling, max pooling, average + max, or concat([average, max]) at model creation. All global pooling is adaptive average by default and compatible with pretrained weights.
* A 'Test Time Pool' wrapper that can wrap any of the included models and usually provides improved performance doing inference with input images larger than the training size. Idea adapted from original DPN implementation when I ported (https://github.com/cypw/DPNs)
* Learning rate schedulers
* Ideas adopted from
* [AllenNLP schedulers](https://github.com/allenai/allennlp/tree/master/allennlp/training/learning_rate_schedulers)
* [FAIRseq lr_scheduler](https://github.com/pytorch/fairseq/tree/master/fairseq/optim/lr_scheduler)
* SGDR: Stochastic Gradient Descent with Warm Restarts (https://arxiv.org/abs/1608.03983)
* Schedulers include `step`, `cosine` w/ restarts, `tanh` w/ restarts, `plateau`
* Space-to-Depth by [mrT23](https://github.com/mrT23/TResNet/blob/master/src/models/tresnet/layers/space_to_depth.py) (https://arxiv.org/abs/1801.04590) -- original paper?
* Adaptive Gradient Clipping (https://arxiv.org/abs/2102.06171, https://github.com/deepmind/deepmind-research/tree/master/nfnets)
* An extensive selection of channel and/or spatial attention modules:
* Bottleneck Transformer - https://arxiv.org/abs/2101.11605
* CBAM - https://arxiv.org/abs/1807.06521
* Effective Squeeze-Excitation (ESE) - https://arxiv.org/abs/1911.06667
* Efficient Channel Attention (ECA) - https://arxiv.org/abs/1910.03151
* Gather-Excite (GE) - https://arxiv.org/abs/1810.12348
* Global Context (GC) - https://arxiv.org/abs/1904.11492
* Halo - https://arxiv.org/abs/2103.12731
* Involution - https://arxiv.org/abs/2103.06255
* Lambda Layer - https://arxiv.org/abs/2102.08602
* Non-Local (NL) - https://arxiv.org/abs/1711.07971
* Squeeze-and-Excitation (SE) - https://arxiv.org/abs/1709.01507
* Selective Kernel (SK) - (https://arxiv.org/abs/1903.06586
* Split (SPLAT) - https://arxiv.org/abs/2004.08955
* Shifted Window (SWIN) - https://arxiv.org/abs/2103.14030
## Results
Model validation results can be found in the [results tables](results/README.md)
## Getting Started (Documentation)
The official documentation can be found at https://huggingface.co/docs/hub/timm. Documentation contributions are welcome.
[Getting Started with PyTorch Image Models (timm): A Practitioner’s Guide](https://towardsdatascience.com/getting-started-with-pytorch-image-models-timm-a-practitioners-guide-4e77b4bf9055) by [Chris Hughes](https://github.com/Chris-hughes10) is an extensive blog post covering many aspects of `timm` in detail.
[timmdocs](http://timm.fast.ai/) is an alternate set of documentation for `timm`. A big thanks to [Aman Arora](https://github.com/amaarora) for his efforts creating timmdocs.
[paperswithcode](https://paperswithcode.com/lib/timm) is a good resource for browsing the models within `timm`.
## Train, Validation, Inference Scripts
The root folder of the repository contains reference train, validation, and inference scripts that work with the included models and other features of this repository. They are adaptable for other datasets and use cases with a little hacking. See [documentation](https://huggingface.co/docs/timm/training_script).
## Awesome PyTorch Resources
One of the greatest assets of PyTorch is the community and their contributions. A few of my favourite resources that pair well with the models and components here are listed below.
### Object Detection, Instance and Semantic Segmentation
* Detectron2 - https://github.com/facebookresearch/detectron2
* Segmentation Models (Semantic) - https://github.com/qubvel/segmentation_models.pytorch
* EfficientDet (Obj Det, Semantic soon) - https://github.com/rwightman/efficientdet-pytorch
### Computer Vision / Image Augmentation
* Albumentations - https://github.com/albumentations-team/albumentations
* Kornia - https://github.com/kornia/kornia
### Knowledge Distillation
* RepDistiller - https://github.com/HobbitLong/RepDistiller
* torchdistill - https://github.com/yoshitomo-matsubara/torchdistill
### Metric Learning
* PyTorch Metric Learning - https://github.com/KevinMusgrave/pytorch-metric-learning
### Training / Frameworks
* fastai - https://github.com/fastai/fastai
## Licenses
### Code
The code here is licensed Apache 2.0. I've taken care to make sure any third party code included or adapted has compatible (permissive) licenses such as MIT, BSD, etc. I've made an effort to avoid any GPL / LGPL conflicts. That said, it is your responsibility to ensure you comply with licenses here and conditions of any dependent licenses. Where applicable, I've linked the sources/references for various components in docstrings. If you think I've missed anything please create an issue.
### Pretrained Weights
So far all of the pretrained weights available here are pretrained on ImageNet with a select few that have some additional pretraining (see extra note below). ImageNet was released for non-commercial research purposes only (https://image-net.org/download). It's not clear what the implications of that are for the use of pretrained weights from that dataset. Any models I have trained with ImageNet are done for research purposes and one should assume that the original dataset license applies to the weights. It's best to seek legal advice if you intend to use the pretrained weights in a commercial product.
#### Pretrained on more than ImageNet
Several weights included or references here were pretrained with proprietary datasets that I do not have access to. These include the Facebook WSL, SSL, SWSL ResNe(Xt) and the Google Noisy Student EfficientNet models. The Facebook models have an explicit non-commercial license (CC-BY-NC 4.0, https://github.com/facebookresearch/semi-supervised-ImageNet1K-models, https://github.com/facebookresearch/WSL-Images). The Google models do not appear to have any restriction beyond the Apache 2.0 license (and ImageNet concerns). In either case, you should contact Facebook or Google with any questions.
## Citing
### BibTeX
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/rwightman/pytorch-image-models}}
}
```
### Latest DOI
[](https://zenodo.org/badge/latestdoi/168799526)
| pytorch-image-models/README.md/0 | {
"file_path": "pytorch-image-models/README.md",
"repo_id": "pytorch-image-models",
"token_count": 19602
} | 160 |
"""
Run this script to generate the model-index files in `models` from the templates in `.templates/models`.
"""
import argparse
from pathlib import Path
from jinja2 import Environment, FileSystemLoader
import modelindex
def generate_readmes(templates_path: Path, dest_path: Path):
"""Add the code snippet template to the readmes"""
readme_templates_path = templates_path / "models"
code_template_path = templates_path / "code_snippets.md"
env = Environment(
loader=FileSystemLoader([readme_templates_path, readme_templates_path.parent]),
)
for readme in readme_templates_path.iterdir():
if readme.suffix == ".md":
template = env.get_template(readme.name)
# get the first model_name for this model family
mi = modelindex.load(str(readme))
model_name = mi.models[0].name
full_content = template.render(model_name=model_name)
# generate full_readme
with open(dest_path / readme.name, "w") as f:
f.write(full_content)
def main():
parser = argparse.ArgumentParser(description="Model index generation config")
parser.add_argument(
"-t",
"--templates",
default=Path(__file__).parent / ".templates",
type=str,
help="Location of the markdown templates",
)
parser.add_argument(
"-d",
"--dest",
default=Path(__file__).parent / "models",
type=str,
help="Destination folder that contains the generated model-index files.",
)
args = parser.parse_args()
templates_path = Path(args.templates)
dest_readmes_path = Path(args.dest)
generate_readmes(
templates_path,
dest_readmes_path,
)
if __name__ == "__main__":
main()
| pytorch-image-models/docs/models/.templates/generate_readmes.py/0 | {
"file_path": "pytorch-image-models/docs/models/.templates/generate_readmes.py",
"repo_id": "pytorch-image-models",
"token_count": 725
} | 161 |
# (Gluon) Inception v3
**Inception v3** is a convolutional neural network architecture from the Inception family that makes several improvements including using [Label Smoothing](https://paperswithcode.com/method/label-smoothing), Factorized 7 x 7 convolutions, and the use of an [auxiliary classifer](https://paperswithcode.com/method/auxiliary-classifier) to propagate label information lower down the network (along with the use of batch normalization for layers in the sidehead). The key building block is an [Inception Module](https://paperswithcode.com/method/inception-v3-module).
The weights from this model were ported from [Gluon](https://cv.gluon.ai/model_zoo/classification.html).
{% include 'code_snippets.md' %}
## How do I train this model?
You can follow the [timm recipe scripts](https://rwightman.github.io/pytorch-image-models/scripts/) for training a new model afresh.
## Citation
```BibTeX
@article{DBLP:journals/corr/SzegedyVISW15,
author = {Christian Szegedy and
Vincent Vanhoucke and
Sergey Ioffe and
Jonathon Shlens and
Zbigniew Wojna},
title = {Rethinking the Inception Architecture for Computer Vision},
journal = {CoRR},
volume = {abs/1512.00567},
year = {2015},
url = {http://arxiv.org/abs/1512.00567},
archivePrefix = {arXiv},
eprint = {1512.00567},
timestamp = {Mon, 13 Aug 2018 16:49:07 +0200},
biburl = {https://dblp.org/rec/journals/corr/SzegedyVISW15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
<!--
Type: model-index
Collections:
- Name: Gloun Inception v3
Paper:
Title: Rethinking the Inception Architecture for Computer Vision
URL: https://paperswithcode.com/paper/rethinking-the-inception-architecture-for
Models:
- Name: gluon_inception_v3
In Collection: Gloun Inception v3
Metadata:
FLOPs: 7352418880
Parameters: 23830000
File Size: 95567055
Architecture:
- 1x1 Convolution
- Auxiliary Classifier
- Average Pooling
- Average Pooling
- Batch Normalization
- Convolution
- Dense Connections
- Dropout
- Inception-v3 Module
- Max Pooling
- ReLU
- Softmax
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: gluon_inception_v3
Crop Pct: '0.875'
Image Size: '299'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/inception_v3.py#L464
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/gluon_inception_v3-9f746940.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 78.8%
Top 5 Accuracy: 94.38%
-->
| pytorch-image-models/docs/models/.templates/models/gloun-inception-v3.md/0 | {
"file_path": "pytorch-image-models/docs/models/.templates/models/gloun-inception-v3.md",
"repo_id": "pytorch-image-models",
"token_count": 1073
} | 162 |
# MobileNet v2
**MobileNetV2** is a convolutional neural network architecture that seeks to perform well on mobile devices. It is based on an [inverted residual structure](https://paperswithcode.com/method/inverted-residual-block) where the residual connections are between the bottleneck layers. The intermediate expansion layer uses lightweight depthwise convolutions to filter features as a source of non-linearity. As a whole, the architecture of MobileNetV2 contains the initial fully convolution layer with 32 filters, followed by 19 residual bottleneck layers.
{% include 'code_snippets.md' %}
## How do I train this model?
You can follow the [timm recipe scripts](https://rwightman.github.io/pytorch-image-models/scripts/) for training a new model afresh.
## Citation
```BibTeX
@article{DBLP:journals/corr/abs-1801-04381,
author = {Mark Sandler and
Andrew G. Howard and
Menglong Zhu and
Andrey Zhmoginov and
Liang{-}Chieh Chen},
title = {Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification,
Detection and Segmentation},
journal = {CoRR},
volume = {abs/1801.04381},
year = {2018},
url = {http://arxiv.org/abs/1801.04381},
archivePrefix = {arXiv},
eprint = {1801.04381},
timestamp = {Tue, 12 Jan 2021 15:30:06 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1801-04381.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
<!--
Type: model-index
Collections:
- Name: MobileNet V2
Paper:
Title: 'MobileNetV2: Inverted Residuals and Linear Bottlenecks'
URL: https://paperswithcode.com/paper/mobilenetv2-inverted-residuals-and-linear
Models:
- Name: mobilenetv2_100
In Collection: MobileNet V2
Metadata:
FLOPs: 401920448
Parameters: 3500000
File Size: 14202571
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Depthwise Separable Convolution
- Dropout
- Inverted Residual Block
- Max Pooling
- ReLU6
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- RMSProp
- Weight Decay
Training Data:
- ImageNet
Training Resources: 16x GPUs
ID: mobilenetv2_100
LR: 0.045
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1536
Image Size: '224'
Weight Decay: 4.0e-05
Interpolation: bicubic
RMSProp Decay: 0.9
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/efficientnet.py#L955
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv2_100_ra-b33bc2c4.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 72.95%
Top 5 Accuracy: 91.0%
- Name: mobilenetv2_110d
In Collection: MobileNet V2
Metadata:
FLOPs: 573958832
Parameters: 4520000
File Size: 18316431
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Depthwise Separable Convolution
- Dropout
- Inverted Residual Block
- Max Pooling
- ReLU6
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- RMSProp
- Weight Decay
Training Data:
- ImageNet
Training Resources: 16x GPUs
ID: mobilenetv2_110d
LR: 0.045
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1536
Image Size: '224'
Weight Decay: 4.0e-05
Interpolation: bicubic
RMSProp Decay: 0.9
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/efficientnet.py#L969
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv2_110d_ra-77090ade.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 75.05%
Top 5 Accuracy: 92.19%
- Name: mobilenetv2_120d
In Collection: MobileNet V2
Metadata:
FLOPs: 888510048
Parameters: 5830000
File Size: 23651121
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Depthwise Separable Convolution
- Dropout
- Inverted Residual Block
- Max Pooling
- ReLU6
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- RMSProp
- Weight Decay
Training Data:
- ImageNet
Training Resources: 16x GPUs
ID: mobilenetv2_120d
LR: 0.045
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1536
Image Size: '224'
Weight Decay: 4.0e-05
Interpolation: bicubic
RMSProp Decay: 0.9
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/efficientnet.py#L977
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv2_120d_ra-5987e2ed.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 77.28%
Top 5 Accuracy: 93.51%
- Name: mobilenetv2_140
In Collection: MobileNet V2
Metadata:
FLOPs: 770196784
Parameters: 6110000
File Size: 24673555
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Depthwise Separable Convolution
- Dropout
- Inverted Residual Block
- Max Pooling
- ReLU6
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- RMSProp
- Weight Decay
Training Data:
- ImageNet
Training Resources: 16x GPUs
ID: mobilenetv2_140
LR: 0.045
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1536
Image Size: '224'
Weight Decay: 4.0e-05
Interpolation: bicubic
RMSProp Decay: 0.9
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/efficientnet.py#L962
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv2_140_ra-21a4e913.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 76.51%
Top 5 Accuracy: 93.0%
-->
| pytorch-image-models/docs/models/.templates/models/mobilenet-v2.md/0 | {
"file_path": "pytorch-image-models/docs/models/.templates/models/mobilenet-v2.md",
"repo_id": "pytorch-image-models",
"token_count": 2583
} | 163 |
# SE-ResNeXt
**SE ResNeXt** is a variant of a [ResNext](https://www.paperswithcode.com/method/resneXt) that employs [squeeze-and-excitation blocks](https://paperswithcode.com/method/squeeze-and-excitation-block) to enable the network to perform dynamic channel-wise feature recalibration.
{% include 'code_snippets.md' %}
## How do I train this model?
You can follow the [timm recipe scripts](https://rwightman.github.io/pytorch-image-models/scripts/) for training a new model afresh.
## Citation
```BibTeX
@misc{hu2019squeezeandexcitation,
title={Squeeze-and-Excitation Networks},
author={Jie Hu and Li Shen and Samuel Albanie and Gang Sun and Enhua Wu},
year={2019},
eprint={1709.01507},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
Type: model-index
Collections:
- Name: SEResNeXt
Paper:
Title: Squeeze-and-Excitation Networks
URL: https://paperswithcode.com/paper/squeeze-and-excitation-networks
Models:
- Name: seresnext26d_32x4d
In Collection: SEResNeXt
Metadata:
FLOPs: 3507053024
Parameters: 16810000
File Size: 67425193
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Techniques:
- Label Smoothing
- SGD with Momentum
- Weight Decay
Training Data:
- ImageNet
Training Resources: 8x NVIDIA Titan X GPUs
ID: seresnext26d_32x4d
LR: 0.6
Epochs: 100
Layers: 26
Dropout: 0.2
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1024
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/a7f95818e44b281137503bcf4b3e3e94d8ffa52f/timm/models/resnet.py#L1234
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26d_32x4d-80fa48a3.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 77.59%
Top 5 Accuracy: 93.61%
- Name: seresnext26t_32x4d
In Collection: SEResNeXt
Metadata:
FLOPs: 3466436448
Parameters: 16820000
File Size: 67414838
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Techniques:
- Label Smoothing
- SGD with Momentum
- Weight Decay
Training Data:
- ImageNet
Training Resources: 8x NVIDIA Titan X GPUs
ID: seresnext26t_32x4d
LR: 0.6
Epochs: 100
Layers: 26
Dropout: 0.2
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1024
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/a7f95818e44b281137503bcf4b3e3e94d8ffa52f/timm/models/resnet.py#L1246
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26tn_32x4d-569cb627.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 77.99%
Top 5 Accuracy: 93.73%
- Name: seresnext50_32x4d
In Collection: SEResNeXt
Metadata:
FLOPs: 5475179184
Parameters: 27560000
File Size: 110569859
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Techniques:
- Label Smoothing
- SGD with Momentum
- Weight Decay
Training Data:
- ImageNet
Training Resources: 8x NVIDIA Titan X GPUs
ID: seresnext50_32x4d
LR: 0.6
Epochs: 100
Layers: 50
Dropout: 0.2
Crop Pct: '0.875'
Momentum: 0.9
Batch Size: 1024
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/a7f95818e44b281137503bcf4b3e3e94d8ffa52f/timm/models/resnet.py#L1267
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext50_32x4d_racm-a304a460.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 81.27%
Top 5 Accuracy: 95.62%
-->
| pytorch-image-models/docs/models/.templates/models/seresnext.md/0 | {
"file_path": "pytorch-image-models/docs/models/.templates/models/seresnext.md",
"repo_id": "pytorch-image-models",
"token_count": 1929
} | 164 |
# Wide ResNet
**Wide Residual Networks** are a variant on [ResNets](https://paperswithcode.com/method/resnet) where we decrease depth and increase the width of residual networks. This is achieved through the use of [wide residual blocks](https://paperswithcode.com/method/wide-residual-block).
{% include 'code_snippets.md' %}
## How do I train this model?
You can follow the [timm recipe scripts](https://rwightman.github.io/pytorch-image-models/scripts/) for training a new model afresh.
## Citation
```BibTeX
@article{DBLP:journals/corr/ZagoruykoK16,
author = {Sergey Zagoruyko and
Nikos Komodakis},
title = {Wide Residual Networks},
journal = {CoRR},
volume = {abs/1605.07146},
year = {2016},
url = {http://arxiv.org/abs/1605.07146},
archivePrefix = {arXiv},
eprint = {1605.07146},
timestamp = {Mon, 13 Aug 2018 16:46:42 +0200},
biburl = {https://dblp.org/rec/journals/corr/ZagoruykoK16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
<!--
Type: model-index
Collections:
- Name: Wide ResNet
Paper:
Title: Wide Residual Networks
URL: https://paperswithcode.com/paper/wide-residual-networks
Models:
- Name: wide_resnet101_2
In Collection: Wide ResNet
Metadata:
FLOPs: 29304929280
Parameters: 126890000
File Size: 254695146
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Max Pooling
- ReLU
- Residual Connection
- Softmax
- Wide Residual Block
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: wide_resnet101_2
Crop Pct: '0.875'
Image Size: '224'
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/5f9aff395c224492e9e44248b15f44b5cc095d9c/timm/models/resnet.py#L802
Weights: https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 78.85%
Top 5 Accuracy: 94.28%
- Name: wide_resnet50_2
In Collection: Wide ResNet
Metadata:
FLOPs: 14688058368
Parameters: 68880000
File Size: 275853271
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Max Pooling
- ReLU
- Residual Connection
- Softmax
- Wide Residual Block
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: wide_resnet50_2
Crop Pct: '0.875'
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/5f9aff395c224492e9e44248b15f44b5cc095d9c/timm/models/resnet.py#L790
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/wide_resnet50_racm-8234f177.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 81.45%
Top 5 Accuracy: 95.52%
-->
| pytorch-image-models/docs/models/.templates/models/wide-resnet.md/0 | {
"file_path": "pytorch-image-models/docs/models/.templates/models/wide-resnet.md",
"repo_id": "pytorch-image-models",
"token_count": 1220
} | 165 |
# CSP-DarkNet
**CSPDarknet53** is a convolutional neural network and backbone for object detection that uses [DarkNet-53](https://paperswithcode.com/method/darknet-53). It employs a CSPNet strategy to partition the feature map of the base layer into two parts and then merges them through a cross-stage hierarchy. The use of a split and merge strategy allows for more gradient flow through the network.
This CNN is used as the backbone for [YOLOv4](https://paperswithcode.com/method/yolov4).
## How do I use this model on an image?
To load a pretrained model:
```py
>>> import timm
>>> model = timm.create_model('cspdarknet53', pretrained=True)
>>> model.eval()
```
To load and preprocess the image:
```py
>>> import urllib
>>> from PIL import Image
>>> from timm.data import resolve_data_config
>>> from timm.data.transforms_factory import create_transform
>>> config = resolve_data_config({}, model=model)
>>> transform = create_transform(**config)
>>> url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
>>> urllib.request.urlretrieve(url, filename)
>>> img = Image.open(filename).convert('RGB')
>>> tensor = transform(img).unsqueeze(0) # transform and add batch dimension
```
To get the model predictions:
```py
>>> import torch
>>> with torch.no_grad():
... out = model(tensor)
>>> probabilities = torch.nn.functional.softmax(out[0], dim=0)
>>> print(probabilities.shape)
>>> # prints: torch.Size([1000])
```
To get the top-5 predictions class names:
```py
>>> # Get imagenet class mappings
>>> url, filename = ("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt", "imagenet_classes.txt")
>>> urllib.request.urlretrieve(url, filename)
>>> with open("imagenet_classes.txt", "r") as f:
... categories = [s.strip() for s in f.readlines()]
>>> # Print top categories per image
>>> top5_prob, top5_catid = torch.topk(probabilities, 5)
>>> for i in range(top5_prob.size(0)):
... print(categories[top5_catid[i]], top5_prob[i].item())
>>> # prints class names and probabilities like:
>>> # [('Samoyed', 0.6425196528434753), ('Pomeranian', 0.04062102362513542), ('keeshond', 0.03186424449086189), ('white wolf', 0.01739676296710968), ('Eskimo dog', 0.011717947199940681)]
```
Replace the model name with the variant you want to use, e.g. `cspdarknet53`. You can find the IDs in the model summaries at the top of this page.
To extract image features with this model, follow the [timm feature extraction examples](../feature_extraction), just change the name of the model you want to use.
## How do I finetune this model?
You can finetune any of the pre-trained models just by changing the classifier (the last layer).
```py
>>> model = timm.create_model('cspdarknet53', pretrained=True, num_classes=NUM_FINETUNE_CLASSES)
```
To finetune on your own dataset, you have to write a training loop or adapt [timm's training
script](https://github.com/rwightman/pytorch-image-models/blob/master/train.py) to use your dataset.
## How do I train this model?
You can follow the [timm recipe scripts](../scripts) for training a new model afresh.
## Citation
```BibTeX
@misc{bochkovskiy2020yolov4,
title={YOLOv4: Optimal Speed and Accuracy of Object Detection},
author={Alexey Bochkovskiy and Chien-Yao Wang and Hong-Yuan Mark Liao},
year={2020},
eprint={2004.10934},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
Type: model-index
Collections:
- Name: CSP DarkNet
Paper:
Title: 'YOLOv4: Optimal Speed and Accuracy of Object Detection'
URL: https://paperswithcode.com/paper/yolov4-optimal-speed-and-accuracy-of-object
Models:
- Name: cspdarknet53
In Collection: CSP DarkNet
Metadata:
FLOPs: 8545018880
Parameters: 27640000
File Size: 110775135
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Mish
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- CutMix
- Label Smoothing
- Mosaic
- Polynomial Learning Rate Decay
- SGD with Momentum
- Self-Adversarial Training
- Weight Decay
Training Data:
- ImageNet
Training Resources: 1x NVIDIA RTX 2070 GPU
ID: cspdarknet53
LR: 0.1
Layers: 53
Crop Pct: '0.887'
Momentum: 0.9
Batch Size: 128
Image Size: '256'
Warmup Steps: 1000
Weight Decay: 0.0005
Interpolation: bilinear
Training Steps: 8000000
FPS (GPU RTX 2070): 66
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/cspnet.py#L441
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/cspdarknet53_ra_256-d05c7c21.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 80.05%
Top 5 Accuracy: 95.09%
--> | pytorch-image-models/hfdocs/source/models/csp-darknet.mdx/0 | {
"file_path": "pytorch-image-models/hfdocs/source/models/csp-darknet.mdx",
"repo_id": "pytorch-image-models",
"token_count": 1756
} | 166 |
# (Gluon) SE-ResNeXt
**SE ResNeXt** is a variant of a [ResNext](https://www.paperswithcode.com/method/resnext) that employs [squeeze-and-excitation blocks](https://paperswithcode.com/method/squeeze-and-excitation-block) to enable the network to perform dynamic channel-wise feature recalibration.
The weights from this model were ported from [Gluon](https://cv.gluon.ai/model_zoo/classification.html).
## How do I use this model on an image?
To load a pretrained model:
```py
>>> import timm
>>> model = timm.create_model('gluon_seresnext101_32x4d', pretrained=True)
>>> model.eval()
```
To load and preprocess the image:
```py
>>> import urllib
>>> from PIL import Image
>>> from timm.data import resolve_data_config
>>> from timm.data.transforms_factory import create_transform
>>> config = resolve_data_config({}, model=model)
>>> transform = create_transform(**config)
>>> url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
>>> urllib.request.urlretrieve(url, filename)
>>> img = Image.open(filename).convert('RGB')
>>> tensor = transform(img).unsqueeze(0) # transform and add batch dimension
```
To get the model predictions:
```py
>>> import torch
>>> with torch.no_grad():
... out = model(tensor)
>>> probabilities = torch.nn.functional.softmax(out[0], dim=0)
>>> print(probabilities.shape)
>>> # prints: torch.Size([1000])
```
To get the top-5 predictions class names:
```py
>>> # Get imagenet class mappings
>>> url, filename = ("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt", "imagenet_classes.txt")
>>> urllib.request.urlretrieve(url, filename)
>>> with open("imagenet_classes.txt", "r") as f:
... categories = [s.strip() for s in f.readlines()]
>>> # Print top categories per image
>>> top5_prob, top5_catid = torch.topk(probabilities, 5)
>>> for i in range(top5_prob.size(0)):
... print(categories[top5_catid[i]], top5_prob[i].item())
>>> # prints class names and probabilities like:
>>> # [('Samoyed', 0.6425196528434753), ('Pomeranian', 0.04062102362513542), ('keeshond', 0.03186424449086189), ('white wolf', 0.01739676296710968), ('Eskimo dog', 0.011717947199940681)]
```
Replace the model name with the variant you want to use, e.g. `gluon_seresnext101_32x4d`. You can find the IDs in the model summaries at the top of this page.
To extract image features with this model, follow the [timm feature extraction examples](../feature_extraction), just change the name of the model you want to use.
## How do I finetune this model?
You can finetune any of the pre-trained models just by changing the classifier (the last layer).
```py
>>> model = timm.create_model('gluon_seresnext101_32x4d', pretrained=True, num_classes=NUM_FINETUNE_CLASSES)
```
To finetune on your own dataset, you have to write a training loop or adapt [timm's training
script](https://github.com/rwightman/pytorch-image-models/blob/master/train.py) to use your dataset.
## How do I train this model?
You can follow the [timm recipe scripts](../scripts) for training a new model afresh.
## Citation
```BibTeX
@misc{hu2019squeezeandexcitation,
title={Squeeze-and-Excitation Networks},
author={Jie Hu and Li Shen and Samuel Albanie and Gang Sun and Enhua Wu},
year={2019},
eprint={1709.01507},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
Type: model-index
Collections:
- Name: Gloun SEResNeXt
Paper:
Title: Squeeze-and-Excitation Networks
URL: https://paperswithcode.com/paper/squeeze-and-excitation-networks
Models:
- Name: gluon_seresnext101_32x4d
In Collection: Gloun SEResNeXt
Metadata:
FLOPs: 10302923504
Parameters: 48960000
File Size: 196505510
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: gluon_seresnext101_32x4d
Crop Pct: '0.875'
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/gluon_resnet.py#L219
Weights: https://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext101_32x4d-cf52900d.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 80.87%
Top 5 Accuracy: 95.29%
- Name: gluon_seresnext101_64x4d
In Collection: Gloun SEResNeXt
Metadata:
FLOPs: 19958950640
Parameters: 88230000
File Size: 353875948
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: gluon_seresnext101_64x4d
Crop Pct: '0.875'
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/gluon_resnet.py#L229
Weights: https://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext101_64x4d-f9926f93.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 80.88%
Top 5 Accuracy: 95.31%
- Name: gluon_seresnext50_32x4d
In Collection: Gloun SEResNeXt
Metadata:
FLOPs: 5475179184
Parameters: 27560000
File Size: 110578827
Architecture:
- 1x1 Convolution
- Batch Normalization
- Convolution
- Global Average Pooling
- Grouped Convolution
- Max Pooling
- ReLU
- ResNeXt Block
- Residual Connection
- Softmax
- Squeeze-and-Excitation Block
Tasks:
- Image Classification
Training Data:
- ImageNet
ID: gluon_seresnext50_32x4d
Crop Pct: '0.875'
Image Size: '224'
Interpolation: bicubic
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/gluon_resnet.py#L209
Weights: https://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext50_32x4d-90cf2d6e.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 79.92%
Top 5 Accuracy: 94.82%
--> | pytorch-image-models/hfdocs/source/models/gloun-seresnext.mdx/0 | {
"file_path": "pytorch-image-models/hfdocs/source/models/gloun-seresnext.mdx",
"repo_id": "pytorch-image-models",
"token_count": 2538
} | 167 |
# PNASNet
**Progressive Neural Architecture Search**, or **PNAS**, is a method for learning the structure of convolutional neural networks (CNNs). It uses a sequential model-based optimization (SMBO) strategy, where we search the space of cell structures, starting with simple (shallow) models and progressing to complex ones, pruning out unpromising structures as we go.
## How do I use this model on an image?
To load a pretrained model:
```py
>>> import timm
>>> model = timm.create_model('pnasnet5large', pretrained=True)
>>> model.eval()
```
To load and preprocess the image:
```py
>>> import urllib
>>> from PIL import Image
>>> from timm.data import resolve_data_config
>>> from timm.data.transforms_factory import create_transform
>>> config = resolve_data_config({}, model=model)
>>> transform = create_transform(**config)
>>> url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
>>> urllib.request.urlretrieve(url, filename)
>>> img = Image.open(filename).convert('RGB')
>>> tensor = transform(img).unsqueeze(0) # transform and add batch dimension
```
To get the model predictions:
```py
>>> import torch
>>> with torch.no_grad():
... out = model(tensor)
>>> probabilities = torch.nn.functional.softmax(out[0], dim=0)
>>> print(probabilities.shape)
>>> # prints: torch.Size([1000])
```
To get the top-5 predictions class names:
```py
>>> # Get imagenet class mappings
>>> url, filename = ("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt", "imagenet_classes.txt")
>>> urllib.request.urlretrieve(url, filename)
>>> with open("imagenet_classes.txt", "r") as f:
... categories = [s.strip() for s in f.readlines()]
>>> # Print top categories per image
>>> top5_prob, top5_catid = torch.topk(probabilities, 5)
>>> for i in range(top5_prob.size(0)):
... print(categories[top5_catid[i]], top5_prob[i].item())
>>> # prints class names and probabilities like:
>>> # [('Samoyed', 0.6425196528434753), ('Pomeranian', 0.04062102362513542), ('keeshond', 0.03186424449086189), ('white wolf', 0.01739676296710968), ('Eskimo dog', 0.011717947199940681)]
```
Replace the model name with the variant you want to use, e.g. `pnasnet5large`. You can find the IDs in the model summaries at the top of this page.
To extract image features with this model, follow the [timm feature extraction examples](../feature_extraction), just change the name of the model you want to use.
## How do I finetune this model?
You can finetune any of the pre-trained models just by changing the classifier (the last layer).
```py
>>> model = timm.create_model('pnasnet5large', pretrained=True, num_classes=NUM_FINETUNE_CLASSES)
```
To finetune on your own dataset, you have to write a training loop or adapt [timm's training
script](https://github.com/rwightman/pytorch-image-models/blob/master/train.py) to use your dataset.
## How do I train this model?
You can follow the [timm recipe scripts](../scripts) for training a new model afresh.
## Citation
```BibTeX
@misc{liu2018progressive,
title={Progressive Neural Architecture Search},
author={Chenxi Liu and Barret Zoph and Maxim Neumann and Jonathon Shlens and Wei Hua and Li-Jia Li and Li Fei-Fei and Alan Yuille and Jonathan Huang and Kevin Murphy},
year={2018},
eprint={1712.00559},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
<!--
Type: model-index
Collections:
- Name: PNASNet
Paper:
Title: Progressive Neural Architecture Search
URL: https://paperswithcode.com/paper/progressive-neural-architecture-search
Models:
- Name: pnasnet5large
In Collection: PNASNet
Metadata:
FLOPs: 31458865950
Parameters: 86060000
File Size: 345153926
Architecture:
- Average Pooling
- Batch Normalization
- Convolution
- Depthwise Separable Convolution
- Dropout
- ReLU
Tasks:
- Image Classification
Training Techniques:
- Label Smoothing
- RMSProp
- Weight Decay
Training Data:
- ImageNet
Training Resources: 100x NVIDIA P100 GPUs
ID: pnasnet5large
LR: 0.015
Dropout: 0.5
Crop Pct: '0.911'
Momentum: 0.9
Batch Size: 1600
Image Size: '331'
Interpolation: bicubic
Label Smoothing: 0.1
Code: https://github.com/rwightman/pytorch-image-models/blob/d8e69206be253892b2956341fea09fdebfaae4e3/timm/models/pnasnet.py#L343
Weights: https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-cadene/pnasnet5large-bf079911.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 0.98%
Top 5 Accuracy: 18.58%
--> | pytorch-image-models/hfdocs/source/models/pnasnet.mdx/0 | {
"file_path": "pytorch-image-models/hfdocs/source/models/pnasnet.mdx",
"repo_id": "pytorch-image-models",
"token_count": 1622
} | 168 |
# SSL ResNet
**Residual Networks**, or **ResNets**, learn residual functions with reference to the layer inputs, instead of learning unreferenced functions. Instead of hoping each few stacked layers directly fit a desired underlying mapping, residual nets let these layers fit a residual mapping. They stack [residual blocks](https://paperswithcode.com/method/residual-block) ontop of each other to form network: e.g. a ResNet-50 has fifty layers using these blocks.
The model in this collection utilises semi-supervised learning to improve the performance of the model. The approach brings important gains to standard architectures for image, video and fine-grained classification.
Please note the CC-BY-NC 4.0 license on theses weights, non-commercial use only.
## How do I use this model on an image?
To load a pretrained model:
```py
>>> import timm
>>> model = timm.create_model('ssl_resnet18', pretrained=True)
>>> model.eval()
```
To load and preprocess the image:
```py
>>> import urllib
>>> from PIL import Image
>>> from timm.data import resolve_data_config
>>> from timm.data.transforms_factory import create_transform
>>> config = resolve_data_config({}, model=model)
>>> transform = create_transform(**config)
>>> url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
>>> urllib.request.urlretrieve(url, filename)
>>> img = Image.open(filename).convert('RGB')
>>> tensor = transform(img).unsqueeze(0) # transform and add batch dimension
```
To get the model predictions:
```py
>>> import torch
>>> with torch.no_grad():
... out = model(tensor)
>>> probabilities = torch.nn.functional.softmax(out[0], dim=0)
>>> print(probabilities.shape)
>>> # prints: torch.Size([1000])
```
To get the top-5 predictions class names:
```py
>>> # Get imagenet class mappings
>>> url, filename = ("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt", "imagenet_classes.txt")
>>> urllib.request.urlretrieve(url, filename)
>>> with open("imagenet_classes.txt", "r") as f:
... categories = [s.strip() for s in f.readlines()]
>>> # Print top categories per image
>>> top5_prob, top5_catid = torch.topk(probabilities, 5)
>>> for i in range(top5_prob.size(0)):
... print(categories[top5_catid[i]], top5_prob[i].item())
>>> # prints class names and probabilities like:
>>> # [('Samoyed', 0.6425196528434753), ('Pomeranian', 0.04062102362513542), ('keeshond', 0.03186424449086189), ('white wolf', 0.01739676296710968), ('Eskimo dog', 0.011717947199940681)]
```
Replace the model name with the variant you want to use, e.g. `ssl_resnet18`. You can find the IDs in the model summaries at the top of this page.
To extract image features with this model, follow the [timm feature extraction examples](../feature_extraction), just change the name of the model you want to use.
## How do I finetune this model?
You can finetune any of the pre-trained models just by changing the classifier (the last layer).
```py
>>> model = timm.create_model('ssl_resnet18', pretrained=True, num_classes=NUM_FINETUNE_CLASSES)
```
To finetune on your own dataset, you have to write a training loop or adapt [timm's training
script](https://github.com/rwightman/pytorch-image-models/blob/master/train.py) to use your dataset.
## How do I train this model?
You can follow the [timm recipe scripts](../scripts) for training a new model afresh.
## Citation
```BibTeX
@article{DBLP:journals/corr/abs-1905-00546,
author = {I. Zeki Yalniz and
Herv{\'{e}} J{\'{e}}gou and
Kan Chen and
Manohar Paluri and
Dhruv Mahajan},
title = {Billion-scale semi-supervised learning for image classification},
journal = {CoRR},
volume = {abs/1905.00546},
year = {2019},
url = {http://arxiv.org/abs/1905.00546},
archivePrefix = {arXiv},
eprint = {1905.00546},
timestamp = {Mon, 28 Sep 2020 08:19:37 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-1905-00546.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```
<!--
Type: model-index
Collections:
- Name: SSL ResNet
Paper:
Title: Billion-scale semi-supervised learning for image classification
URL: https://paperswithcode.com/paper/billion-scale-semi-supervised-learning-for
Models:
- Name: ssl_resnet18
In Collection: SSL ResNet
Metadata:
FLOPs: 2337073152
Parameters: 11690000
File Size: 46811375
Architecture:
- 1x1 Convolution
- Batch Normalization
- Bottleneck Residual Block
- Convolution
- Global Average Pooling
- Max Pooling
- ReLU
- Residual Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- ImageNet
- YFCC-100M
Training Resources: 64x GPUs
ID: ssl_resnet18
LR: 0.0015
Epochs: 30
Layers: 18
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L894
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet18-d92f0530.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 72.62%
Top 5 Accuracy: 91.42%
- Name: ssl_resnet50
In Collection: SSL ResNet
Metadata:
FLOPs: 5282531328
Parameters: 25560000
File Size: 102480594
Architecture:
- 1x1 Convolution
- Batch Normalization
- Bottleneck Residual Block
- Convolution
- Global Average Pooling
- Max Pooling
- ReLU
- Residual Block
- Residual Connection
- Softmax
Tasks:
- Image Classification
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Data:
- ImageNet
- YFCC-100M
Training Resources: 64x GPUs
ID: ssl_resnet50
LR: 0.0015
Epochs: 30
Layers: 50
Crop Pct: '0.875'
Batch Size: 1536
Image Size: '224'
Weight Decay: 0.0001
Interpolation: bilinear
Code: https://github.com/rwightman/pytorch-image-models/blob/9a25fdf3ad0414b4d66da443fe60ae0aa14edc84/timm/models/resnet.py#L904
Weights: https://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet50-08389792.pth
Results:
- Task: Image Classification
Dataset: ImageNet
Metrics:
Top 1 Accuracy: 79.24%
Top 5 Accuracy: 94.83%
--> | pytorch-image-models/hfdocs/source/models/ssl-resnet.mdx/0 | {
"file_path": "pytorch-image-models/hfdocs/source/models/ssl-resnet.mdx",
"repo_id": "pytorch-image-models",
"token_count": 2425
} | 169 |
# Learning Rate Schedulers
This page contains the API reference documentation for learning rate schedulers included in `timm`.
## Schedulers
### Factory functions
[[autodoc]] timm.scheduler.scheduler_factory.create_scheduler
[[autodoc]] timm.scheduler.scheduler_factory.create_scheduler_v2
### Scheduler Classes
[[autodoc]] timm.scheduler.cosine_lr.CosineLRScheduler
[[autodoc]] timm.scheduler.multistep_lr.MultiStepLRScheduler
[[autodoc]] timm.scheduler.plateau_lr.PlateauLRScheduler
[[autodoc]] timm.scheduler.poly_lr.PolyLRScheduler
[[autodoc]] timm.scheduler.step_lr.StepLRScheduler
[[autodoc]] timm.scheduler.tanh_lr.TanhLRScheduler
| pytorch-image-models/hfdocs/source/reference/schedulers.mdx/0 | {
"file_path": "pytorch-image-models/hfdocs/source/reference/schedulers.mdx",
"repo_id": "pytorch-image-models",
"token_count": 242
} | 170 |
""" AutoAugment, RandAugment, AugMix, and 3-Augment for PyTorch
This code implements the searched ImageNet policies with various tweaks and improvements and
does not include any of the search code.
AA and RA Implementation adapted from:
https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/autoaugment.py
AugMix adapted from:
https://github.com/google-research/augmix
3-Augment based on: https://github.com/facebookresearch/deit/blob/main/README_revenge.md
Papers:
AutoAugment: Learning Augmentation Policies from Data - https://arxiv.org/abs/1805.09501
Learning Data Augmentation Strategies for Object Detection - https://arxiv.org/abs/1906.11172
RandAugment: Practical automated data augmentation... - https://arxiv.org/abs/1909.13719
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty - https://arxiv.org/abs/1912.02781
3-Augment: DeiT III: Revenge of the ViT - https://arxiv.org/abs/2204.07118
Hacked together by / Copyright 2019, Ross Wightman
"""
import random
import math
import re
from functools import partial
from typing import Dict, List, Optional, Union
from PIL import Image, ImageOps, ImageEnhance, ImageChops, ImageFilter
import PIL
import numpy as np
_PIL_VER = tuple([int(x) for x in PIL.__version__.split('.')[:2]])
_FILL = (128, 128, 128)
_LEVEL_DENOM = 10. # denominator for conversion from 'Mx' magnitude scale to fractional aug level for op arguments
_HPARAMS_DEFAULT = dict(
translate_const=250,
img_mean=_FILL,
)
if hasattr(Image, "Resampling"):
_RANDOM_INTERPOLATION = (Image.Resampling.BILINEAR, Image.Resampling.BICUBIC)
_DEFAULT_INTERPOLATION = Image.Resampling.BICUBIC
else:
_RANDOM_INTERPOLATION = (Image.BILINEAR, Image.BICUBIC)
_DEFAULT_INTERPOLATION = Image.BICUBIC
def _interpolation(kwargs):
interpolation = kwargs.pop('resample', _DEFAULT_INTERPOLATION)
if isinstance(interpolation, (list, tuple)):
return random.choice(interpolation)
return interpolation
def _check_args_tf(kwargs):
if 'fillcolor' in kwargs and _PIL_VER < (5, 0):
kwargs.pop('fillcolor')
kwargs['resample'] = _interpolation(kwargs)
def shear_x(img, factor, **kwargs):
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, factor, 0, 0, 1, 0), **kwargs)
def shear_y(img, factor, **kwargs):
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, 0, 0, factor, 1, 0), **kwargs)
def translate_x_rel(img, pct, **kwargs):
pixels = pct * img.size[0]
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, 0, pixels, 0, 1, 0), **kwargs)
def translate_y_rel(img, pct, **kwargs):
pixels = pct * img.size[1]
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, 0, 0, 0, 1, pixels), **kwargs)
def translate_x_abs(img, pixels, **kwargs):
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, 0, pixels, 0, 1, 0), **kwargs)
def translate_y_abs(img, pixels, **kwargs):
_check_args_tf(kwargs)
return img.transform(img.size, Image.AFFINE, (1, 0, 0, 0, 1, pixels), **kwargs)
def rotate(img, degrees, **kwargs):
_check_args_tf(kwargs)
if _PIL_VER >= (5, 2):
return img.rotate(degrees, **kwargs)
if _PIL_VER >= (5, 0):
w, h = img.size
post_trans = (0, 0)
rotn_center = (w / 2.0, h / 2.0)
angle = -math.radians(degrees)
matrix = [
round(math.cos(angle), 15),
round(math.sin(angle), 15),
0.0,
round(-math.sin(angle), 15),
round(math.cos(angle), 15),
0.0,
]
def transform(x, y, matrix):
(a, b, c, d, e, f) = matrix
return a * x + b * y + c, d * x + e * y + f
matrix[2], matrix[5] = transform(
-rotn_center[0] - post_trans[0], -rotn_center[1] - post_trans[1], matrix
)
matrix[2] += rotn_center[0]
matrix[5] += rotn_center[1]
return img.transform(img.size, Image.AFFINE, matrix, **kwargs)
return img.rotate(degrees, resample=kwargs['resample'])
def auto_contrast(img, **__):
return ImageOps.autocontrast(img)
def invert(img, **__):
return ImageOps.invert(img)
def equalize(img, **__):
return ImageOps.equalize(img)
def solarize(img, thresh, **__):
return ImageOps.solarize(img, thresh)
def solarize_add(img, add, thresh=128, **__):
lut = []
for i in range(256):
if i < thresh:
lut.append(min(255, i + add))
else:
lut.append(i)
if img.mode in ("L", "RGB"):
if img.mode == "RGB" and len(lut) == 256:
lut = lut + lut + lut
return img.point(lut)
return img
def posterize(img, bits_to_keep, **__):
if bits_to_keep >= 8:
return img
return ImageOps.posterize(img, bits_to_keep)
def contrast(img, factor, **__):
return ImageEnhance.Contrast(img).enhance(factor)
def color(img, factor, **__):
return ImageEnhance.Color(img).enhance(factor)
def brightness(img, factor, **__):
return ImageEnhance.Brightness(img).enhance(factor)
def sharpness(img, factor, **__):
return ImageEnhance.Sharpness(img).enhance(factor)
def gaussian_blur(img, factor, **__):
img = img.filter(ImageFilter.GaussianBlur(radius=factor))
return img
def gaussian_blur_rand(img, factor, **__):
radius_min = 0.1
radius_max = 2.0
img = img.filter(ImageFilter.GaussianBlur(radius=random.uniform(radius_min, radius_max * factor)))
return img
def desaturate(img, factor, **_):
factor = min(1., max(0., 1. - factor))
# enhance factor 0 = grayscale, 1.0 = no-change
return ImageEnhance.Color(img).enhance(factor)
def _randomly_negate(v):
"""With 50% prob, negate the value"""
return -v if random.random() > 0.5 else v
def _rotate_level_to_arg(level, _hparams):
# range [-30, 30]
level = (level / _LEVEL_DENOM) * 30.
level = _randomly_negate(level)
return level,
def _enhance_level_to_arg(level, _hparams):
# range [0.1, 1.9]
return (level / _LEVEL_DENOM) * 1.8 + 0.1,
def _enhance_increasing_level_to_arg(level, _hparams):
# the 'no change' level is 1.0, moving away from that towards 0. or 2.0 increases the enhancement blend
# range [0.1, 1.9] if level <= _LEVEL_DENOM
level = (level / _LEVEL_DENOM) * .9
level = max(0.1, 1.0 + _randomly_negate(level)) # keep it >= 0.1
return level,
def _minmax_level_to_arg(level, _hparams, min_val=0., max_val=1.0, clamp=True):
level = (level / _LEVEL_DENOM)
level = min_val + (max_val - min_val) * level
if clamp:
level = max(min_val, min(max_val, level))
return level,
def _shear_level_to_arg(level, _hparams):
# range [-0.3, 0.3]
level = (level / _LEVEL_DENOM) * 0.3
level = _randomly_negate(level)
return level,
def _translate_abs_level_to_arg(level, hparams):
translate_const = hparams['translate_const']
level = (level / _LEVEL_DENOM) * float(translate_const)
level = _randomly_negate(level)
return level,
def _translate_rel_level_to_arg(level, hparams):
# default range [-0.45, 0.45]
translate_pct = hparams.get('translate_pct', 0.45)
level = (level / _LEVEL_DENOM) * translate_pct
level = _randomly_negate(level)
return level,
def _posterize_level_to_arg(level, _hparams):
# As per Tensorflow TPU EfficientNet impl
# range [0, 4], 'keep 0 up to 4 MSB of original image'
# intensity/severity of augmentation decreases with level
return int((level / _LEVEL_DENOM) * 4),
def _posterize_increasing_level_to_arg(level, hparams):
# As per Tensorflow models research and UDA impl
# range [4, 0], 'keep 4 down to 0 MSB of original image',
# intensity/severity of augmentation increases with level
return 4 - _posterize_level_to_arg(level, hparams)[0],
def _posterize_original_level_to_arg(level, _hparams):
# As per original AutoAugment paper description
# range [4, 8], 'keep 4 up to 8 MSB of image'
# intensity/severity of augmentation decreases with level
return int((level / _LEVEL_DENOM) * 4) + 4,
def _solarize_level_to_arg(level, _hparams):
# range [0, 256]
# intensity/severity of augmentation decreases with level
return min(256, int((level / _LEVEL_DENOM) * 256)),
def _solarize_increasing_level_to_arg(level, _hparams):
# range [0, 256]
# intensity/severity of augmentation increases with level
return 256 - _solarize_level_to_arg(level, _hparams)[0],
def _solarize_add_level_to_arg(level, _hparams):
# range [0, 110]
return min(128, int((level / _LEVEL_DENOM) * 110)),
LEVEL_TO_ARG = {
'AutoContrast': None,
'Equalize': None,
'Invert': None,
'Rotate': _rotate_level_to_arg,
# There are several variations of the posterize level scaling in various Tensorflow/Google repositories/papers
'Posterize': _posterize_level_to_arg,
'PosterizeIncreasing': _posterize_increasing_level_to_arg,
'PosterizeOriginal': _posterize_original_level_to_arg,
'Solarize': _solarize_level_to_arg,
'SolarizeIncreasing': _solarize_increasing_level_to_arg,
'SolarizeAdd': _solarize_add_level_to_arg,
'Color': _enhance_level_to_arg,
'ColorIncreasing': _enhance_increasing_level_to_arg,
'Contrast': _enhance_level_to_arg,
'ContrastIncreasing': _enhance_increasing_level_to_arg,
'Brightness': _enhance_level_to_arg,
'BrightnessIncreasing': _enhance_increasing_level_to_arg,
'Sharpness': _enhance_level_to_arg,
'SharpnessIncreasing': _enhance_increasing_level_to_arg,
'ShearX': _shear_level_to_arg,
'ShearY': _shear_level_to_arg,
'TranslateX': _translate_abs_level_to_arg,
'TranslateY': _translate_abs_level_to_arg,
'TranslateXRel': _translate_rel_level_to_arg,
'TranslateYRel': _translate_rel_level_to_arg,
'Desaturate': partial(_minmax_level_to_arg, min_val=0.5, max_val=1.0),
'GaussianBlur': partial(_minmax_level_to_arg, min_val=0.1, max_val=2.0),
'GaussianBlurRand': _minmax_level_to_arg,
}
NAME_TO_OP = {
'AutoContrast': auto_contrast,
'Equalize': equalize,
'Invert': invert,
'Rotate': rotate,
'Posterize': posterize,
'PosterizeIncreasing': posterize,
'PosterizeOriginal': posterize,
'Solarize': solarize,
'SolarizeIncreasing': solarize,
'SolarizeAdd': solarize_add,
'Color': color,
'ColorIncreasing': color,
'Contrast': contrast,
'ContrastIncreasing': contrast,
'Brightness': brightness,
'BrightnessIncreasing': brightness,
'Sharpness': sharpness,
'SharpnessIncreasing': sharpness,
'ShearX': shear_x,
'ShearY': shear_y,
'TranslateX': translate_x_abs,
'TranslateY': translate_y_abs,
'TranslateXRel': translate_x_rel,
'TranslateYRel': translate_y_rel,
'Desaturate': desaturate,
'GaussianBlur': gaussian_blur,
'GaussianBlurRand': gaussian_blur_rand,
}
class AugmentOp:
def __init__(self, name, prob=0.5, magnitude=10, hparams=None):
hparams = hparams or _HPARAMS_DEFAULT
self.name = name
self.aug_fn = NAME_TO_OP[name]
self.level_fn = LEVEL_TO_ARG[name]
self.prob = prob
self.magnitude = magnitude
self.hparams = hparams.copy()
self.kwargs = dict(
fillcolor=hparams['img_mean'] if 'img_mean' in hparams else _FILL,
resample=hparams['interpolation'] if 'interpolation' in hparams else _RANDOM_INTERPOLATION,
)
# If magnitude_std is > 0, we introduce some randomness
# in the usually fixed policy and sample magnitude from a normal distribution
# with mean `magnitude` and std-dev of `magnitude_std`.
# NOTE This is my own hack, being tested, not in papers or reference impls.
# If magnitude_std is inf, we sample magnitude from a uniform distribution
self.magnitude_std = self.hparams.get('magnitude_std', 0)
self.magnitude_max = self.hparams.get('magnitude_max', None)
def __call__(self, img):
if self.prob < 1.0 and random.random() > self.prob:
return img
magnitude = self.magnitude
if self.magnitude_std > 0:
# magnitude randomization enabled
if self.magnitude_std == float('inf'):
# inf == uniform sampling
magnitude = random.uniform(0, magnitude)
elif self.magnitude_std > 0:
magnitude = random.gauss(magnitude, self.magnitude_std)
# default upper_bound for the timm RA impl is _LEVEL_DENOM (10)
# setting magnitude_max overrides this to allow M > 10 (behaviour closer to Google TF RA impl)
upper_bound = self.magnitude_max or _LEVEL_DENOM
magnitude = max(0., min(magnitude, upper_bound))
level_args = self.level_fn(magnitude, self.hparams) if self.level_fn is not None else tuple()
return self.aug_fn(img, *level_args, **self.kwargs)
def __repr__(self):
fs = self.__class__.__name__ + f'(name={self.name}, p={self.prob}'
fs += f', m={self.magnitude}, mstd={self.magnitude_std}'
if self.magnitude_max is not None:
fs += f', mmax={self.magnitude_max}'
fs += ')'
return fs
def auto_augment_policy_v0(hparams):
# ImageNet v0 policy from TPU EfficientNet impl, cannot find a paper reference.
policy = [
[('Equalize', 0.8, 1), ('ShearY', 0.8, 4)],
[('Color', 0.4, 9), ('Equalize', 0.6, 3)],
[('Color', 0.4, 1), ('Rotate', 0.6, 8)],
[('Solarize', 0.8, 3), ('Equalize', 0.4, 7)],
[('Solarize', 0.4, 2), ('Solarize', 0.6, 2)],
[('Color', 0.2, 0), ('Equalize', 0.8, 8)],
[('Equalize', 0.4, 8), ('SolarizeAdd', 0.8, 3)],
[('ShearX', 0.2, 9), ('Rotate', 0.6, 8)],
[('Color', 0.6, 1), ('Equalize', 1.0, 2)],
[('Invert', 0.4, 9), ('Rotate', 0.6, 0)],
[('Equalize', 1.0, 9), ('ShearY', 0.6, 3)],
[('Color', 0.4, 7), ('Equalize', 0.6, 0)],
[('Posterize', 0.4, 6), ('AutoContrast', 0.4, 7)],
[('Solarize', 0.6, 8), ('Color', 0.6, 9)],
[('Solarize', 0.2, 4), ('Rotate', 0.8, 9)],
[('Rotate', 1.0, 7), ('TranslateYRel', 0.8, 9)],
[('ShearX', 0.0, 0), ('Solarize', 0.8, 4)],
[('ShearY', 0.8, 0), ('Color', 0.6, 4)],
[('Color', 1.0, 0), ('Rotate', 0.6, 2)],
[('Equalize', 0.8, 4), ('Equalize', 0.0, 8)],
[('Equalize', 1.0, 4), ('AutoContrast', 0.6, 2)],
[('ShearY', 0.4, 7), ('SolarizeAdd', 0.6, 7)],
[('Posterize', 0.8, 2), ('Solarize', 0.6, 10)], # This results in black image with Tpu posterize
[('Solarize', 0.6, 8), ('Equalize', 0.6, 1)],
[('Color', 0.8, 6), ('Rotate', 0.4, 5)],
]
pc = [[AugmentOp(*a, hparams=hparams) for a in sp] for sp in policy]
return pc
def auto_augment_policy_v0r(hparams):
# ImageNet v0 policy from TPU EfficientNet impl, with variation of Posterize used
# in Google research implementation (number of bits discarded increases with magnitude)
policy = [
[('Equalize', 0.8, 1), ('ShearY', 0.8, 4)],
[('Color', 0.4, 9), ('Equalize', 0.6, 3)],
[('Color', 0.4, 1), ('Rotate', 0.6, 8)],
[('Solarize', 0.8, 3), ('Equalize', 0.4, 7)],
[('Solarize', 0.4, 2), ('Solarize', 0.6, 2)],
[('Color', 0.2, 0), ('Equalize', 0.8, 8)],
[('Equalize', 0.4, 8), ('SolarizeAdd', 0.8, 3)],
[('ShearX', 0.2, 9), ('Rotate', 0.6, 8)],
[('Color', 0.6, 1), ('Equalize', 1.0, 2)],
[('Invert', 0.4, 9), ('Rotate', 0.6, 0)],
[('Equalize', 1.0, 9), ('ShearY', 0.6, 3)],
[('Color', 0.4, 7), ('Equalize', 0.6, 0)],
[('PosterizeIncreasing', 0.4, 6), ('AutoContrast', 0.4, 7)],
[('Solarize', 0.6, 8), ('Color', 0.6, 9)],
[('Solarize', 0.2, 4), ('Rotate', 0.8, 9)],
[('Rotate', 1.0, 7), ('TranslateYRel', 0.8, 9)],
[('ShearX', 0.0, 0), ('Solarize', 0.8, 4)],
[('ShearY', 0.8, 0), ('Color', 0.6, 4)],
[('Color', 1.0, 0), ('Rotate', 0.6, 2)],
[('Equalize', 0.8, 4), ('Equalize', 0.0, 8)],
[('Equalize', 1.0, 4), ('AutoContrast', 0.6, 2)],
[('ShearY', 0.4, 7), ('SolarizeAdd', 0.6, 7)],
[('PosterizeIncreasing', 0.8, 2), ('Solarize', 0.6, 10)],
[('Solarize', 0.6, 8), ('Equalize', 0.6, 1)],
[('Color', 0.8, 6), ('Rotate', 0.4, 5)],
]
pc = [[AugmentOp(*a, hparams=hparams) for a in sp] for sp in policy]
return pc
def auto_augment_policy_original(hparams):
# ImageNet policy from https://arxiv.org/abs/1805.09501
policy = [
[('PosterizeOriginal', 0.4, 8), ('Rotate', 0.6, 9)],
[('Solarize', 0.6, 5), ('AutoContrast', 0.6, 5)],
[('Equalize', 0.8, 8), ('Equalize', 0.6, 3)],
[('PosterizeOriginal', 0.6, 7), ('PosterizeOriginal', 0.6, 6)],
[('Equalize', 0.4, 7), ('Solarize', 0.2, 4)],
[('Equalize', 0.4, 4), ('Rotate', 0.8, 8)],
[('Solarize', 0.6, 3), ('Equalize', 0.6, 7)],
[('PosterizeOriginal', 0.8, 5), ('Equalize', 1.0, 2)],
[('Rotate', 0.2, 3), ('Solarize', 0.6, 8)],
[('Equalize', 0.6, 8), ('PosterizeOriginal', 0.4, 6)],
[('Rotate', 0.8, 8), ('Color', 0.4, 0)],
[('Rotate', 0.4, 9), ('Equalize', 0.6, 2)],
[('Equalize', 0.0, 7), ('Equalize', 0.8, 8)],
[('Invert', 0.6, 4), ('Equalize', 1.0, 8)],
[('Color', 0.6, 4), ('Contrast', 1.0, 8)],
[('Rotate', 0.8, 8), ('Color', 1.0, 2)],
[('Color', 0.8, 8), ('Solarize', 0.8, 7)],
[('Sharpness', 0.4, 7), ('Invert', 0.6, 8)],
[('ShearX', 0.6, 5), ('Equalize', 1.0, 9)],
[('Color', 0.4, 0), ('Equalize', 0.6, 3)],
[('Equalize', 0.4, 7), ('Solarize', 0.2, 4)],
[('Solarize', 0.6, 5), ('AutoContrast', 0.6, 5)],
[('Invert', 0.6, 4), ('Equalize', 1.0, 8)],
[('Color', 0.6, 4), ('Contrast', 1.0, 8)],
[('Equalize', 0.8, 8), ('Equalize', 0.6, 3)],
]
pc = [[AugmentOp(*a, hparams=hparams) for a in sp] for sp in policy]
return pc
def auto_augment_policy_originalr(hparams):
# ImageNet policy from https://arxiv.org/abs/1805.09501 with research posterize variation
policy = [
[('PosterizeIncreasing', 0.4, 8), ('Rotate', 0.6, 9)],
[('Solarize', 0.6, 5), ('AutoContrast', 0.6, 5)],
[('Equalize', 0.8, 8), ('Equalize', 0.6, 3)],
[('PosterizeIncreasing', 0.6, 7), ('PosterizeIncreasing', 0.6, 6)],
[('Equalize', 0.4, 7), ('Solarize', 0.2, 4)],
[('Equalize', 0.4, 4), ('Rotate', 0.8, 8)],
[('Solarize', 0.6, 3), ('Equalize', 0.6, 7)],
[('PosterizeIncreasing', 0.8, 5), ('Equalize', 1.0, 2)],
[('Rotate', 0.2, 3), ('Solarize', 0.6, 8)],
[('Equalize', 0.6, 8), ('PosterizeIncreasing', 0.4, 6)],
[('Rotate', 0.8, 8), ('Color', 0.4, 0)],
[('Rotate', 0.4, 9), ('Equalize', 0.6, 2)],
[('Equalize', 0.0, 7), ('Equalize', 0.8, 8)],
[('Invert', 0.6, 4), ('Equalize', 1.0, 8)],
[('Color', 0.6, 4), ('Contrast', 1.0, 8)],
[('Rotate', 0.8, 8), ('Color', 1.0, 2)],
[('Color', 0.8, 8), ('Solarize', 0.8, 7)],
[('Sharpness', 0.4, 7), ('Invert', 0.6, 8)],
[('ShearX', 0.6, 5), ('Equalize', 1.0, 9)],
[('Color', 0.4, 0), ('Equalize', 0.6, 3)],
[('Equalize', 0.4, 7), ('Solarize', 0.2, 4)],
[('Solarize', 0.6, 5), ('AutoContrast', 0.6, 5)],
[('Invert', 0.6, 4), ('Equalize', 1.0, 8)],
[('Color', 0.6, 4), ('Contrast', 1.0, 8)],
[('Equalize', 0.8, 8), ('Equalize', 0.6, 3)],
]
pc = [[AugmentOp(*a, hparams=hparams) for a in sp] for sp in policy]
return pc
def auto_augment_policy_3a(hparams):
policy = [
[('Solarize', 1.0, 5)], # 128 solarize threshold @ 5 magnitude
[('Desaturate', 1.0, 10)], # grayscale at 10 magnitude
[('GaussianBlurRand', 1.0, 10)],
]
pc = [[AugmentOp(*a, hparams=hparams) for a in sp] for sp in policy]
return pc
def auto_augment_policy(name='v0', hparams=None):
hparams = hparams or _HPARAMS_DEFAULT
if name == 'original':
return auto_augment_policy_original(hparams)
if name == 'originalr':
return auto_augment_policy_originalr(hparams)
if name == 'v0':
return auto_augment_policy_v0(hparams)
if name == 'v0r':
return auto_augment_policy_v0r(hparams)
if name == '3a':
return auto_augment_policy_3a(hparams)
assert False, f'Unknown AA policy {name}'
class AutoAugment:
def __init__(self, policy):
self.policy = policy
def __call__(self, img):
sub_policy = random.choice(self.policy)
for op in sub_policy:
img = op(img)
return img
def __repr__(self):
fs = self.__class__.__name__ + '(policy='
for p in self.policy:
fs += '\n\t['
fs += ', '.join([str(op) for op in p])
fs += ']'
fs += ')'
return fs
def auto_augment_transform(config_str: str, hparams: Optional[Dict] = None):
"""
Create a AutoAugment transform
Args:
config_str: String defining configuration of auto augmentation. Consists of multiple sections separated by
dashes ('-').
The first section defines the AutoAugment policy (one of 'v0', 'v0r', 'original', 'originalr').
The remaining sections:
'mstd' - float std deviation of magnitude noise applied
Ex 'original-mstd0.5' results in AutoAugment with original policy, magnitude_std 0.5
hparams: Other hparams (kwargs) for the AutoAugmentation scheme
Returns:
A PyTorch compatible Transform
"""
config = config_str.split('-')
policy_name = config[0]
config = config[1:]
for c in config:
cs = re.split(r'(\d.*)', c)
if len(cs) < 2:
continue
key, val = cs[:2]
if key == 'mstd':
# noise param injected via hparams for now
hparams.setdefault('magnitude_std', float(val))
else:
assert False, 'Unknown AutoAugment config section'
aa_policy = auto_augment_policy(policy_name, hparams=hparams)
return AutoAugment(aa_policy)
_RAND_TRANSFORMS = [
'AutoContrast',
'Equalize',
'Invert',
'Rotate',
'Posterize',
'Solarize',
'SolarizeAdd',
'Color',
'Contrast',
'Brightness',
'Sharpness',
'ShearX',
'ShearY',
'TranslateXRel',
'TranslateYRel',
# 'Cutout' # NOTE I've implement this as random erasing separately
]
_RAND_INCREASING_TRANSFORMS = [
'AutoContrast',
'Equalize',
'Invert',
'Rotate',
'PosterizeIncreasing',
'SolarizeIncreasing',
'SolarizeAdd',
'ColorIncreasing',
'ContrastIncreasing',
'BrightnessIncreasing',
'SharpnessIncreasing',
'ShearX',
'ShearY',
'TranslateXRel',
'TranslateYRel',
# 'Cutout' # NOTE I've implement this as random erasing separately
]
_RAND_3A = [
'SolarizeIncreasing',
'Desaturate',
'GaussianBlur',
]
_RAND_WEIGHTED_3A = {
'SolarizeIncreasing': 6,
'Desaturate': 6,
'GaussianBlur': 6,
'Rotate': 3,
'ShearX': 2,
'ShearY': 2,
'PosterizeIncreasing': 1,
'AutoContrast': 1,
'ColorIncreasing': 1,
'SharpnessIncreasing': 1,
'ContrastIncreasing': 1,
'BrightnessIncreasing': 1,
'Equalize': 1,
'Invert': 1,
}
# These experimental weights are based loosely on the relative improvements mentioned in paper.
# They may not result in increased performance, but could likely be tuned to so.
_RAND_WEIGHTED_0 = {
'Rotate': 3,
'ShearX': 2,
'ShearY': 2,
'TranslateXRel': 1,
'TranslateYRel': 1,
'ColorIncreasing': .25,
'SharpnessIncreasing': 0.25,
'AutoContrast': 0.25,
'SolarizeIncreasing': .05,
'SolarizeAdd': .05,
'ContrastIncreasing': .05,
'BrightnessIncreasing': .05,
'Equalize': .05,
'PosterizeIncreasing': 0.05,
'Invert': 0.05,
}
def _get_weighted_transforms(transforms: Dict):
transforms, probs = list(zip(*transforms.items()))
probs = np.array(probs)
probs = probs / np.sum(probs)
return transforms, probs
def rand_augment_choices(name: str, increasing=True):
if name == 'weights':
return _RAND_WEIGHTED_0
if name == '3aw':
return _RAND_WEIGHTED_3A
if name == '3a':
return _RAND_3A
return _RAND_INCREASING_TRANSFORMS if increasing else _RAND_TRANSFORMS
def rand_augment_ops(
magnitude: Union[int, float] = 10,
prob: float = 0.5,
hparams: Optional[Dict] = None,
transforms: Optional[Union[Dict, List]] = None,
):
hparams = hparams or _HPARAMS_DEFAULT
transforms = transforms or _RAND_TRANSFORMS
return [AugmentOp(
name, prob=prob, magnitude=magnitude, hparams=hparams) for name in transforms]
class RandAugment:
def __init__(self, ops, num_layers=2, choice_weights=None):
self.ops = ops
self.num_layers = num_layers
self.choice_weights = choice_weights
def __call__(self, img):
# no replacement when using weighted choice
ops = np.random.choice(
self.ops,
self.num_layers,
replace=self.choice_weights is None,
p=self.choice_weights,
)
for op in ops:
img = op(img)
return img
def __repr__(self):
fs = self.__class__.__name__ + f'(n={self.num_layers}, ops='
for op in self.ops:
fs += f'\n\t{op}'
fs += ')'
return fs
def rand_augment_transform(
config_str: str,
hparams: Optional[Dict] = None,
transforms: Optional[Union[str, Dict, List]] = None,
):
"""
Create a RandAugment transform
Args:
config_str (str): String defining configuration of random augmentation. Consists of multiple sections separated
by dashes ('-'). The first section defines the specific variant of rand augment (currently only 'rand').
The remaining sections, not order sepecific determine
'm' - integer magnitude of rand augment
'n' - integer num layers (number of transform ops selected per image)
'p' - float probability of applying each layer (default 0.5)
'mstd' - float std deviation of magnitude noise applied, or uniform sampling if infinity (or > 100)
'mmax' - set upper bound for magnitude to something other than default of _LEVEL_DENOM (10)
'inc' - integer (bool), use augmentations that increase in severity with magnitude (default: 0)
't' - str name of transform set to use
Ex 'rand-m9-n3-mstd0.5' results in RandAugment with magnitude 9, num_layers 3, magnitude_std 0.5
'rand-mstd1-tweights' results in mag std 1.0, weighted transforms, default mag of 10 and num_layers 2
hparams (dict): Other hparams (kwargs) for the RandAugmentation scheme
Returns:
A PyTorch compatible Transform
"""
magnitude = _LEVEL_DENOM # default to _LEVEL_DENOM for magnitude (currently 10)
num_layers = 2 # default to 2 ops per image
increasing = False
prob = 0.5
config = config_str.split('-')
assert config[0] == 'rand'
config = config[1:]
for c in config:
if c.startswith('t'):
# NOTE old 'w' key was removed, 'w0' is not equivalent to 'tweights'
val = str(c[1:])
if transforms is None:
transforms = val
else:
# numeric options
cs = re.split(r'(\d.*)', c)
if len(cs) < 2:
continue
key, val = cs[:2]
if key == 'mstd':
# noise param / randomization of magnitude values
mstd = float(val)
if mstd > 100:
# use uniform sampling in 0 to magnitude if mstd is > 100
mstd = float('inf')
hparams.setdefault('magnitude_std', mstd)
elif key == 'mmax':
# clip magnitude between [0, mmax] instead of default [0, _LEVEL_DENOM]
hparams.setdefault('magnitude_max', int(val))
elif key == 'inc':
if bool(val):
increasing = True
elif key == 'm':
magnitude = int(val)
elif key == 'n':
num_layers = int(val)
elif key == 'p':
prob = float(val)
else:
assert False, 'Unknown RandAugment config section'
if isinstance(transforms, str):
transforms = rand_augment_choices(transforms, increasing=increasing)
elif transforms is None:
transforms = _RAND_INCREASING_TRANSFORMS if increasing else _RAND_TRANSFORMS
choice_weights = None
if isinstance(transforms, Dict):
transforms, choice_weights = _get_weighted_transforms(transforms)
ra_ops = rand_augment_ops(magnitude=magnitude, prob=prob, hparams=hparams, transforms=transforms)
return RandAugment(ra_ops, num_layers, choice_weights=choice_weights)
_AUGMIX_TRANSFORMS = [
'AutoContrast',
'ColorIncreasing', # not in paper
'ContrastIncreasing', # not in paper
'BrightnessIncreasing', # not in paper
'SharpnessIncreasing', # not in paper
'Equalize',
'Rotate',
'PosterizeIncreasing',
'SolarizeIncreasing',
'ShearX',
'ShearY',
'TranslateXRel',
'TranslateYRel',
]
def augmix_ops(
magnitude: Union[int, float] = 10,
hparams: Optional[Dict] = None,
transforms: Optional[Union[str, Dict, List]] = None,
):
hparams = hparams or _HPARAMS_DEFAULT
transforms = transforms or _AUGMIX_TRANSFORMS
return [AugmentOp(
name,
prob=1.0,
magnitude=magnitude,
hparams=hparams
) for name in transforms]
class AugMixAugment:
""" AugMix Transform
Adapted and improved from impl here: https://github.com/google-research/augmix/blob/master/imagenet.py
From paper: 'AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty -
https://arxiv.org/abs/1912.02781
"""
def __init__(self, ops, alpha=1., width=3, depth=-1, blended=False):
self.ops = ops
self.alpha = alpha
self.width = width
self.depth = depth
self.blended = blended # blended mode is faster but not well tested
def _calc_blended_weights(self, ws, m):
ws = ws * m
cump = 1.
rws = []
for w in ws[::-1]:
alpha = w / cump
cump *= (1 - alpha)
rws.append(alpha)
return np.array(rws[::-1], dtype=np.float32)
def _apply_blended(self, img, mixing_weights, m):
# This is my first crack and implementing a slightly faster mixed augmentation. Instead
# of accumulating the mix for each chain in a Numpy array and then blending with original,
# it recomputes the blending coefficients and applies one PIL image blend per chain.
# TODO the results appear in the right ballpark but they differ by more than rounding.
img_orig = img.copy()
ws = self._calc_blended_weights(mixing_weights, m)
for w in ws:
depth = self.depth if self.depth > 0 else np.random.randint(1, 4)
ops = np.random.choice(self.ops, depth, replace=True)
img_aug = img_orig # no ops are in-place, deep copy not necessary
for op in ops:
img_aug = op(img_aug)
img = Image.blend(img, img_aug, w)
return img
def _apply_basic(self, img, mixing_weights, m):
# This is a literal adaptation of the paper/official implementation without normalizations and
# PIL <-> Numpy conversions between every op. It is still quite CPU compute heavy compared to the
# typical augmentation transforms, could use a GPU / Kornia implementation.
img_shape = img.size[0], img.size[1], len(img.getbands())
mixed = np.zeros(img_shape, dtype=np.float32)
for mw in mixing_weights:
depth = self.depth if self.depth > 0 else np.random.randint(1, 4)
ops = np.random.choice(self.ops, depth, replace=True)
img_aug = img # no ops are in-place, deep copy not necessary
for op in ops:
img_aug = op(img_aug)
mixed += mw * np.asarray(img_aug, dtype=np.float32)
np.clip(mixed, 0, 255., out=mixed)
mixed = Image.fromarray(mixed.astype(np.uint8))
return Image.blend(img, mixed, m)
def __call__(self, img):
mixing_weights = np.float32(np.random.dirichlet([self.alpha] * self.width))
m = np.float32(np.random.beta(self.alpha, self.alpha))
if self.blended:
mixed = self._apply_blended(img, mixing_weights, m)
else:
mixed = self._apply_basic(img, mixing_weights, m)
return mixed
def __repr__(self):
fs = self.__class__.__name__ + f'(alpha={self.alpha}, width={self.width}, depth={self.depth}, ops='
for op in self.ops:
fs += f'\n\t{op}'
fs += ')'
return fs
def augment_and_mix_transform(config_str: str, hparams: Optional[Dict] = None):
""" Create AugMix PyTorch transform
Args:
config_str (str): String defining configuration of random augmentation. Consists of multiple sections separated
by dashes ('-'). The first section defines the specific variant of rand augment (currently only 'rand').
The remaining sections, not order sepecific determine
'm' - integer magnitude (severity) of augmentation mix (default: 3)
'w' - integer width of augmentation chain (default: 3)
'd' - integer depth of augmentation chain (-1 is random [1, 3], default: -1)
'b' - integer (bool), blend each branch of chain into end result without a final blend, less CPU (default: 0)
'mstd' - float std deviation of magnitude noise applied (default: 0)
Ex 'augmix-m5-w4-d2' results in AugMix with severity 5, chain width 4, chain depth 2
hparams: Other hparams (kwargs) for the Augmentation transforms
Returns:
A PyTorch compatible Transform
"""
magnitude = 3
width = 3
depth = -1
alpha = 1.
blended = False
config = config_str.split('-')
assert config[0] == 'augmix'
config = config[1:]
for c in config:
cs = re.split(r'(\d.*)', c)
if len(cs) < 2:
continue
key, val = cs[:2]
if key == 'mstd':
# noise param injected via hparams for now
hparams.setdefault('magnitude_std', float(val))
elif key == 'm':
magnitude = int(val)
elif key == 'w':
width = int(val)
elif key == 'd':
depth = int(val)
elif key == 'a':
alpha = float(val)
elif key == 'b':
blended = bool(val)
else:
assert False, 'Unknown AugMix config section'
hparams.setdefault('magnitude_std', float('inf')) # default to uniform sampling (if not set via mstd arg)
ops = augmix_ops(magnitude=magnitude, hparams=hparams)
return AugMixAugment(ops, alpha=alpha, width=width, depth=depth, blended=blended)
| pytorch-image-models/timm/data/auto_augment.py/0 | {
"file_path": "pytorch-image-models/timm/data/auto_augment.py",
"repo_id": "pytorch-image-models",
"token_count": 15929
} | 171 |
""" Dataset reader that wraps Hugging Face datasets
Hacked together by / Copyright 2022 Ross Wightman
"""
import io
import math
from typing import Optional
import torch
import torch.distributed as dist
from PIL import Image
try:
import datasets
except ImportError as e:
print("Please install Hugging Face datasets package `pip install datasets`.")
raise e
from .class_map import load_class_map
from .reader import Reader
def get_class_labels(info, label_key='label'):
if 'label' not in info.features:
return {}
class_label = info.features[label_key]
class_to_idx = {n: class_label.str2int(n) for n in class_label.names}
return class_to_idx
class ReaderHfds(Reader):
def __init__(
self,
name: str,
root: Optional[str] = None,
split: str = 'train',
class_map: dict = None,
image_key: str = 'image',
target_key: str = 'label',
download: bool = False,
):
"""
"""
super().__init__()
self.root = root
self.split = split
self.dataset = datasets.load_dataset(
name, # 'name' maps to path arg in hf datasets
split=split,
cache_dir=self.root, # timm doesn't expect hidden cache dir for datasets, specify a path
)
# leave decode for caller, plus we want easy access to original path names...
self.dataset = self.dataset.cast_column(image_key, datasets.Image(decode=False))
self.image_key = image_key
self.label_key = target_key
self.remap_class = False
if class_map:
self.class_to_idx = load_class_map(class_map)
self.remap_class = True
else:
self.class_to_idx = get_class_labels(self.dataset.info, self.label_key)
self.split_info = self.dataset.info.splits[split]
self.num_samples = self.split_info.num_examples
def __getitem__(self, index):
item = self.dataset[index]
image = item[self.image_key]
if 'bytes' in image and image['bytes']:
image = io.BytesIO(image['bytes'])
else:
assert 'path' in image and image['path']
image = open(image['path'], 'rb')
label = item[self.label_key]
if self.remap_class:
label = self.class_to_idx[label]
return image, label
def __len__(self):
return len(self.dataset)
def _filename(self, index, basename=False, absolute=False):
item = self.dataset[index]
return item[self.image_key]['path']
| pytorch-image-models/timm/data/readers/reader_hfds.py/0 | {
"file_path": "pytorch-image-models/timm/data/readers/reader_hfds.py",
"repo_id": "pytorch-image-models",
"token_count": 1150
} | 172 |
""" PyTorch selectable adaptive pooling
Adaptive pooling with the ability to select the type of pooling from:
* 'avg' - Average pooling
* 'max' - Max pooling
* 'avgmax' - Sum of average and max pooling re-scaled by 0.5
* 'avgmaxc' - Concatenation of average and max pooling along feature dim, doubles feature dim
Both a functional and a nn.Module version of the pooling is provided.
Hacked together by / Copyright 2020 Ross Wightman
"""
from typing import Optional, Tuple, Union
import torch
import torch.nn as nn
import torch.nn.functional as F
from .format import get_spatial_dim, get_channel_dim
_int_tuple_2_t = Union[int, Tuple[int, int]]
def adaptive_pool_feat_mult(pool_type='avg'):
if pool_type.endswith('catavgmax'):
return 2
else:
return 1
def adaptive_avgmax_pool2d(x, output_size: _int_tuple_2_t = 1):
x_avg = F.adaptive_avg_pool2d(x, output_size)
x_max = F.adaptive_max_pool2d(x, output_size)
return 0.5 * (x_avg + x_max)
def adaptive_catavgmax_pool2d(x, output_size: _int_tuple_2_t = 1):
x_avg = F.adaptive_avg_pool2d(x, output_size)
x_max = F.adaptive_max_pool2d(x, output_size)
return torch.cat((x_avg, x_max), 1)
def select_adaptive_pool2d(x, pool_type='avg', output_size: _int_tuple_2_t = 1):
"""Selectable global pooling function with dynamic input kernel size
"""
if pool_type == 'avg':
x = F.adaptive_avg_pool2d(x, output_size)
elif pool_type == 'avgmax':
x = adaptive_avgmax_pool2d(x, output_size)
elif pool_type == 'catavgmax':
x = adaptive_catavgmax_pool2d(x, output_size)
elif pool_type == 'max':
x = F.adaptive_max_pool2d(x, output_size)
else:
assert False, 'Invalid pool type: %s' % pool_type
return x
class FastAdaptiveAvgPool(nn.Module):
def __init__(self, flatten: bool = False, input_fmt: F = 'NCHW'):
super(FastAdaptiveAvgPool, self).__init__()
self.flatten = flatten
self.dim = get_spatial_dim(input_fmt)
def forward(self, x):
return x.mean(self.dim, keepdim=not self.flatten)
class FastAdaptiveMaxPool(nn.Module):
def __init__(self, flatten: bool = False, input_fmt: str = 'NCHW'):
super(FastAdaptiveMaxPool, self).__init__()
self.flatten = flatten
self.dim = get_spatial_dim(input_fmt)
def forward(self, x):
return x.amax(self.dim, keepdim=not self.flatten)
class FastAdaptiveAvgMaxPool(nn.Module):
def __init__(self, flatten: bool = False, input_fmt: str = 'NCHW'):
super(FastAdaptiveAvgMaxPool, self).__init__()
self.flatten = flatten
self.dim = get_spatial_dim(input_fmt)
def forward(self, x):
x_avg = x.mean(self.dim, keepdim=not self.flatten)
x_max = x.amax(self.dim, keepdim=not self.flatten)
return 0.5 * x_avg + 0.5 * x_max
class FastAdaptiveCatAvgMaxPool(nn.Module):
def __init__(self, flatten: bool = False, input_fmt: str = 'NCHW'):
super(FastAdaptiveCatAvgMaxPool, self).__init__()
self.flatten = flatten
self.dim_reduce = get_spatial_dim(input_fmt)
if flatten:
self.dim_cat = 1
else:
self.dim_cat = get_channel_dim(input_fmt)
def forward(self, x):
x_avg = x.mean(self.dim_reduce, keepdim=not self.flatten)
x_max = x.amax(self.dim_reduce, keepdim=not self.flatten)
return torch.cat((x_avg, x_max), self.dim_cat)
class AdaptiveAvgMaxPool2d(nn.Module):
def __init__(self, output_size: _int_tuple_2_t = 1):
super(AdaptiveAvgMaxPool2d, self).__init__()
self.output_size = output_size
def forward(self, x):
return adaptive_avgmax_pool2d(x, self.output_size)
class AdaptiveCatAvgMaxPool2d(nn.Module):
def __init__(self, output_size: _int_tuple_2_t = 1):
super(AdaptiveCatAvgMaxPool2d, self).__init__()
self.output_size = output_size
def forward(self, x):
return adaptive_catavgmax_pool2d(x, self.output_size)
class SelectAdaptivePool2d(nn.Module):
"""Selectable global pooling layer with dynamic input kernel size
"""
def __init__(
self,
output_size: _int_tuple_2_t = 1,
pool_type: str = 'fast',
flatten: bool = False,
input_fmt: str = 'NCHW',
):
super(SelectAdaptivePool2d, self).__init__()
assert input_fmt in ('NCHW', 'NHWC')
self.pool_type = pool_type or '' # convert other falsy values to empty string for consistent TS typing
if not pool_type:
self.pool = nn.Identity() # pass through
self.flatten = nn.Flatten(1) if flatten else nn.Identity()
elif pool_type.startswith('fast') or input_fmt != 'NCHW':
assert output_size == 1, 'Fast pooling and non NCHW input formats require output_size == 1.'
if pool_type.endswith('catavgmax'):
self.pool = FastAdaptiveCatAvgMaxPool(flatten, input_fmt=input_fmt)
elif pool_type.endswith('avgmax'):
self.pool = FastAdaptiveAvgMaxPool(flatten, input_fmt=input_fmt)
elif pool_type.endswith('max'):
self.pool = FastAdaptiveMaxPool(flatten, input_fmt=input_fmt)
else:
self.pool = FastAdaptiveAvgPool(flatten, input_fmt=input_fmt)
self.flatten = nn.Identity()
else:
assert input_fmt == 'NCHW'
if pool_type == 'avgmax':
self.pool = AdaptiveAvgMaxPool2d(output_size)
elif pool_type == 'catavgmax':
self.pool = AdaptiveCatAvgMaxPool2d(output_size)
elif pool_type == 'max':
self.pool = nn.AdaptiveMaxPool2d(output_size)
else:
self.pool = nn.AdaptiveAvgPool2d(output_size)
self.flatten = nn.Flatten(1) if flatten else nn.Identity()
def is_identity(self):
return not self.pool_type
def forward(self, x):
x = self.pool(x)
x = self.flatten(x)
return x
def feat_mult(self):
return adaptive_pool_feat_mult(self.pool_type)
def __repr__(self):
return self.__class__.__name__ + '(' \
+ 'pool_type=' + self.pool_type \
+ ', flatten=' + str(self.flatten) + ')'
| pytorch-image-models/timm/layers/adaptive_avgmax_pool.py/0 | {
"file_path": "pytorch-image-models/timm/layers/adaptive_avgmax_pool.py",
"repo_id": "pytorch-image-models",
"token_count": 2903
} | 173 |
""" DropBlock, DropPath
PyTorch implementations of DropBlock and DropPath (Stochastic Depth) regularization layers.
Papers:
DropBlock: A regularization method for convolutional networks (https://arxiv.org/abs/1810.12890)
Deep Networks with Stochastic Depth (https://arxiv.org/abs/1603.09382)
Code:
DropBlock impl inspired by two Tensorflow impl that I liked:
- https://github.com/tensorflow/tpu/blob/master/models/official/resnet/resnet_model.py#L74
- https://github.com/clovaai/assembled-cnn/blob/master/nets/blocks.py
Hacked together by / Copyright 2020 Ross Wightman
"""
import torch
import torch.nn as nn
import torch.nn.functional as F
from .grid import ndgrid
def drop_block_2d(
x,
drop_prob: float = 0.1,
block_size: int = 7,
gamma_scale: float = 1.0,
with_noise: bool = False,
inplace: bool = False,
batchwise: bool = False
):
""" DropBlock. See https://arxiv.org/pdf/1810.12890.pdf
DropBlock with an experimental gaussian noise option. This layer has been tested on a few training
runs with success, but needs further validation and possibly optimization for lower runtime impact.
"""
B, C, H, W = x.shape
total_size = W * H
clipped_block_size = min(block_size, min(W, H))
# seed_drop_rate, the gamma parameter
gamma = gamma_scale * drop_prob * total_size / clipped_block_size ** 2 / (
(W - block_size + 1) * (H - block_size + 1))
# Forces the block to be inside the feature map.
w_i, h_i = ndgrid(torch.arange(W, device=x.device), torch.arange(H, device=x.device))
valid_block = ((w_i >= clipped_block_size // 2) & (w_i < W - (clipped_block_size - 1) // 2)) & \
((h_i >= clipped_block_size // 2) & (h_i < H - (clipped_block_size - 1) // 2))
valid_block = torch.reshape(valid_block, (1, 1, H, W)).to(dtype=x.dtype)
if batchwise:
# one mask for whole batch, quite a bit faster
uniform_noise = torch.rand((1, C, H, W), dtype=x.dtype, device=x.device)
else:
uniform_noise = torch.rand_like(x)
block_mask = ((2 - gamma - valid_block + uniform_noise) >= 1).to(dtype=x.dtype)
block_mask = -F.max_pool2d(
-block_mask,
kernel_size=clipped_block_size, # block_size,
stride=1,
padding=clipped_block_size // 2)
if with_noise:
normal_noise = torch.randn((1, C, H, W), dtype=x.dtype, device=x.device) if batchwise else torch.randn_like(x)
if inplace:
x.mul_(block_mask).add_(normal_noise * (1 - block_mask))
else:
x = x * block_mask + normal_noise * (1 - block_mask)
else:
normalize_scale = (block_mask.numel() / block_mask.to(dtype=torch.float32).sum().add(1e-7)).to(x.dtype)
if inplace:
x.mul_(block_mask * normalize_scale)
else:
x = x * block_mask * normalize_scale
return x
def drop_block_fast_2d(
x: torch.Tensor,
drop_prob: float = 0.1,
block_size: int = 7,
gamma_scale: float = 1.0,
with_noise: bool = False,
inplace: bool = False,
):
""" DropBlock. See https://arxiv.org/pdf/1810.12890.pdf
DropBlock with an experimental gaussian noise option. Simplied from above without concern for valid
block mask at edges.
"""
B, C, H, W = x.shape
total_size = W * H
clipped_block_size = min(block_size, min(W, H))
gamma = gamma_scale * drop_prob * total_size / clipped_block_size ** 2 / (
(W - block_size + 1) * (H - block_size + 1))
block_mask = torch.empty_like(x).bernoulli_(gamma)
block_mask = F.max_pool2d(
block_mask.to(x.dtype), kernel_size=clipped_block_size, stride=1, padding=clipped_block_size // 2)
if with_noise:
normal_noise = torch.empty_like(x).normal_()
if inplace:
x.mul_(1. - block_mask).add_(normal_noise * block_mask)
else:
x = x * (1. - block_mask) + normal_noise * block_mask
else:
block_mask = 1 - block_mask
normalize_scale = (block_mask.numel() / block_mask.to(dtype=torch.float32).sum().add(1e-6)).to(dtype=x.dtype)
if inplace:
x.mul_(block_mask * normalize_scale)
else:
x = x * block_mask * normalize_scale
return x
class DropBlock2d(nn.Module):
""" DropBlock. See https://arxiv.org/pdf/1810.12890.pdf
"""
def __init__(
self,
drop_prob: float = 0.1,
block_size: int = 7,
gamma_scale: float = 1.0,
with_noise: bool = False,
inplace: bool = False,
batchwise: bool = False,
fast: bool = True):
super(DropBlock2d, self).__init__()
self.drop_prob = drop_prob
self.gamma_scale = gamma_scale
self.block_size = block_size
self.with_noise = with_noise
self.inplace = inplace
self.batchwise = batchwise
self.fast = fast # FIXME finish comparisons of fast vs not
def forward(self, x):
if not self.training or not self.drop_prob:
return x
if self.fast:
return drop_block_fast_2d(
x, self.drop_prob, self.block_size, self.gamma_scale, self.with_noise, self.inplace)
else:
return drop_block_2d(
x, self.drop_prob, self.block_size, self.gamma_scale, self.with_noise, self.inplace, self.batchwise)
def drop_path(x, drop_prob: float = 0., training: bool = False, scale_by_keep: bool = True):
"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).
This is the same as the DropConnect impl I created for EfficientNet, etc networks, however,
the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper...
See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 ... I've opted for
changing the layer and argument names to 'drop path' rather than mix DropConnect as a layer name and use
'survival rate' as the argument.
"""
if drop_prob == 0. or not training:
return x
keep_prob = 1 - drop_prob
shape = (x.shape[0],) + (1,) * (x.ndim - 1) # work with diff dim tensors, not just 2D ConvNets
random_tensor = x.new_empty(shape).bernoulli_(keep_prob)
if keep_prob > 0.0 and scale_by_keep:
random_tensor.div_(keep_prob)
return x * random_tensor
class DropPath(nn.Module):
"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).
"""
def __init__(self, drop_prob: float = 0., scale_by_keep: bool = True):
super(DropPath, self).__init__()
self.drop_prob = drop_prob
self.scale_by_keep = scale_by_keep
def forward(self, x):
return drop_path(x, self.drop_prob, self.training, self.scale_by_keep)
def extra_repr(self):
return f'drop_prob={round(self.drop_prob,3):0.3f}'
| pytorch-image-models/timm/layers/drop.py/0 | {
"file_path": "pytorch-image-models/timm/layers/drop.py",
"repo_id": "pytorch-image-models",
"token_count": 3016
} | 174 |
""" Median Pool
Hacked together by / Copyright 2020 Ross Wightman
"""
import torch.nn as nn
import torch.nn.functional as F
from .helpers import to_2tuple, to_4tuple
class MedianPool2d(nn.Module):
""" Median pool (usable as median filter when stride=1) module.
Args:
kernel_size: size of pooling kernel, int or 2-tuple
stride: pool stride, int or 2-tuple
padding: pool padding, int or 4-tuple (l, r, t, b) as in pytorch F.pad
same: override padding and enforce same padding, boolean
"""
def __init__(self, kernel_size=3, stride=1, padding=0, same=False):
super(MedianPool2d, self).__init__()
self.k = to_2tuple(kernel_size)
self.stride = to_2tuple(stride)
self.padding = to_4tuple(padding) # convert to l, r, t, b
self.same = same
def _padding(self, x):
if self.same:
ih, iw = x.size()[2:]
if ih % self.stride[0] == 0:
ph = max(self.k[0] - self.stride[0], 0)
else:
ph = max(self.k[0] - (ih % self.stride[0]), 0)
if iw % self.stride[1] == 0:
pw = max(self.k[1] - self.stride[1], 0)
else:
pw = max(self.k[1] - (iw % self.stride[1]), 0)
pl = pw // 2
pr = pw - pl
pt = ph // 2
pb = ph - pt
padding = (pl, pr, pt, pb)
else:
padding = self.padding
return padding
def forward(self, x):
x = F.pad(x, self._padding(x), mode='reflect')
x = x.unfold(2, self.k[0], self.stride[0]).unfold(3, self.k[1], self.stride[1])
x = x.contiguous().view(x.size()[:4] + (-1,)).median(dim=-1)[0]
return x
| pytorch-image-models/timm/layers/median_pool.py/0 | {
"file_path": "pytorch-image-models/timm/layers/median_pool.py",
"repo_id": "pytorch-image-models",
"token_count": 883
} | 175 |
import torch
import torch.nn as nn
class SpaceToDepth(nn.Module):
bs: torch.jit.Final[int]
def __init__(self, block_size=4):
super().__init__()
assert block_size == 4
self.bs = block_size
def forward(self, x):
N, C, H, W = x.size()
x = x.view(N, C, H // self.bs, self.bs, W // self.bs, self.bs) # (N, C, H//bs, bs, W//bs, bs)
x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # (N, bs, bs, C, H//bs, W//bs)
x = x.view(N, C * self.bs * self.bs, H // self.bs, W // self.bs) # (N, C*bs^2, H//bs, W//bs)
return x
@torch.jit.script
class SpaceToDepthJit:
def __call__(self, x: torch.Tensor):
# assuming hard-coded that block_size==4 for acceleration
N, C, H, W = x.size()
x = x.view(N, C, H // 4, 4, W // 4, 4) # (N, C, H//bs, bs, W//bs, bs)
x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # (N, bs, bs, C, H//bs, W//bs)
x = x.view(N, C * 16, H // 4, W // 4) # (N, C*bs^2, H//bs, W//bs)
return x
class SpaceToDepthModule(nn.Module):
def __init__(self, no_jit=False):
super().__init__()
if not no_jit:
self.op = SpaceToDepthJit()
else:
self.op = SpaceToDepth()
def forward(self, x):
return self.op(x)
class DepthToSpace(nn.Module):
def __init__(self, block_size):
super().__init__()
self.bs = block_size
def forward(self, x):
N, C, H, W = x.size()
x = x.view(N, self.bs, self.bs, C // (self.bs ** 2), H, W) # (N, bs, bs, C//bs^2, H, W)
x = x.permute(0, 3, 4, 1, 5, 2).contiguous() # (N, C//bs^2, H, bs, W, bs)
x = x.view(N, C // (self.bs ** 2), H * self.bs, W * self.bs) # (N, C//bs^2, H * bs, W * bs)
return x
| pytorch-image-models/timm/layers/space_to_depth.py/0 | {
"file_path": "pytorch-image-models/timm/layers/space_to_depth.py",
"repo_id": "pytorch-image-models",
"token_count": 938
} | 176 |
""" EfficientNet, MobileNetV3, etc Blocks
Hacked together by / Copyright 2019, Ross Wightman
"""
import torch
import torch.nn as nn
from torch.nn import functional as F
from timm.layers import create_conv2d, DropPath, make_divisible, create_act_layer, get_norm_act_layer
__all__ = [
'SqueezeExcite', 'ConvBnAct', 'DepthwiseSeparableConv', 'InvertedResidual', 'CondConvResidual', 'EdgeResidual']
def num_groups(group_size, channels):
if not group_size: # 0 or None
return 1 # normal conv with 1 group
else:
# NOTE group_size == 1 -> depthwise conv
assert channels % group_size == 0
return channels // group_size
class SqueezeExcite(nn.Module):
""" Squeeze-and-Excitation w/ specific features for EfficientNet/MobileNet family
Args:
in_chs (int): input channels to layer
rd_ratio (float): ratio of squeeze reduction
act_layer (nn.Module): activation layer of containing block
gate_layer (Callable): attention gate function
force_act_layer (nn.Module): override block's activation fn if this is set/bound
rd_round_fn (Callable): specify a fn to calculate rounding of reduced chs
"""
def __init__(
self, in_chs, rd_ratio=0.25, rd_channels=None, act_layer=nn.ReLU,
gate_layer=nn.Sigmoid, force_act_layer=None, rd_round_fn=None):
super(SqueezeExcite, self).__init__()
if rd_channels is None:
rd_round_fn = rd_round_fn or round
rd_channels = rd_round_fn(in_chs * rd_ratio)
act_layer = force_act_layer or act_layer
self.conv_reduce = nn.Conv2d(in_chs, rd_channels, 1, bias=True)
self.act1 = create_act_layer(act_layer, inplace=True)
self.conv_expand = nn.Conv2d(rd_channels, in_chs, 1, bias=True)
self.gate = create_act_layer(gate_layer)
def forward(self, x):
x_se = x.mean((2, 3), keepdim=True)
x_se = self.conv_reduce(x_se)
x_se = self.act1(x_se)
x_se = self.conv_expand(x_se)
return x * self.gate(x_se)
class ConvBnAct(nn.Module):
""" Conv + Norm Layer + Activation w/ optional skip connection
"""
def __init__(
self, in_chs, out_chs, kernel_size, stride=1, dilation=1, group_size=0, pad_type='',
skip=False, act_layer=nn.ReLU, norm_layer=nn.BatchNorm2d, drop_path_rate=0.):
super(ConvBnAct, self).__init__()
norm_act_layer = get_norm_act_layer(norm_layer, act_layer)
groups = num_groups(group_size, in_chs)
self.has_skip = skip and stride == 1 and in_chs == out_chs
self.conv = create_conv2d(
in_chs, out_chs, kernel_size, stride=stride, dilation=dilation, groups=groups, padding=pad_type)
self.bn1 = norm_act_layer(out_chs, inplace=True)
self.drop_path = DropPath(drop_path_rate) if drop_path_rate else nn.Identity()
def feature_info(self, location):
if location == 'expansion': # output of conv after act, same as block coutput
return dict(module='bn1', hook_type='forward', num_chs=self.conv.out_channels)
else: # location == 'bottleneck', block output
return dict(module='', num_chs=self.conv.out_channels)
def forward(self, x):
shortcut = x
x = self.conv(x)
x = self.bn1(x)
if self.has_skip:
x = self.drop_path(x) + shortcut
return x
class DepthwiseSeparableConv(nn.Module):
""" DepthwiseSeparable block
Used for DS convs in MobileNet-V1 and in the place of IR blocks that have no expansion
(factor of 1.0). This is an alternative to having a IR with an optional first pw conv.
"""
def __init__(
self, in_chs, out_chs, dw_kernel_size=3, stride=1, dilation=1, group_size=1, pad_type='',
noskip=False, pw_kernel_size=1, pw_act=False, act_layer=nn.ReLU, norm_layer=nn.BatchNorm2d,
se_layer=None, drop_path_rate=0.):
super(DepthwiseSeparableConv, self).__init__()
norm_act_layer = get_norm_act_layer(norm_layer, act_layer)
groups = num_groups(group_size, in_chs)
self.has_skip = (stride == 1 and in_chs == out_chs) and not noskip
self.has_pw_act = pw_act # activation after point-wise conv
self.conv_dw = create_conv2d(
in_chs, in_chs, dw_kernel_size, stride=stride, dilation=dilation, padding=pad_type, groups=groups)
self.bn1 = norm_act_layer(in_chs, inplace=True)
# Squeeze-and-excitation
self.se = se_layer(in_chs, act_layer=act_layer) if se_layer else nn.Identity()
self.conv_pw = create_conv2d(in_chs, out_chs, pw_kernel_size, padding=pad_type)
self.bn2 = norm_act_layer(out_chs, inplace=True, apply_act=self.has_pw_act)
self.drop_path = DropPath(drop_path_rate) if drop_path_rate else nn.Identity()
def feature_info(self, location):
if location == 'expansion': # after SE, input to PW
return dict(module='conv_pw', hook_type='forward_pre', num_chs=self.conv_pw.in_channels)
else: # location == 'bottleneck', block output
return dict(module='', num_chs=self.conv_pw.out_channels)
def forward(self, x):
shortcut = x
x = self.conv_dw(x)
x = self.bn1(x)
x = self.se(x)
x = self.conv_pw(x)
x = self.bn2(x)
if self.has_skip:
x = self.drop_path(x) + shortcut
return x
class InvertedResidual(nn.Module):
""" Inverted residual block w/ optional SE
Originally used in MobileNet-V2 - https://arxiv.org/abs/1801.04381v4, this layer is often
referred to as 'MBConv' for (Mobile inverted bottleneck conv) and is also used in
* MNasNet - https://arxiv.org/abs/1807.11626
* EfficientNet - https://arxiv.org/abs/1905.11946
* MobileNet-V3 - https://arxiv.org/abs/1905.02244
"""
def __init__(
self, in_chs, out_chs, dw_kernel_size=3, stride=1, dilation=1, group_size=1, pad_type='',
noskip=False, exp_ratio=1.0, exp_kernel_size=1, pw_kernel_size=1, act_layer=nn.ReLU,
norm_layer=nn.BatchNorm2d, se_layer=None, conv_kwargs=None, drop_path_rate=0.):
super(InvertedResidual, self).__init__()
norm_act_layer = get_norm_act_layer(norm_layer, act_layer)
conv_kwargs = conv_kwargs or {}
mid_chs = make_divisible(in_chs * exp_ratio)
groups = num_groups(group_size, mid_chs)
self.has_skip = (in_chs == out_chs and stride == 1) and not noskip
# Point-wise expansion
self.conv_pw = create_conv2d(in_chs, mid_chs, exp_kernel_size, padding=pad_type, **conv_kwargs)
self.bn1 = norm_act_layer(mid_chs, inplace=True)
# Depth-wise convolution
self.conv_dw = create_conv2d(
mid_chs, mid_chs, dw_kernel_size, stride=stride, dilation=dilation,
groups=groups, padding=pad_type, **conv_kwargs)
self.bn2 = norm_act_layer(mid_chs, inplace=True)
# Squeeze-and-excitation
self.se = se_layer(mid_chs, act_layer=act_layer) if se_layer else nn.Identity()
# Point-wise linear projection
self.conv_pwl = create_conv2d(mid_chs, out_chs, pw_kernel_size, padding=pad_type, **conv_kwargs)
self.bn3 = norm_act_layer(out_chs, apply_act=False)
self.drop_path = DropPath(drop_path_rate) if drop_path_rate else nn.Identity()
def feature_info(self, location):
if location == 'expansion': # after SE, input to PWL
return dict(module='conv_pwl', hook_type='forward_pre', num_chs=self.conv_pwl.in_channels)
else: # location == 'bottleneck', block output
return dict(module='', num_chs=self.conv_pwl.out_channels)
def forward(self, x):
shortcut = x
x = self.conv_pw(x)
x = self.bn1(x)
x = self.conv_dw(x)
x = self.bn2(x)
x = self.se(x)
x = self.conv_pwl(x)
x = self.bn3(x)
if self.has_skip:
x = self.drop_path(x) + shortcut
return x
class CondConvResidual(InvertedResidual):
""" Inverted residual block w/ CondConv routing"""
def __init__(
self, in_chs, out_chs, dw_kernel_size=3, stride=1, dilation=1, group_size=1, pad_type='',
noskip=False, exp_ratio=1.0, exp_kernel_size=1, pw_kernel_size=1, act_layer=nn.ReLU,
norm_layer=nn.BatchNorm2d, se_layer=None, num_experts=0, drop_path_rate=0.):
self.num_experts = num_experts
conv_kwargs = dict(num_experts=self.num_experts)
super(CondConvResidual, self).__init__(
in_chs, out_chs, dw_kernel_size=dw_kernel_size, stride=stride, dilation=dilation, group_size=group_size,
pad_type=pad_type, act_layer=act_layer, noskip=noskip, exp_ratio=exp_ratio, exp_kernel_size=exp_kernel_size,
pw_kernel_size=pw_kernel_size, se_layer=se_layer, norm_layer=norm_layer, conv_kwargs=conv_kwargs,
drop_path_rate=drop_path_rate)
self.routing_fn = nn.Linear(in_chs, self.num_experts)
def forward(self, x):
shortcut = x
pooled_inputs = F.adaptive_avg_pool2d(x, 1).flatten(1) # CondConv routing
routing_weights = torch.sigmoid(self.routing_fn(pooled_inputs))
x = self.conv_pw(x, routing_weights)
x = self.bn1(x)
x = self.conv_dw(x, routing_weights)
x = self.bn2(x)
x = self.se(x)
x = self.conv_pwl(x, routing_weights)
x = self.bn3(x)
if self.has_skip:
x = self.drop_path(x) + shortcut
return x
class EdgeResidual(nn.Module):
""" Residual block with expansion convolution followed by pointwise-linear w/ stride
Originally introduced in `EfficientNet-EdgeTPU: Creating Accelerator-Optimized Neural Networks with AutoML`
- https://ai.googleblog.com/2019/08/efficientnet-edgetpu-creating.html
This layer is also called FusedMBConv in the MobileDet, EfficientNet-X, and EfficientNet-V2 papers
* MobileDet - https://arxiv.org/abs/2004.14525
* EfficientNet-X - https://arxiv.org/abs/2102.05610
* EfficientNet-V2 - https://arxiv.org/abs/2104.00298
"""
def __init__(
self, in_chs, out_chs, exp_kernel_size=3, stride=1, dilation=1, group_size=0, pad_type='',
force_in_chs=0, noskip=False, exp_ratio=1.0, pw_kernel_size=1, act_layer=nn.ReLU,
norm_layer=nn.BatchNorm2d, se_layer=None, drop_path_rate=0.):
super(EdgeResidual, self).__init__()
norm_act_layer = get_norm_act_layer(norm_layer, act_layer)
if force_in_chs > 0:
mid_chs = make_divisible(force_in_chs * exp_ratio)
else:
mid_chs = make_divisible(in_chs * exp_ratio)
groups = num_groups(group_size, in_chs)
self.has_skip = (in_chs == out_chs and stride == 1) and not noskip
# Expansion convolution
self.conv_exp = create_conv2d(
in_chs, mid_chs, exp_kernel_size, stride=stride, dilation=dilation, groups=groups, padding=pad_type)
self.bn1 = norm_act_layer(mid_chs, inplace=True)
# Squeeze-and-excitation
self.se = se_layer(mid_chs, act_layer=act_layer) if se_layer else nn.Identity()
# Point-wise linear projection
self.conv_pwl = create_conv2d(mid_chs, out_chs, pw_kernel_size, padding=pad_type)
self.bn2 = norm_act_layer(out_chs, apply_act=False)
self.drop_path = DropPath(drop_path_rate) if drop_path_rate else nn.Identity()
def feature_info(self, location):
if location == 'expansion': # after SE, before PWL
return dict(module='conv_pwl', hook_type='forward_pre', num_chs=self.conv_pwl.in_channels)
else: # location == 'bottleneck', block output
return dict(module='', num_chs=self.conv_pwl.out_channels)
def forward(self, x):
shortcut = x
x = self.conv_exp(x)
x = self.bn1(x)
x = self.se(x)
x = self.conv_pwl(x)
x = self.bn2(x)
if self.has_skip:
x = self.drop_path(x) + shortcut
return x
| pytorch-image-models/timm/models/_efficientnet_blocks.py/0 | {
"file_path": "pytorch-image-models/timm/models/_efficientnet_blocks.py",
"repo_id": "pytorch-image-models",
"token_count": 5589
} | 177 |
""" BEiT: BERT Pre-Training of Image Transformers (https://arxiv.org/abs/2106.08254)
Model from official source: https://github.com/microsoft/unilm/tree/master/beit
@inproceedings{beit,
title={{BEiT}: {BERT} Pre-Training of Image Transformers},
author={Hangbo Bao and Li Dong and Songhao Piao and Furu Wei},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=p-BhZSz59o4}
}
BEiT-v2 from https://github.com/microsoft/unilm/tree/master/beit2
@article{beitv2,
title={{BEiT v2}: Masked Image Modeling with Vector-Quantized Visual Tokenizers},
author={Zhiliang Peng and Li Dong and Hangbo Bao and Qixiang Ye and Furu Wei},
year={2022},
eprint={2208.06366},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
At this point only the 1k fine-tuned classification weights and model configs have been added,
see original source above for pre-training models and procedure.
Modifications by / Copyright 2021 Ross Wightman, original copyrights below
"""
# --------------------------------------------------------
# BEIT: BERT Pre-Training of Image Transformers (https://arxiv.org/abs/2106.08254)
# Github source: https://github.com/microsoft/unilm/tree/master/beit
# Copyright (c) 2021 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# By Hangbo Bao
# Based on timm and DeiT code bases
# https://github.com/rwightman/pytorch-image-models/tree/master/timm
# https://github.com/facebookresearch/deit/
# https://github.com/facebookresearch/dino
# --------------------------------------------------------'
import math
from typing import Callable, Optional, Tuple, Union
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.utils.checkpoint import checkpoint
from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.layers import PatchEmbed, Mlp, SwiGLU, LayerNorm, DropPath, trunc_normal_, use_fused_attn
from timm.layers import resample_patch_embed, resample_abs_pos_embed, resize_rel_pos_bias_table, ndgrid
from ._builder import build_model_with_cfg
from ._registry import generate_default_cfgs, register_model
from .vision_transformer import checkpoint_filter_fn
__all__ = ['Beit']
def gen_relative_position_index(window_size: Tuple[int, int]) -> torch.Tensor:
num_relative_distance = (2 * window_size[0] - 1) * (2 * window_size[1] - 1) + 3
# cls to token & token 2 cls & cls to cls
# get pair-wise relative position index for each token inside the window
window_area = window_size[0] * window_size[1]
coords = torch.stack(ndgrid(torch.arange(window_size[0]), torch.arange(window_size[1]))) # 2, Wh, Ww
coords_flatten = torch.flatten(coords, 1) # 2, Wh*Ww
relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :] # 2, Wh*Ww, Wh*Ww
relative_coords = relative_coords.permute(1, 2, 0).contiguous() # Wh*Ww, Wh*Ww, 2
relative_coords[:, :, 0] += window_size[0] - 1 # shift to start from 0
relative_coords[:, :, 1] += window_size[1] - 1
relative_coords[:, :, 0] *= 2 * window_size[1] - 1
relative_position_index = torch.zeros(size=(window_area + 1,) * 2, dtype=relative_coords.dtype)
relative_position_index[1:, 1:] = relative_coords.sum(-1) # Wh*Ww, Wh*Ww
relative_position_index[0, 0:] = num_relative_distance - 3
relative_position_index[0:, 0] = num_relative_distance - 2
relative_position_index[0, 0] = num_relative_distance - 1
return relative_position_index
class Attention(nn.Module):
fused_attn: torch.jit.Final[bool]
def __init__(
self,
dim: int,
num_heads: int = 8,
qkv_bias: bool = False,
attn_drop: float = 0.,
proj_drop: float = 0.,
window_size: Optional[Tuple[int, int]] = None,
attn_head_dim: Optional[int] = None,
):
super().__init__()
self.num_heads = num_heads
head_dim = dim // num_heads
if attn_head_dim is not None:
head_dim = attn_head_dim
all_head_dim = head_dim * self.num_heads
self.scale = head_dim ** -0.5
self.fused_attn = use_fused_attn()
self.qkv = nn.Linear(dim, all_head_dim * 3, bias=False)
if qkv_bias:
self.q_bias = nn.Parameter(torch.zeros(all_head_dim))
self.register_buffer('k_bias', torch.zeros(all_head_dim), persistent=False)
self.v_bias = nn.Parameter(torch.zeros(all_head_dim))
else:
self.q_bias = None
self.k_bias = None
self.v_bias = None
if window_size:
self.window_size = window_size
self.num_relative_distance = (2 * window_size[0] - 1) * (2 * window_size[1] - 1) + 3
self.relative_position_bias_table = nn.Parameter(
torch.zeros(self.num_relative_distance, num_heads)) # 2*Wh-1 * 2*Ww-1, nH
self.register_buffer("relative_position_index", gen_relative_position_index(window_size), persistent=False)
else:
self.window_size = None
self.relative_position_bias_table = None
self.relative_position_index = None
self.attn_drop = nn.Dropout(attn_drop)
self.proj = nn.Linear(all_head_dim, dim)
self.proj_drop = nn.Dropout(proj_drop)
def _get_rel_pos_bias(self):
relative_position_bias = self.relative_position_bias_table[
self.relative_position_index.view(-1)].view(
self.window_size[0] * self.window_size[1] + 1,
self.window_size[0] * self.window_size[1] + 1, -1) # Wh*Ww,Wh*Ww,nH
relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous() # nH, Wh*Ww, Wh*Ww
return relative_position_bias.unsqueeze(0)
def forward(self, x, shared_rel_pos_bias: Optional[torch.Tensor] = None):
B, N, C = x.shape
qkv_bias = torch.cat((self.q_bias, self.k_bias, self.v_bias)) if self.q_bias is not None else None
qkv = F.linear(input=x, weight=self.qkv.weight, bias=qkv_bias)
qkv = qkv.reshape(B, N, 3, self.num_heads, -1).permute(2, 0, 3, 1, 4)
q, k, v = qkv.unbind(0) # B, num_heads, N, head_dim
if self.fused_attn:
rel_pos_bias = None
if self.relative_position_bias_table is not None:
rel_pos_bias = self._get_rel_pos_bias()
if shared_rel_pos_bias is not None:
rel_pos_bias = rel_pos_bias + shared_rel_pos_bias
elif shared_rel_pos_bias is not None:
rel_pos_bias = shared_rel_pos_bias
x = F.scaled_dot_product_attention(
q, k, v,
attn_mask=rel_pos_bias,
dropout_p=self.attn_drop.p if self.training else 0.,
)
else:
q = q * self.scale
attn = (q @ k.transpose(-2, -1))
if self.relative_position_bias_table is not None:
attn = attn + self._get_rel_pos_bias()
if shared_rel_pos_bias is not None:
attn = attn + shared_rel_pos_bias
attn = attn.softmax(dim=-1)
attn = self.attn_drop(attn)
x = attn @ v
x = x.transpose(1, 2).reshape(B, N, C)
x = self.proj(x)
x = self.proj_drop(x)
return x
class Block(nn.Module):
def __init__(
self,
dim: int,
num_heads: int,
qkv_bias: bool = False,
mlp_ratio: float = 4.,
scale_mlp: bool = False,
swiglu_mlp: bool = False,
proj_drop: float = 0.,
attn_drop: float = 0.,
drop_path: float = 0.,
init_values: Optional[float] = None,
act_layer: Callable = nn.GELU,
norm_layer: Callable = LayerNorm,
window_size: Optional[Tuple[int, int]] = None,
attn_head_dim: Optional[int] = None,
):
super().__init__()
self.norm1 = norm_layer(dim)
self.attn = Attention(
dim,
num_heads=num_heads,
qkv_bias=qkv_bias,
attn_drop=attn_drop,
proj_drop=proj_drop,
window_size=window_size,
attn_head_dim=attn_head_dim,
)
# NOTE: drop path for stochastic depth, we shall see if this is better than dropout here
self.drop_path1 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.norm2 = norm_layer(dim)
if swiglu_mlp:
self.mlp = SwiGLU(
in_features=dim,
hidden_features=int(dim * mlp_ratio),
norm_layer=norm_layer if scale_mlp else None,
drop=proj_drop,
)
else:
self.mlp = Mlp(
in_features=dim,
hidden_features=int(dim * mlp_ratio),
act_layer=act_layer,
norm_layer=norm_layer if scale_mlp else None,
drop=proj_drop,
)
self.drop_path2 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
if init_values:
self.gamma_1 = nn.Parameter(init_values * torch.ones(dim))
self.gamma_2 = nn.Parameter(init_values * torch.ones(dim))
else:
self.gamma_1, self.gamma_2 = None, None
def forward(self, x, shared_rel_pos_bias: Optional[torch.Tensor] = None):
if self.gamma_1 is None:
x = x + self.drop_path1(self.attn(self.norm1(x), shared_rel_pos_bias=shared_rel_pos_bias))
x = x + self.drop_path2(self.mlp(self.norm2(x)))
else:
x = x + self.drop_path1(self.gamma_1 * self.attn(self.norm1(x), shared_rel_pos_bias=shared_rel_pos_bias))
x = x + self.drop_path2(self.gamma_2 * self.mlp(self.norm2(x)))
return x
class RelativePositionBias(nn.Module):
def __init__(self, window_size, num_heads):
super().__init__()
self.window_size = window_size
self.window_area = window_size[0] * window_size[1]
num_relative_distance = (2 * window_size[0] - 1) * (2 * window_size[1] - 1) + 3
self.relative_position_bias_table = nn.Parameter(torch.zeros(num_relative_distance, num_heads))
# trunc_normal_(self.relative_position_bias_table, std=.02)
self.register_buffer("relative_position_index", gen_relative_position_index(window_size))
def forward(self):
relative_position_bias = self.relative_position_bias_table[self.relative_position_index.view(-1)].view(
self.window_area + 1, self.window_area + 1, -1) # Wh*Ww,Wh*Ww,nH
return relative_position_bias.permute(2, 0, 1).contiguous() # nH, Wh*Ww, Wh*Ww
class Beit(nn.Module):
""" Vision Transformer with support for patch or hybrid CNN input stage
"""
def __init__(
self,
img_size: Union[int, Tuple[int, int]] = 224,
patch_size: Union[int, Tuple[int, int]] = 16,
in_chans: int = 3,
num_classes: int = 1000,
global_pool: str = 'avg',
embed_dim: int = 768,
depth: int = 12,
num_heads: int = 12,
qkv_bias: bool = True,
mlp_ratio: float = 4.,
swiglu_mlp: bool = False,
scale_mlp: bool = False,
drop_rate: float = 0.,
pos_drop_rate: float = 0.,
proj_drop_rate: float = 0.,
attn_drop_rate: float = 0.,
drop_path_rate: float = 0.,
norm_layer: Callable = LayerNorm,
init_values: Optional[float] = None,
use_abs_pos_emb: bool = True,
use_rel_pos_bias: bool = False,
use_shared_rel_pos_bias: bool = False,
head_init_scale: float = 0.001,
):
super().__init__()
self.num_classes = num_classes
self.global_pool = global_pool
self.num_features = self.embed_dim = embed_dim # num_features for consistency with other models
self.num_prefix_tokens = 1
self.grad_checkpointing = False
self.patch_embed = PatchEmbed(
img_size=img_size,
patch_size=patch_size,
in_chans=in_chans,
embed_dim=embed_dim,
)
num_patches = self.patch_embed.num_patches
self.cls_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
# self.mask_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
self.pos_embed = nn.Parameter(torch.zeros(1, num_patches + 1, embed_dim)) if use_abs_pos_emb else None
self.pos_drop = nn.Dropout(p=pos_drop_rate)
if use_shared_rel_pos_bias:
self.rel_pos_bias = RelativePositionBias(
window_size=self.patch_embed.grid_size,
num_heads=num_heads,
)
else:
self.rel_pos_bias = None
dpr = [x.item() for x in torch.linspace(0, drop_path_rate, depth)] # stochastic depth decay rule
self.blocks = nn.ModuleList([
Block(
dim=embed_dim,
num_heads=num_heads,
qkv_bias=qkv_bias,
mlp_ratio=mlp_ratio,
scale_mlp=scale_mlp,
swiglu_mlp=swiglu_mlp,
proj_drop=proj_drop_rate,
attn_drop=attn_drop_rate,
drop_path=dpr[i],
norm_layer=norm_layer,
init_values=init_values,
window_size=self.patch_embed.grid_size if use_rel_pos_bias else None,
)
for i in range(depth)])
use_fc_norm = self.global_pool == 'avg'
self.norm = nn.Identity() if use_fc_norm else norm_layer(embed_dim)
self.fc_norm = norm_layer(embed_dim) if use_fc_norm else nn.Identity()
self.head_drop = nn.Dropout(drop_rate)
self.head = nn.Linear(embed_dim, num_classes) if num_classes > 0 else nn.Identity()
self.apply(self._init_weights)
if self.pos_embed is not None:
trunc_normal_(self.pos_embed, std=.02)
trunc_normal_(self.cls_token, std=.02)
self.fix_init_weight()
if isinstance(self.head, nn.Linear):
trunc_normal_(self.head.weight, std=.02)
self.head.weight.data.mul_(head_init_scale)
self.head.bias.data.mul_(head_init_scale)
def fix_init_weight(self):
def rescale(param, layer_id):
param.div_(math.sqrt(2.0 * layer_id))
for layer_id, layer in enumerate(self.blocks):
rescale(layer.attn.proj.weight.data, layer_id + 1)
rescale(layer.mlp.fc2.weight.data, layer_id + 1)
def _init_weights(self, m):
if isinstance(m, nn.Linear):
trunc_normal_(m.weight, std=.02)
if isinstance(m, nn.Linear) and m.bias is not None:
nn.init.constant_(m.bias, 0)
elif isinstance(m, nn.LayerNorm):
nn.init.constant_(m.bias, 0)
nn.init.constant_(m.weight, 1.0)
@torch.jit.ignore
def no_weight_decay(self):
nwd = {'pos_embed', 'cls_token'}
for n, _ in self.named_parameters():
if 'relative_position_bias_table' in n:
nwd.add(n)
return nwd
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
self.grad_checkpointing = enable
@torch.jit.ignore
def group_matcher(self, coarse=False):
matcher = dict(
stem=r'^cls_token|pos_embed|patch_embed|rel_pos_bias', # stem and embed
blocks=[(r'^blocks\.(\d+)', None), (r'^norm', (99999,))],
)
return matcher
@torch.jit.ignore
def get_classifier(self):
return self.head
def reset_classifier(self, num_classes, global_pool=None):
self.num_classes = num_classes
if global_pool is not None:
self.global_pool = global_pool
self.head = nn.Linear(self.embed_dim, num_classes) if num_classes > 0 else nn.Identity()
def forward_features(self, x):
x = self.patch_embed(x)
x = torch.cat((self.cls_token.expand(x.shape[0], -1, -1), x), dim=1)
if self.pos_embed is not None:
x = x + self.pos_embed
x = self.pos_drop(x)
rel_pos_bias = self.rel_pos_bias() if self.rel_pos_bias is not None else None
for blk in self.blocks:
if self.grad_checkpointing and not torch.jit.is_scripting():
x = checkpoint(blk, x, shared_rel_pos_bias=rel_pos_bias)
else:
x = blk(x, shared_rel_pos_bias=rel_pos_bias)
x = self.norm(x)
return x
def forward_head(self, x, pre_logits: bool = False):
if self.global_pool:
x = x[:, self.num_prefix_tokens:].mean(dim=1) if self.global_pool == 'avg' else x[:, 0]
x = self.fc_norm(x)
x = self.head_drop(x)
return x if pre_logits else self.head(x)
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': None,
'crop_pct': .9, 'interpolation': 'bicubic', 'fixed_input_size': True,
'mean': (0.5, 0.5, 0.5), 'std': (0.5, 0.5, 0.5),
'first_conv': 'patch_embed.proj', 'classifier': 'head',
**kwargs
}
default_cfgs = generate_default_cfgs({
'beit_base_patch16_224.in22k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_base_patch16_224_pt22k_ft22kto1k.pth',
hf_hub_id='timm/'),
'beit_base_patch16_384.in22k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_base_patch16_384_pt22k_ft22kto1k.pth',
hf_hub_id='timm/',
input_size=(3, 384, 384), crop_pct=1.0,
),
'beit_base_patch16_224.in22k_ft_in22k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_base_patch16_224_pt22k_ft22k.pth',
hf_hub_id='timm/',
num_classes=21841,
),
'beit_large_patch16_224.in22k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_large_patch16_224_pt22k_ft22kto1k.pth',
hf_hub_id='timm/'),
'beit_large_patch16_384.in22k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_large_patch16_384_pt22k_ft22kto1k.pth',
hf_hub_id='timm/',
input_size=(3, 384, 384), crop_pct=1.0,
),
'beit_large_patch16_512.in22k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_large_patch16_512_pt22k_ft22kto1k.pth',
hf_hub_id='timm/',
input_size=(3, 512, 512), crop_pct=1.0,
),
'beit_large_patch16_224.in22k_ft_in22k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beit/beit_large_patch16_224_pt22k_ft22k.pth',
hf_hub_id='timm/',
num_classes=21841,
),
'beitv2_base_patch16_224.in1k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_base_patch16_224_pt1k_ft21kto1k.pth',
hf_hub_id='timm/',
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
'beitv2_base_patch16_224.in1k_ft_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_base_patch16_224_pt1k_ft1k.pth',
hf_hub_id='timm/',
mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
'beitv2_base_patch16_224.in1k_ft_in22k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_base_patch16_224_pt1k_ft21k.pth',
hf_hub_id='timm/',
num_classes=21841, mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
'beitv2_large_patch16_224.in1k_ft_in22k_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_large_patch16_224_pt1k_ft21kto1k.pth',
hf_hub_id='timm/',
crop_pct=0.95, mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
'beitv2_large_patch16_224.in1k_ft_in1k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_large_patch16_224_pt1k_ft1k.pth',
hf_hub_id='timm/',
crop_pct=0.95, mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
'beitv2_large_patch16_224.in1k_ft_in22k': _cfg(
#url='https://conversationhub.blob.core.windows.net/beit-share-public/beitv2/beitv2_large_patch16_224_pt1k_ft21k.pth',
hf_hub_id='timm/',
num_classes=21841, mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD
),
})
def _beit_checkpoint_filter_fn(state_dict, model, interpolation='bicubic', antialias=True):
state_dict = state_dict.get('model', state_dict)
state_dict = state_dict.get('module', state_dict)
# beit v2 didn't strip module
out_dict = {}
for k, v in state_dict.items():
if 'relative_position_index' in k:
continue
if 'patch_embed.proj.weight' in k:
O, I, H, W = model.patch_embed.proj.weight.shape
if v.shape[-1] != W or v.shape[-2] != H:
v = resample_patch_embed(
v,
(H, W),
interpolation=interpolation,
antialias=antialias,
verbose=True,
)
elif k == 'pos_embed' and v.shape[1] != model.pos_embed.shape[1]:
# To resize pos embedding when using model at different size from pretrained weights
num_prefix_tokens = 1
v = resample_abs_pos_embed(
v,
new_size=model.patch_embed.grid_size,
num_prefix_tokens=num_prefix_tokens,
interpolation=interpolation,
antialias=antialias,
verbose=True,
)
elif k.endswith('relative_position_bias_table'):
m = model.get_submodule(k[:-29])
if v.shape != m.relative_position_bias_table.shape or m.window_size[0] != m.window_size[1]:
v = resize_rel_pos_bias_table(
v,
new_window_size=m.window_size,
new_bias_shape=m.relative_position_bias_table.shape,
)
out_dict[k] = v
return out_dict
def _create_beit(variant, pretrained=False, **kwargs):
if kwargs.get('features_only', None):
raise RuntimeError('features_only not implemented for BEiT models.')
model = build_model_with_cfg(
Beit, variant, pretrained,
# FIXME an updated filter fn needed to interpolate rel pos emb if fine tuning to diff model sizes
pretrained_filter_fn=_beit_checkpoint_filter_fn,
**kwargs)
return model
@register_model
def beit_base_patch16_224(pretrained=False, **kwargs) -> Beit:
model_args = dict(
patch_size=16, embed_dim=768, depth=12, num_heads=12, mlp_ratio=4,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=0.1)
model = _create_beit('beit_base_patch16_224', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beit_base_patch16_384(pretrained=False, **kwargs) -> Beit:
model_args = dict(
img_size=384, patch_size=16, embed_dim=768, depth=12, num_heads=12,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=0.1)
model = _create_beit('beit_base_patch16_384', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beit_large_patch16_224(pretrained=False, **kwargs) -> Beit:
model_args = dict(
patch_size=16, embed_dim=1024, depth=24, num_heads=16,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=1e-5)
model = _create_beit('beit_large_patch16_224', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beit_large_patch16_384(pretrained=False, **kwargs) -> Beit:
model_args = dict(
img_size=384, patch_size=16, embed_dim=1024, depth=24, num_heads=16,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=1e-5)
model = _create_beit('beit_large_patch16_384', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beit_large_patch16_512(pretrained=False, **kwargs) -> Beit:
model_args = dict(
img_size=512, patch_size=16, embed_dim=1024, depth=24, num_heads=16,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=1e-5)
model = _create_beit('beit_large_patch16_512', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beitv2_base_patch16_224(pretrained=False, **kwargs) -> Beit:
model_args = dict(
patch_size=16, embed_dim=768, depth=12, num_heads=12, mlp_ratio=4,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=1e-5)
model = _create_beit('beitv2_base_patch16_224', pretrained=pretrained, **dict(model_args, **kwargs))
return model
@register_model
def beitv2_large_patch16_224(pretrained=False, **kwargs) -> Beit:
model_args = dict(
patch_size=16, embed_dim=1024, depth=24, num_heads=16,
use_abs_pos_emb=False, use_rel_pos_bias=True, init_values=1e-5)
model = _create_beit('beitv2_large_patch16_224', pretrained=pretrained, **dict(model_args, **kwargs))
return model
| pytorch-image-models/timm/models/beit.py/0 | {
"file_path": "pytorch-image-models/timm/models/beit.py",
"repo_id": "pytorch-image-models",
"token_count": 12467
} | 178 |
""" EfficientFormer
@article{li2022efficientformer,
title={EfficientFormer: Vision Transformers at MobileNet Speed},
author={Li, Yanyu and Yuan, Geng and Wen, Yang and Hu, Eric and Evangelidis, Georgios and Tulyakov,
Sergey and Wang, Yanzhi and Ren, Jian},
journal={arXiv preprint arXiv:2206.01191},
year={2022}
}
Based on Apache 2.0 licensed code at https://github.com/snap-research/EfficientFormer, Copyright (c) 2022 Snap Inc.
Modifications and timm support by / Copyright 2022, Ross Wightman
"""
from typing import Dict
import torch
import torch.nn as nn
from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.layers import DropPath, trunc_normal_, to_2tuple, Mlp, ndgrid
from ._builder import build_model_with_cfg
from ._manipulate import checkpoint_seq
from ._registry import generate_default_cfgs, register_model
__all__ = ['EfficientFormer'] # model_registry will add each entrypoint fn to this
EfficientFormer_width = {
'l1': (48, 96, 224, 448),
'l3': (64, 128, 320, 512),
'l7': (96, 192, 384, 768),
}
EfficientFormer_depth = {
'l1': (3, 2, 6, 4),
'l3': (4, 4, 12, 6),
'l7': (6, 6, 18, 8),
}
class Attention(torch.nn.Module):
attention_bias_cache: Dict[str, torch.Tensor]
def __init__(
self,
dim=384,
key_dim=32,
num_heads=8,
attn_ratio=4,
resolution=7
):
super().__init__()
self.num_heads = num_heads
self.scale = key_dim ** -0.5
self.key_dim = key_dim
self.key_attn_dim = key_dim * num_heads
self.val_dim = int(attn_ratio * key_dim)
self.val_attn_dim = self.val_dim * num_heads
self.attn_ratio = attn_ratio
self.qkv = nn.Linear(dim, self.key_attn_dim * 2 + self.val_attn_dim)
self.proj = nn.Linear(self.val_attn_dim, dim)
resolution = to_2tuple(resolution)
pos = torch.stack(ndgrid(torch.arange(resolution[0]), torch.arange(resolution[1]))).flatten(1)
rel_pos = (pos[..., :, None] - pos[..., None, :]).abs()
rel_pos = (rel_pos[0] * resolution[1]) + rel_pos[1]
self.attention_biases = torch.nn.Parameter(torch.zeros(num_heads, resolution[0] * resolution[1]))
self.register_buffer('attention_bias_idxs', rel_pos)
self.attention_bias_cache = {} # per-device attention_biases cache (data-parallel compat)
@torch.no_grad()
def train(self, mode=True):
super().train(mode)
if mode and self.attention_bias_cache:
self.attention_bias_cache = {} # clear ab cache
def get_attention_biases(self, device: torch.device) -> torch.Tensor:
if torch.jit.is_tracing() or self.training:
return self.attention_biases[:, self.attention_bias_idxs]
else:
device_key = str(device)
if device_key not in self.attention_bias_cache:
self.attention_bias_cache[device_key] = self.attention_biases[:, self.attention_bias_idxs]
return self.attention_bias_cache[device_key]
def forward(self, x): # x (B,N,C)
B, N, C = x.shape
qkv = self.qkv(x)
qkv = qkv.reshape(B, N, self.num_heads, -1).permute(0, 2, 1, 3)
q, k, v = qkv.split([self.key_dim, self.key_dim, self.val_dim], dim=3)
attn = (q @ k.transpose(-2, -1)) * self.scale
attn = attn + self.get_attention_biases(x.device)
attn = attn.softmax(dim=-1)
x = (attn @ v).transpose(1, 2).reshape(B, N, self.val_attn_dim)
x = self.proj(x)
return x
class Stem4(nn.Sequential):
def __init__(self, in_chs, out_chs, act_layer=nn.ReLU, norm_layer=nn.BatchNorm2d):
super().__init__()
self.stride = 4
self.add_module('conv1', nn.Conv2d(in_chs, out_chs // 2, kernel_size=3, stride=2, padding=1))
self.add_module('norm1', norm_layer(out_chs // 2))
self.add_module('act1', act_layer())
self.add_module('conv2', nn.Conv2d(out_chs // 2, out_chs, kernel_size=3, stride=2, padding=1))
self.add_module('norm2', norm_layer(out_chs))
self.add_module('act2', act_layer())
class Downsample(nn.Module):
"""
Downsampling via strided conv w/ norm
Input: tensor in shape [B, C, H, W]
Output: tensor in shape [B, C, H/stride, W/stride]
"""
def __init__(self, in_chs, out_chs, kernel_size=3, stride=2, padding=None, norm_layer=nn.BatchNorm2d):
super().__init__()
if padding is None:
padding = kernel_size // 2
self.conv = nn.Conv2d(in_chs, out_chs, kernel_size=kernel_size, stride=stride, padding=padding)
self.norm = norm_layer(out_chs)
def forward(self, x):
x = self.conv(x)
x = self.norm(x)
return x
class Flat(nn.Module):
def __init__(self, ):
super().__init__()
def forward(self, x):
x = x.flatten(2).transpose(1, 2)
return x
class Pooling(nn.Module):
"""
Implementation of pooling for PoolFormer
--pool_size: pooling size
"""
def __init__(self, pool_size=3):
super().__init__()
self.pool = nn.AvgPool2d(pool_size, stride=1, padding=pool_size // 2, count_include_pad=False)
def forward(self, x):
return self.pool(x) - x
class ConvMlpWithNorm(nn.Module):
"""
Implementation of MLP with 1*1 convolutions.
Input: tensor with shape [B, C, H, W]
"""
def __init__(
self,
in_features,
hidden_features=None,
out_features=None,
act_layer=nn.GELU,
norm_layer=nn.BatchNorm2d,
drop=0.
):
super().__init__()
out_features = out_features or in_features
hidden_features = hidden_features or in_features
self.fc1 = nn.Conv2d(in_features, hidden_features, 1)
self.norm1 = norm_layer(hidden_features) if norm_layer is not None else nn.Identity()
self.act = act_layer()
self.fc2 = nn.Conv2d(hidden_features, out_features, 1)
self.norm2 = norm_layer(out_features) if norm_layer is not None else nn.Identity()
self.drop = nn.Dropout(drop)
def forward(self, x):
x = self.fc1(x)
x = self.norm1(x)
x = self.act(x)
x = self.drop(x)
x = self.fc2(x)
x = self.norm2(x)
x = self.drop(x)
return x
class LayerScale(nn.Module):
def __init__(self, dim, init_values=1e-5, inplace=False):
super().__init__()
self.inplace = inplace
self.gamma = nn.Parameter(init_values * torch.ones(dim))
def forward(self, x):
return x.mul_(self.gamma) if self.inplace else x * self.gamma
class MetaBlock1d(nn.Module):
def __init__(
self,
dim,
mlp_ratio=4.,
act_layer=nn.GELU,
norm_layer=nn.LayerNorm,
proj_drop=0.,
drop_path=0.,
layer_scale_init_value=1e-5
):
super().__init__()
self.norm1 = norm_layer(dim)
self.token_mixer = Attention(dim)
self.norm2 = norm_layer(dim)
self.mlp = Mlp(
in_features=dim,
hidden_features=int(dim * mlp_ratio),
act_layer=act_layer,
drop=proj_drop,
)
self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.ls1 = LayerScale(dim, layer_scale_init_value)
self.ls2 = LayerScale(dim, layer_scale_init_value)
def forward(self, x):
x = x + self.drop_path(self.ls1(self.token_mixer(self.norm1(x))))
x = x + self.drop_path(self.ls2(self.mlp(self.norm2(x))))
return x
class LayerScale2d(nn.Module):
def __init__(self, dim, init_values=1e-5, inplace=False):
super().__init__()
self.inplace = inplace
self.gamma = nn.Parameter(init_values * torch.ones(dim))
def forward(self, x):
gamma = self.gamma.view(1, -1, 1, 1)
return x.mul_(gamma) if self.inplace else x * gamma
class MetaBlock2d(nn.Module):
def __init__(
self,
dim,
pool_size=3,
mlp_ratio=4.,
act_layer=nn.GELU,
norm_layer=nn.BatchNorm2d,
proj_drop=0.,
drop_path=0.,
layer_scale_init_value=1e-5
):
super().__init__()
self.token_mixer = Pooling(pool_size=pool_size)
self.ls1 = LayerScale2d(dim, layer_scale_init_value)
self.drop_path1 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.mlp = ConvMlpWithNorm(
dim,
hidden_features=int(dim * mlp_ratio),
act_layer=act_layer,
norm_layer=norm_layer,
drop=proj_drop,
)
self.ls2 = LayerScale2d(dim, layer_scale_init_value)
self.drop_path2 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
def forward(self, x):
x = x + self.drop_path1(self.ls1(self.token_mixer(x)))
x = x + self.drop_path2(self.ls2(self.mlp(x)))
return x
class EfficientFormerStage(nn.Module):
def __init__(
self,
dim,
dim_out,
depth,
downsample=True,
num_vit=1,
pool_size=3,
mlp_ratio=4.,
act_layer=nn.GELU,
norm_layer=nn.BatchNorm2d,
norm_layer_cl=nn.LayerNorm,
proj_drop=.0,
drop_path=0.,
layer_scale_init_value=1e-5,
):
super().__init__()
self.grad_checkpointing = False
if downsample:
self.downsample = Downsample(in_chs=dim, out_chs=dim_out, norm_layer=norm_layer)
dim = dim_out
else:
assert dim == dim_out
self.downsample = nn.Identity()
blocks = []
if num_vit and num_vit >= depth:
blocks.append(Flat())
for block_idx in range(depth):
remain_idx = depth - block_idx - 1
if num_vit and num_vit > remain_idx:
blocks.append(
MetaBlock1d(
dim,
mlp_ratio=mlp_ratio,
act_layer=act_layer,
norm_layer=norm_layer_cl,
proj_drop=proj_drop,
drop_path=drop_path[block_idx],
layer_scale_init_value=layer_scale_init_value,
))
else:
blocks.append(
MetaBlock2d(
dim,
pool_size=pool_size,
mlp_ratio=mlp_ratio,
act_layer=act_layer,
norm_layer=norm_layer,
proj_drop=proj_drop,
drop_path=drop_path[block_idx],
layer_scale_init_value=layer_scale_init_value,
))
if num_vit and num_vit == remain_idx:
blocks.append(Flat())
self.blocks = nn.Sequential(*blocks)
def forward(self, x):
x = self.downsample(x)
if self.grad_checkpointing and not torch.jit.is_scripting():
x = checkpoint_seq(self.blocks, x)
else:
x = self.blocks(x)
return x
class EfficientFormer(nn.Module):
def __init__(
self,
depths,
embed_dims=None,
in_chans=3,
num_classes=1000,
global_pool='avg',
downsamples=None,
num_vit=0,
mlp_ratios=4,
pool_size=3,
layer_scale_init_value=1e-5,
act_layer=nn.GELU,
norm_layer=nn.BatchNorm2d,
norm_layer_cl=nn.LayerNorm,
drop_rate=0.,
proj_drop_rate=0.,
drop_path_rate=0.,
**kwargs
):
super().__init__()
self.num_classes = num_classes
self.global_pool = global_pool
self.stem = Stem4(in_chans, embed_dims[0], norm_layer=norm_layer)
prev_dim = embed_dims[0]
# stochastic depth decay rule
dpr = [x.tolist() for x in torch.linspace(0, drop_path_rate, sum(depths)).split(depths)]
downsamples = downsamples or (False,) + (True,) * (len(depths) - 1)
stages = []
for i in range(len(depths)):
stage = EfficientFormerStage(
prev_dim,
embed_dims[i],
depths[i],
downsample=downsamples[i],
num_vit=num_vit if i == 3 else 0,
pool_size=pool_size,
mlp_ratio=mlp_ratios,
act_layer=act_layer,
norm_layer_cl=norm_layer_cl,
norm_layer=norm_layer,
proj_drop=proj_drop_rate,
drop_path=dpr[i],
layer_scale_init_value=layer_scale_init_value,
)
prev_dim = embed_dims[i]
stages.append(stage)
self.stages = nn.Sequential(*stages)
# Classifier head
self.num_features = embed_dims[-1]
self.norm = norm_layer_cl(self.num_features)
self.head_drop = nn.Dropout(drop_rate)
self.head = nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity()
# assuming model is always distilled (valid for current checkpoints, will split def if that changes)
self.head_dist = nn.Linear(embed_dims[-1], num_classes) if num_classes > 0 else nn.Identity()
self.distilled_training = False # must set this True to train w/ distillation token
self.apply(self._init_weights)
# init for classification
def _init_weights(self, m):
if isinstance(m, nn.Linear):
trunc_normal_(m.weight, std=.02)
if isinstance(m, nn.Linear) and m.bias is not None:
nn.init.constant_(m.bias, 0)
@torch.jit.ignore
def no_weight_decay(self):
return {k for k, _ in self.named_parameters() if 'attention_biases' in k}
@torch.jit.ignore
def group_matcher(self, coarse=False):
matcher = dict(
stem=r'^stem', # stem and embed
blocks=[(r'^stages\.(\d+)', None), (r'^norm', (99999,))]
)
return matcher
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
for s in self.stages:
s.grad_checkpointing = enable
@torch.jit.ignore
def get_classifier(self):
return self.head, self.head_dist
def reset_classifier(self, num_classes, global_pool=None):
self.num_classes = num_classes
if global_pool is not None:
self.global_pool = global_pool
self.head = nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity()
self.head_dist = nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity()
@torch.jit.ignore
def set_distilled_training(self, enable=True):
self.distilled_training = enable
def forward_features(self, x):
x = self.stem(x)
x = self.stages(x)
x = self.norm(x)
return x
def forward_head(self, x, pre_logits: bool = False):
if self.global_pool == 'avg':
x = x.mean(dim=1)
x = self.head_drop(x)
if pre_logits:
return x
x, x_dist = self.head(x), self.head_dist(x)
if self.distilled_training and self.training and not torch.jit.is_scripting():
# only return separate classification predictions when training in distilled mode
return x, x_dist
else:
# during standard train/finetune, inference average the classifier predictions
return (x + x_dist) / 2
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def _checkpoint_filter_fn(state_dict, model):
""" Remap original checkpoints -> timm """
if 'stem.0.weight' in state_dict:
return state_dict # non-original checkpoint, no remapping needed
out_dict = {}
import re
stage_idx = 0
for k, v in state_dict.items():
if k.startswith('patch_embed'):
k = k.replace('patch_embed.0', 'stem.conv1')
k = k.replace('patch_embed.1', 'stem.norm1')
k = k.replace('patch_embed.3', 'stem.conv2')
k = k.replace('patch_embed.4', 'stem.norm2')
if re.match(r'network\.(\d+)\.proj\.weight', k):
stage_idx += 1
k = re.sub(r'network.(\d+).(\d+)', f'stages.{stage_idx}.blocks.\\2', k)
k = re.sub(r'network.(\d+).proj', f'stages.{stage_idx}.downsample.conv', k)
k = re.sub(r'network.(\d+).norm', f'stages.{stage_idx}.downsample.norm', k)
k = re.sub(r'layer_scale_([0-9])', r'ls\1.gamma', k)
k = k.replace('dist_head', 'head_dist')
out_dict[k] = v
return out_dict
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': None, 'fixed_input_size': True,
'crop_pct': .95, 'interpolation': 'bicubic',
'mean': IMAGENET_DEFAULT_MEAN, 'std': IMAGENET_DEFAULT_STD,
'first_conv': 'stem.conv1', 'classifier': ('head', 'head_dist'),
**kwargs
}
default_cfgs = generate_default_cfgs({
'efficientformer_l1.snap_dist_in1k': _cfg(
hf_hub_id='timm/',
),
'efficientformer_l3.snap_dist_in1k': _cfg(
hf_hub_id='timm/',
),
'efficientformer_l7.snap_dist_in1k': _cfg(
hf_hub_id='timm/',
),
})
def _create_efficientformer(variant, pretrained=False, **kwargs):
if kwargs.get('features_only', None):
raise RuntimeError('features_only not implemented for EfficientFormer models.')
model = build_model_with_cfg(
EfficientFormer, variant, pretrained,
pretrained_filter_fn=_checkpoint_filter_fn,
**kwargs)
return model
@register_model
def efficientformer_l1(pretrained=False, **kwargs) -> EfficientFormer:
model_args = dict(
depths=EfficientFormer_depth['l1'],
embed_dims=EfficientFormer_width['l1'],
num_vit=1,
)
return _create_efficientformer('efficientformer_l1', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def efficientformer_l3(pretrained=False, **kwargs) -> EfficientFormer:
model_args = dict(
depths=EfficientFormer_depth['l3'],
embed_dims=EfficientFormer_width['l3'],
num_vit=4,
)
return _create_efficientformer('efficientformer_l3', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def efficientformer_l7(pretrained=False, **kwargs) -> EfficientFormer:
model_args = dict(
depths=EfficientFormer_depth['l7'],
embed_dims=EfficientFormer_width['l7'],
num_vit=8,
)
return _create_efficientformer('efficientformer_l7', pretrained=pretrained, **dict(model_args, **kwargs))
| pytorch-image-models/timm/models/efficientformer.py/0 | {
"file_path": "pytorch-image-models/timm/models/efficientformer.py",
"repo_id": "pytorch-image-models",
"token_count": 9481
} | 179 |
""" Pooling-based Vision Transformer (PiT) in PyTorch
A PyTorch implement of Pooling-based Vision Transformers as described in
'Rethinking Spatial Dimensions of Vision Transformers' - https://arxiv.org/abs/2103.16302
This code was adapted from the original version at https://github.com/naver-ai/pit, original copyright below.
Modifications for timm by / Copyright 2020 Ross Wightman
"""
# PiT
# Copyright 2021-present NAVER Corp.
# Apache License v2.0
import math
import re
from functools import partial
from typing import Sequence, Tuple
import torch
from torch import nn
from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.layers import trunc_normal_, to_2tuple, LayerNorm
from ._builder import build_model_with_cfg
from ._registry import register_model, generate_default_cfgs
from .vision_transformer import Block
__all__ = ['PoolingVisionTransformer'] # model_registry will add each entrypoint fn to this
class SequentialTuple(nn.Sequential):
""" This module exists to work around torchscript typing issues list -> list"""
def __init__(self, *args):
super(SequentialTuple, self).__init__(*args)
def forward(self, x: Tuple[torch.Tensor, torch.Tensor]) -> Tuple[torch.Tensor, torch.Tensor]:
for module in self:
x = module(x)
return x
class Transformer(nn.Module):
def __init__(
self,
base_dim,
depth,
heads,
mlp_ratio,
pool=None,
proj_drop=.0,
attn_drop=.0,
drop_path_prob=None,
norm_layer=None,
):
super(Transformer, self).__init__()
embed_dim = base_dim * heads
self.pool = pool
self.norm = norm_layer(embed_dim) if norm_layer else nn.Identity()
self.blocks = nn.Sequential(*[
Block(
dim=embed_dim,
num_heads=heads,
mlp_ratio=mlp_ratio,
qkv_bias=True,
proj_drop=proj_drop,
attn_drop=attn_drop,
drop_path=drop_path_prob[i],
norm_layer=partial(nn.LayerNorm, eps=1e-6)
)
for i in range(depth)])
def forward(self, x: Tuple[torch.Tensor, torch.Tensor]) -> Tuple[torch.Tensor, torch.Tensor]:
x, cls_tokens = x
token_length = cls_tokens.shape[1]
if self.pool is not None:
x, cls_tokens = self.pool(x, cls_tokens)
B, C, H, W = x.shape
x = x.flatten(2).transpose(1, 2)
x = torch.cat((cls_tokens, x), dim=1)
x = self.norm(x)
x = self.blocks(x)
cls_tokens = x[:, :token_length]
x = x[:, token_length:]
x = x.transpose(1, 2).reshape(B, C, H, W)
return x, cls_tokens
class Pooling(nn.Module):
def __init__(self, in_feature, out_feature, stride, padding_mode='zeros'):
super(Pooling, self).__init__()
self.conv = nn.Conv2d(
in_feature,
out_feature,
kernel_size=stride + 1,
padding=stride // 2,
stride=stride,
padding_mode=padding_mode,
groups=in_feature,
)
self.fc = nn.Linear(in_feature, out_feature)
def forward(self, x, cls_token) -> Tuple[torch.Tensor, torch.Tensor]:
x = self.conv(x)
cls_token = self.fc(cls_token)
return x, cls_token
class ConvEmbedding(nn.Module):
def __init__(
self,
in_channels,
out_channels,
img_size: int = 224,
patch_size: int = 16,
stride: int = 8,
padding: int = 0,
):
super(ConvEmbedding, self).__init__()
padding = padding
self.img_size = to_2tuple(img_size)
self.patch_size = to_2tuple(patch_size)
self.height = math.floor((self.img_size[0] + 2 * padding - self.patch_size[0]) / stride + 1)
self.width = math.floor((self.img_size[1] + 2 * padding - self.patch_size[1]) / stride + 1)
self.grid_size = (self.height, self.width)
self.conv = nn.Conv2d(
in_channels, out_channels, kernel_size=patch_size,
stride=stride, padding=padding, bias=True)
def forward(self, x):
x = self.conv(x)
return x
class PoolingVisionTransformer(nn.Module):
""" Pooling-based Vision Transformer
A PyTorch implement of 'Rethinking Spatial Dimensions of Vision Transformers'
- https://arxiv.org/abs/2103.16302
"""
def __init__(
self,
img_size: int = 224,
patch_size: int = 16,
stride: int = 8,
stem_type: str = 'overlap',
base_dims: Sequence[int] = (48, 48, 48),
depth: Sequence[int] = (2, 6, 4),
heads: Sequence[int] = (2, 4, 8),
mlp_ratio: float = 4,
num_classes=1000,
in_chans=3,
global_pool='token',
distilled=False,
drop_rate=0.,
pos_drop_drate=0.,
proj_drop_rate=0.,
attn_drop_rate=0.,
drop_path_rate=0.,
):
super(PoolingVisionTransformer, self).__init__()
assert global_pool in ('token',)
self.base_dims = base_dims
self.heads = heads
embed_dim = base_dims[0] * heads[0]
self.num_classes = num_classes
self.global_pool = global_pool
self.num_tokens = 2 if distilled else 1
self.feature_info = []
self.patch_embed = ConvEmbedding(in_chans, embed_dim, img_size, patch_size, stride)
self.pos_embed = nn.Parameter(torch.randn(1, embed_dim, self.patch_embed.height, self.patch_embed.width))
self.cls_token = nn.Parameter(torch.randn(1, self.num_tokens, embed_dim))
self.pos_drop = nn.Dropout(p=pos_drop_drate)
transformers = []
# stochastic depth decay rule
dpr = [x.tolist() for x in torch.linspace(0, drop_path_rate, sum(depth)).split(depth)]
prev_dim = embed_dim
for i in range(len(depth)):
pool = None
embed_dim = base_dims[i] * heads[i]
if i > 0:
pool = Pooling(
prev_dim,
embed_dim,
stride=2,
)
transformers += [Transformer(
base_dims[i],
depth[i],
heads[i],
mlp_ratio,
pool=pool,
proj_drop=proj_drop_rate,
attn_drop=attn_drop_rate,
drop_path_prob=dpr[i],
)]
prev_dim = embed_dim
self.feature_info += [dict(num_chs=prev_dim, reduction=(stride - 1) * 2**i, module=f'transformers.{i}')]
self.transformers = SequentialTuple(*transformers)
self.norm = nn.LayerNorm(base_dims[-1] * heads[-1], eps=1e-6)
self.num_features = self.embed_dim = embed_dim
# Classifier head
self.head_drop = nn.Dropout(drop_rate)
self.head = nn.Linear(self.embed_dim, num_classes) if num_classes > 0 else nn.Identity()
self.head_dist = None
if distilled:
self.head_dist = nn.Linear(self.embed_dim, self.num_classes) if num_classes > 0 else nn.Identity()
self.distilled_training = False # must set this True to train w/ distillation token
trunc_normal_(self.pos_embed, std=.02)
trunc_normal_(self.cls_token, std=.02)
self.apply(self._init_weights)
def _init_weights(self, m):
if isinstance(m, nn.LayerNorm):
nn.init.constant_(m.bias, 0)
nn.init.constant_(m.weight, 1.0)
@torch.jit.ignore
def no_weight_decay(self):
return {'pos_embed', 'cls_token'}
@torch.jit.ignore
def set_distilled_training(self, enable=True):
self.distilled_training = enable
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
assert not enable, 'gradient checkpointing not supported'
def get_classifier(self):
if self.head_dist is not None:
return self.head, self.head_dist
else:
return self.head
def reset_classifier(self, num_classes, global_pool=None):
self.num_classes = num_classes
self.head = nn.Linear(self.embed_dim, num_classes) if num_classes > 0 else nn.Identity()
if self.head_dist is not None:
self.head_dist = nn.Linear(self.embed_dim, self.num_classes) if num_classes > 0 else nn.Identity()
def forward_features(self, x):
x = self.patch_embed(x)
x = self.pos_drop(x + self.pos_embed)
cls_tokens = self.cls_token.expand(x.shape[0], -1, -1)
x, cls_tokens = self.transformers((x, cls_tokens))
cls_tokens = self.norm(cls_tokens)
return cls_tokens
def forward_head(self, x, pre_logits: bool = False) -> torch.Tensor:
if self.head_dist is not None:
assert self.global_pool == 'token'
x, x_dist = x[:, 0], x[:, 1]
x = self.head_drop(x)
x_dist = self.head_drop(x)
if not pre_logits:
x = self.head(x)
x_dist = self.head_dist(x_dist)
if self.distilled_training and self.training and not torch.jit.is_scripting():
# only return separate classification predictions when training in distilled mode
return x, x_dist
else:
# during standard train / finetune, inference average the classifier predictions
return (x + x_dist) / 2
else:
if self.global_pool == 'token':
x = x[:, 0]
x = self.head_drop(x)
if not pre_logits:
x = self.head(x)
return x
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def checkpoint_filter_fn(state_dict, model):
""" preprocess checkpoints """
out_dict = {}
p_blocks = re.compile(r'pools\.(\d)\.')
for k, v in state_dict.items():
# FIXME need to update resize for PiT impl
# if k == 'pos_embed' and v.shape != model.pos_embed.shape:
# # To resize pos embedding when using model at different size from pretrained weights
# v = resize_pos_embed(v, model.pos_embed)
k = p_blocks.sub(lambda exp: f'transformers.{int(exp.group(1)) + 1}.pool.', k)
out_dict[k] = v
return out_dict
def _create_pit(variant, pretrained=False, **kwargs):
default_out_indices = tuple(range(3))
out_indices = kwargs.pop('out_indices', default_out_indices)
model = build_model_with_cfg(
PoolingVisionTransformer,
variant,
pretrained,
pretrained_filter_fn=checkpoint_filter_fn,
feature_cfg=dict(feature_cls='hook', no_rewrite=True, out_indices=out_indices),
**kwargs,
)
return model
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': None,
'crop_pct': .9, 'interpolation': 'bicubic', 'fixed_input_size': True,
'mean': IMAGENET_DEFAULT_MEAN, 'std': IMAGENET_DEFAULT_STD,
'first_conv': 'patch_embed.conv', 'classifier': 'head',
**kwargs
}
default_cfgs = generate_default_cfgs({
# deit models (FB weights)
'pit_ti_224.in1k': _cfg(hf_hub_id='timm/'),
'pit_xs_224.in1k': _cfg(hf_hub_id='timm/'),
'pit_s_224.in1k': _cfg(hf_hub_id='timm/'),
'pit_b_224.in1k': _cfg(hf_hub_id='timm/'),
'pit_ti_distilled_224.in1k': _cfg(
hf_hub_id='timm/',
classifier=('head', 'head_dist')),
'pit_xs_distilled_224.in1k': _cfg(
hf_hub_id='timm/',
classifier=('head', 'head_dist')),
'pit_s_distilled_224.in1k': _cfg(
hf_hub_id='timm/',
classifier=('head', 'head_dist')),
'pit_b_distilled_224.in1k': _cfg(
hf_hub_id='timm/',
classifier=('head', 'head_dist')),
})
@register_model
def pit_b_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=14,
stride=7,
base_dims=[64, 64, 64],
depth=[3, 6, 4],
heads=[4, 8, 16],
mlp_ratio=4,
)
return _create_pit('pit_b_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_s_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[48, 48, 48],
depth=[2, 6, 4],
heads=[3, 6, 12],
mlp_ratio=4,
)
return _create_pit('pit_s_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_xs_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[48, 48, 48],
depth=[2, 6, 4],
heads=[2, 4, 8],
mlp_ratio=4,
)
return _create_pit('pit_xs_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_ti_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[32, 32, 32],
depth=[2, 6, 4],
heads=[2, 4, 8],
mlp_ratio=4,
)
return _create_pit('pit_ti_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_b_distilled_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=14,
stride=7,
base_dims=[64, 64, 64],
depth=[3, 6, 4],
heads=[4, 8, 16],
mlp_ratio=4,
distilled=True,
)
return _create_pit('pit_b_distilled_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_s_distilled_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[48, 48, 48],
depth=[2, 6, 4],
heads=[3, 6, 12],
mlp_ratio=4,
distilled=True,
)
return _create_pit('pit_s_distilled_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_xs_distilled_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[48, 48, 48],
depth=[2, 6, 4],
heads=[2, 4, 8],
mlp_ratio=4,
distilled=True,
)
return _create_pit('pit_xs_distilled_224', pretrained, **dict(model_args, **kwargs))
@register_model
def pit_ti_distilled_224(pretrained=False, **kwargs) -> PoolingVisionTransformer:
model_args = dict(
patch_size=16,
stride=8,
base_dims=[32, 32, 32],
depth=[2, 6, 4],
heads=[2, 4, 8],
mlp_ratio=4,
distilled=True,
)
return _create_pit('pit_ti_distilled_224', pretrained, **dict(model_args, **kwargs))
| pytorch-image-models/timm/models/pit.py/0 | {
"file_path": "pytorch-image-models/timm/models/pit.py",
"repo_id": "pytorch-image-models",
"token_count": 7347
} | 180 |
""" Swin Transformer
A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows`
- https://arxiv.org/pdf/2103.14030
Code/weights from https://github.com/microsoft/Swin-Transformer, original copyright/license info below
S3 (AutoFormerV2, https://arxiv.org/abs/2111.14725) Swin weights from
- https://github.com/microsoft/Cream/tree/main/AutoFormerV2
Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
"""
# --------------------------------------------------------
# Swin Transformer
# Copyright (c) 2021 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ze Liu
# --------------------------------------------------------
import logging
import math
from typing import Callable, List, Optional, Tuple, Union
import torch
import torch.nn as nn
from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
from timm.layers import PatchEmbed, Mlp, DropPath, ClassifierHead, to_2tuple, to_ntuple, trunc_normal_, \
_assert, use_fused_attn, resize_rel_pos_bias_table, resample_patch_embed, ndgrid
from ._builder import build_model_with_cfg
from ._features_fx import register_notrace_function
from ._manipulate import checkpoint_seq, named_apply
from ._registry import generate_default_cfgs, register_model, register_model_deprecations
from .vision_transformer import get_init_weights_vit
__all__ = ['SwinTransformer'] # model_registry will add each entrypoint fn to this
_logger = logging.getLogger(__name__)
_int_or_tuple_2_t = Union[int, Tuple[int, int]]
def window_partition(
x: torch.Tensor,
window_size: Tuple[int, int],
) -> torch.Tensor:
"""
Partition into non-overlapping windows with padding if needed.
Args:
x (tensor): input tokens with [B, H, W, C].
window_size (int): window size.
Returns:
windows: windows after partition with [B * num_windows, window_size, window_size, C].
(Hp, Wp): padded height and width before partition
"""
B, H, W, C = x.shape
x = x.view(B, H // window_size[0], window_size[0], W // window_size[1], window_size[1], C)
windows = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, window_size[0], window_size[1], C)
return windows
@register_notrace_function # reason: int argument is a Proxy
def window_reverse(windows, window_size: Tuple[int, int], H: int, W: int):
"""
Args:
windows: (num_windows*B, window_size, window_size, C)
window_size (int): Window size
H (int): Height of image
W (int): Width of image
Returns:
x: (B, H, W, C)
"""
C = windows.shape[-1]
x = windows.view(-1, H // window_size[0], W // window_size[1], window_size[0], window_size[1], C)
x = x.permute(0, 1, 3, 2, 4, 5).contiguous().view(-1, H, W, C)
return x
def get_relative_position_index(win_h: int, win_w: int):
# get pair-wise relative position index for each token inside the window
coords = torch.stack(ndgrid(torch.arange(win_h), torch.arange(win_w))) # 2, Wh, Ww
coords_flatten = torch.flatten(coords, 1) # 2, Wh*Ww
relative_coords = coords_flatten[:, :, None] - coords_flatten[:, None, :] # 2, Wh*Ww, Wh*Ww
relative_coords = relative_coords.permute(1, 2, 0).contiguous() # Wh*Ww, Wh*Ww, 2
relative_coords[:, :, 0] += win_h - 1 # shift to start from 0
relative_coords[:, :, 1] += win_w - 1
relative_coords[:, :, 0] *= 2 * win_w - 1
return relative_coords.sum(-1) # Wh*Ww, Wh*Ww
class WindowAttention(nn.Module):
""" Window based multi-head self attention (W-MSA) module with relative position bias.
It supports shifted and non-shifted windows.
"""
fused_attn: torch.jit.Final[bool]
def __init__(
self,
dim: int,
num_heads: int,
head_dim: Optional[int] = None,
window_size: _int_or_tuple_2_t = 7,
qkv_bias: bool = True,
attn_drop: float = 0.,
proj_drop: float = 0.,
):
"""
Args:
dim: Number of input channels.
num_heads: Number of attention heads.
head_dim: Number of channels per head (dim // num_heads if not set)
window_size: The height and width of the window.
qkv_bias: If True, add a learnable bias to query, key, value.
attn_drop: Dropout ratio of attention weight.
proj_drop: Dropout ratio of output.
"""
super().__init__()
self.dim = dim
self.window_size = to_2tuple(window_size) # Wh, Ww
win_h, win_w = self.window_size
self.window_area = win_h * win_w
self.num_heads = num_heads
head_dim = head_dim or dim // num_heads
attn_dim = head_dim * num_heads
self.scale = head_dim ** -0.5
self.fused_attn = use_fused_attn(experimental=True) # NOTE not tested for prime-time yet
# define a parameter table of relative position bias, shape: 2*Wh-1 * 2*Ww-1, nH
self.relative_position_bias_table = nn.Parameter(torch.zeros((2 * win_h - 1) * (2 * win_w - 1), num_heads))
# get pair-wise relative position index for each token inside the window
self.register_buffer("relative_position_index", get_relative_position_index(win_h, win_w), persistent=False)
self.qkv = nn.Linear(dim, attn_dim * 3, bias=qkv_bias)
self.attn_drop = nn.Dropout(attn_drop)
self.proj = nn.Linear(attn_dim, dim)
self.proj_drop = nn.Dropout(proj_drop)
trunc_normal_(self.relative_position_bias_table, std=.02)
self.softmax = nn.Softmax(dim=-1)
def _get_rel_pos_bias(self) -> torch.Tensor:
relative_position_bias = self.relative_position_bias_table[
self.relative_position_index.view(-1)].view(self.window_area, self.window_area, -1) # Wh*Ww,Wh*Ww,nH
relative_position_bias = relative_position_bias.permute(2, 0, 1).contiguous() # nH, Wh*Ww, Wh*Ww
return relative_position_bias.unsqueeze(0)
def forward(self, x, mask: Optional[torch.Tensor] = None):
"""
Args:
x: input features with shape of (num_windows*B, N, C)
mask: (0/-inf) mask with shape of (num_windows, Wh*Ww, Wh*Ww) or None
"""
B_, N, C = x.shape
qkv = self.qkv(x).reshape(B_, N, 3, self.num_heads, -1).permute(2, 0, 3, 1, 4)
q, k, v = qkv.unbind(0)
if self.fused_attn:
attn_mask = self._get_rel_pos_bias()
if mask is not None:
num_win = mask.shape[0]
mask = mask.view(1, num_win, 1, N, N).expand(B_ // num_win, -1, self.num_heads, -1, -1)
attn_mask = attn_mask + mask.reshape(-1, self.num_heads, N, N)
x = torch.nn.functional.scaled_dot_product_attention(
q, k, v,
attn_mask=attn_mask,
dropout_p=self.attn_drop.p if self.training else 0.,
)
else:
q = q * self.scale
attn = q @ k.transpose(-2, -1)
attn = attn + self._get_rel_pos_bias()
if mask is not None:
num_win = mask.shape[0]
attn = attn.view(-1, num_win, self.num_heads, N, N) + mask.unsqueeze(1).unsqueeze(0)
attn = attn.view(-1, self.num_heads, N, N)
attn = self.softmax(attn)
attn = self.attn_drop(attn)
x = attn @ v
x = x.transpose(1, 2).reshape(B_, N, -1)
x = self.proj(x)
x = self.proj_drop(x)
return x
class SwinTransformerBlock(nn.Module):
""" Swin Transformer Block.
"""
def __init__(
self,
dim: int,
input_resolution: _int_or_tuple_2_t,
num_heads: int = 4,
head_dim: Optional[int] = None,
window_size: _int_or_tuple_2_t = 7,
shift_size: int = 0,
mlp_ratio: float = 4.,
qkv_bias: bool = True,
proj_drop: float = 0.,
attn_drop: float = 0.,
drop_path: float = 0.,
act_layer: Callable = nn.GELU,
norm_layer: Callable = nn.LayerNorm,
):
"""
Args:
dim: Number of input channels.
input_resolution: Input resolution.
window_size: Window size.
num_heads: Number of attention heads.
head_dim: Enforce the number of channels per head
shift_size: Shift size for SW-MSA.
mlp_ratio: Ratio of mlp hidden dim to embedding dim.
qkv_bias: If True, add a learnable bias to query, key, value.
proj_drop: Dropout rate.
attn_drop: Attention dropout rate.
drop_path: Stochastic depth rate.
act_layer: Activation layer.
norm_layer: Normalization layer.
"""
super().__init__()
self.dim = dim
self.input_resolution = input_resolution
ws, ss = self._calc_window_shift(window_size, shift_size)
self.window_size: Tuple[int, int] = ws
self.shift_size: Tuple[int, int] = ss
self.window_area = self.window_size[0] * self.window_size[1]
self.mlp_ratio = mlp_ratio
self.norm1 = norm_layer(dim)
self.attn = WindowAttention(
dim,
num_heads=num_heads,
head_dim=head_dim,
window_size=to_2tuple(self.window_size),
qkv_bias=qkv_bias,
attn_drop=attn_drop,
proj_drop=proj_drop,
)
self.drop_path1 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
self.norm2 = norm_layer(dim)
self.mlp = Mlp(
in_features=dim,
hidden_features=int(dim * mlp_ratio),
act_layer=act_layer,
drop=proj_drop,
)
self.drop_path2 = DropPath(drop_path) if drop_path > 0. else nn.Identity()
if any(self.shift_size):
# calculate attention mask for SW-MSA
H, W = self.input_resolution
H = math.ceil(H / self.window_size[0]) * self.window_size[0]
W = math.ceil(W / self.window_size[1]) * self.window_size[1]
img_mask = torch.zeros((1, H, W, 1)) # 1 H W 1
cnt = 0
for h in (
slice(0, -self.window_size[0]),
slice(-self.window_size[0], -self.shift_size[0]),
slice(-self.shift_size[0], None)):
for w in (
slice(0, -self.window_size[1]),
slice(-self.window_size[1], -self.shift_size[1]),
slice(-self.shift_size[1], None)):
img_mask[:, h, w, :] = cnt
cnt += 1
mask_windows = window_partition(img_mask, self.window_size) # nW, window_size, window_size, 1
mask_windows = mask_windows.view(-1, self.window_area)
attn_mask = mask_windows.unsqueeze(1) - mask_windows.unsqueeze(2)
attn_mask = attn_mask.masked_fill(attn_mask != 0, float(-100.0)).masked_fill(attn_mask == 0, float(0.0))
else:
attn_mask = None
self.register_buffer("attn_mask", attn_mask, persistent=False)
def _calc_window_shift(self, target_window_size, target_shift_size) -> Tuple[Tuple[int, int], Tuple[int, int]]:
target_window_size = to_2tuple(target_window_size)
target_shift_size = to_2tuple(target_shift_size)
window_size = [r if r <= w else w for r, w in zip(self.input_resolution, target_window_size)]
shift_size = [0 if r <= w else s for r, w, s in zip(self.input_resolution, window_size, target_shift_size)]
return tuple(window_size), tuple(shift_size)
def _attn(self, x):
B, H, W, C = x.shape
# cyclic shift
has_shift = any(self.shift_size)
if has_shift:
shifted_x = torch.roll(x, shifts=(-self.shift_size[0], -self.shift_size[1]), dims=(1, 2))
else:
shifted_x = x
# pad for resolution not divisible by window size
pad_h = (self.window_size[0] - H % self.window_size[0]) % self.window_size[0]
pad_w = (self.window_size[1] - W % self.window_size[1]) % self.window_size[1]
shifted_x = torch.nn.functional.pad(shifted_x, (0, 0, 0, pad_w, 0, pad_h))
Hp, Wp = H + pad_h, W + pad_w
# partition windows
x_windows = window_partition(shifted_x, self.window_size) # nW*B, window_size, window_size, C
x_windows = x_windows.view(-1, self.window_area, C) # nW*B, window_size*window_size, C
# W-MSA/SW-MSA
attn_windows = self.attn(x_windows, mask=self.attn_mask) # nW*B, window_size*window_size, C
# merge windows
attn_windows = attn_windows.view(-1, self.window_size[0], self.window_size[1], C)
shifted_x = window_reverse(attn_windows, self.window_size, Hp, Wp) # B H' W' C
shifted_x = shifted_x[:, :H, :W, :].contiguous()
# reverse cyclic shift
if has_shift:
x = torch.roll(shifted_x, shifts=self.shift_size, dims=(1, 2))
else:
x = shifted_x
return x
def forward(self, x):
B, H, W, C = x.shape
x = x + self.drop_path1(self._attn(self.norm1(x)))
x = x.reshape(B, -1, C)
x = x + self.drop_path2(self.mlp(self.norm2(x)))
x = x.reshape(B, H, W, C)
return x
class PatchMerging(nn.Module):
""" Patch Merging Layer.
"""
def __init__(
self,
dim: int,
out_dim: Optional[int] = None,
norm_layer: Callable = nn.LayerNorm,
):
"""
Args:
dim: Number of input channels.
out_dim: Number of output channels (or 2 * dim if None)
norm_layer: Normalization layer.
"""
super().__init__()
self.dim = dim
self.out_dim = out_dim or 2 * dim
self.norm = norm_layer(4 * dim)
self.reduction = nn.Linear(4 * dim, self.out_dim, bias=False)
def forward(self, x):
B, H, W, C = x.shape
_assert(H % 2 == 0, f"x height ({H}) is not even.")
_assert(W % 2 == 0, f"x width ({W}) is not even.")
x = x.reshape(B, H // 2, 2, W // 2, 2, C).permute(0, 1, 3, 4, 2, 5).flatten(3)
x = self.norm(x)
x = self.reduction(x)
return x
class SwinTransformerStage(nn.Module):
""" A basic Swin Transformer layer for one stage.
"""
def __init__(
self,
dim: int,
out_dim: int,
input_resolution: Tuple[int, int],
depth: int,
downsample: bool = True,
num_heads: int = 4,
head_dim: Optional[int] = None,
window_size: _int_or_tuple_2_t = 7,
mlp_ratio: float = 4.,
qkv_bias: bool = True,
proj_drop: float = 0.,
attn_drop: float = 0.,
drop_path: Union[List[float], float] = 0.,
norm_layer: Callable = nn.LayerNorm,
):
"""
Args:
dim: Number of input channels.
out_dim: Number of output channels.
input_resolution: Input resolution.
depth: Number of blocks.
downsample: Downsample layer at the end of the layer.
num_heads: Number of attention heads.
head_dim: Channels per head (dim // num_heads if not set)
window_size: Local window size.
mlp_ratio: Ratio of mlp hidden dim to embedding dim.
qkv_bias: If True, add a learnable bias to query, key, value.
proj_drop: Projection dropout rate.
attn_drop: Attention dropout rate.
drop_path: Stochastic depth rate.
norm_layer: Normalization layer.
"""
super().__init__()
self.dim = dim
self.input_resolution = input_resolution
self.output_resolution = tuple(i // 2 for i in input_resolution) if downsample else input_resolution
self.depth = depth
self.grad_checkpointing = False
window_size = to_2tuple(window_size)
shift_size = tuple([w // 2 for w in window_size])
# patch merging layer
if downsample:
self.downsample = PatchMerging(
dim=dim,
out_dim=out_dim,
norm_layer=norm_layer,
)
else:
assert dim == out_dim
self.downsample = nn.Identity()
# build blocks
self.blocks = nn.Sequential(*[
SwinTransformerBlock(
dim=out_dim,
input_resolution=self.output_resolution,
num_heads=num_heads,
head_dim=head_dim,
window_size=window_size,
shift_size=0 if (i % 2 == 0) else shift_size,
mlp_ratio=mlp_ratio,
qkv_bias=qkv_bias,
proj_drop=proj_drop,
attn_drop=attn_drop,
drop_path=drop_path[i] if isinstance(drop_path, list) else drop_path,
norm_layer=norm_layer,
)
for i in range(depth)])
def forward(self, x):
x = self.downsample(x)
if self.grad_checkpointing and not torch.jit.is_scripting():
x = checkpoint_seq(self.blocks, x)
else:
x = self.blocks(x)
return x
class SwinTransformer(nn.Module):
""" Swin Transformer
A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shifted Windows` -
https://arxiv.org/pdf/2103.14030
"""
def __init__(
self,
img_size: _int_or_tuple_2_t = 224,
patch_size: int = 4,
in_chans: int = 3,
num_classes: int = 1000,
global_pool: str = 'avg',
embed_dim: int = 96,
depths: Tuple[int, ...] = (2, 2, 6, 2),
num_heads: Tuple[int, ...] = (3, 6, 12, 24),
head_dim: Optional[int] = None,
window_size: _int_or_tuple_2_t = 7,
mlp_ratio: float = 4.,
qkv_bias: bool = True,
drop_rate: float = 0.,
proj_drop_rate: float = 0.,
attn_drop_rate: float = 0.,
drop_path_rate: float = 0.1,
embed_layer: Callable = PatchEmbed,
norm_layer: Union[str, Callable] = nn.LayerNorm,
weight_init: str = '',
**kwargs,
):
"""
Args:
img_size: Input image size.
patch_size: Patch size.
in_chans: Number of input image channels.
num_classes: Number of classes for classification head.
embed_dim: Patch embedding dimension.
depths: Depth of each Swin Transformer layer.
num_heads: Number of attention heads in different layers.
head_dim: Dimension of self-attention heads.
window_size: Window size.
mlp_ratio: Ratio of mlp hidden dim to embedding dim.
qkv_bias: If True, add a learnable bias to query, key, value.
drop_rate: Dropout rate.
attn_drop_rate (float): Attention dropout rate.
drop_path_rate (float): Stochastic depth rate.
embed_layer: Patch embedding layer.
norm_layer (nn.Module): Normalization layer.
"""
super().__init__()
assert global_pool in ('', 'avg')
self.num_classes = num_classes
self.global_pool = global_pool
self.output_fmt = 'NHWC'
self.num_layers = len(depths)
self.embed_dim = embed_dim
self.num_features = int(embed_dim * 2 ** (self.num_layers - 1))
self.feature_info = []
if not isinstance(embed_dim, (tuple, list)):
embed_dim = [int(embed_dim * 2 ** i) for i in range(self.num_layers)]
# split image into non-overlapping patches
self.patch_embed = embed_layer(
img_size=img_size,
patch_size=patch_size,
in_chans=in_chans,
embed_dim=embed_dim[0],
norm_layer=norm_layer,
output_fmt='NHWC',
)
self.patch_grid = self.patch_embed.grid_size
# build layers
head_dim = to_ntuple(self.num_layers)(head_dim)
if not isinstance(window_size, (list, tuple)):
window_size = to_ntuple(self.num_layers)(window_size)
elif len(window_size) == 2:
window_size = (window_size,) * self.num_layers
assert len(window_size) == self.num_layers
mlp_ratio = to_ntuple(self.num_layers)(mlp_ratio)
dpr = [x.tolist() for x in torch.linspace(0, drop_path_rate, sum(depths)).split(depths)]
layers = []
in_dim = embed_dim[0]
scale = 1
for i in range(self.num_layers):
out_dim = embed_dim[i]
layers += [SwinTransformerStage(
dim=in_dim,
out_dim=out_dim,
input_resolution=(
self.patch_grid[0] // scale,
self.patch_grid[1] // scale
),
depth=depths[i],
downsample=i > 0,
num_heads=num_heads[i],
head_dim=head_dim[i],
window_size=window_size[i],
mlp_ratio=mlp_ratio[i],
qkv_bias=qkv_bias,
proj_drop=proj_drop_rate,
attn_drop=attn_drop_rate,
drop_path=dpr[i],
norm_layer=norm_layer,
)]
in_dim = out_dim
if i > 0:
scale *= 2
self.feature_info += [dict(num_chs=out_dim, reduction=4 * scale, module=f'layers.{i}')]
self.layers = nn.Sequential(*layers)
self.norm = norm_layer(self.num_features)
self.head = ClassifierHead(
self.num_features,
num_classes,
pool_type=global_pool,
drop_rate=drop_rate,
input_fmt=self.output_fmt,
)
if weight_init != 'skip':
self.init_weights(weight_init)
@torch.jit.ignore
def init_weights(self, mode=''):
assert mode in ('jax', 'jax_nlhb', 'moco', '')
head_bias = -math.log(self.num_classes) if 'nlhb' in mode else 0.
named_apply(get_init_weights_vit(mode, head_bias=head_bias), self)
@torch.jit.ignore
def no_weight_decay(self):
nwd = set()
for n, _ in self.named_parameters():
if 'relative_position_bias_table' in n:
nwd.add(n)
return nwd
@torch.jit.ignore
def group_matcher(self, coarse=False):
return dict(
stem=r'^patch_embed', # stem and embed
blocks=r'^layers\.(\d+)' if coarse else [
(r'^layers\.(\d+).downsample', (0,)),
(r'^layers\.(\d+)\.\w+\.(\d+)', None),
(r'^norm', (99999,)),
]
)
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
for l in self.layers:
l.grad_checkpointing = enable
@torch.jit.ignore
def get_classifier(self):
return self.head.fc
def reset_classifier(self, num_classes, global_pool=None):
self.num_classes = num_classes
self.head.reset(num_classes, pool_type=global_pool)
def forward_features(self, x):
x = self.patch_embed(x)
x = self.layers(x)
x = self.norm(x)
return x
def forward_head(self, x, pre_logits: bool = False):
return self.head(x, pre_logits=True) if pre_logits else self.head(x)
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def checkpoint_filter_fn(state_dict, model):
""" convert patch embedding weight from manual patchify + linear proj to conv"""
old_weights = True
if 'head.fc.weight' in state_dict:
old_weights = False
import re
out_dict = {}
state_dict = state_dict.get('model', state_dict)
state_dict = state_dict.get('state_dict', state_dict)
for k, v in state_dict.items():
if any([n in k for n in ('relative_position_index', 'attn_mask')]):
continue # skip buffers that should not be persistent
if 'patch_embed.proj.weight' in k:
_, _, H, W = model.patch_embed.proj.weight.shape
if v.shape[-2] != H or v.shape[-1] != W:
v = resample_patch_embed(
v,
(H, W),
interpolation='bicubic',
antialias=True,
verbose=True,
)
if k.endswith('relative_position_bias_table'):
m = model.get_submodule(k[:-29])
if v.shape != m.relative_position_bias_table.shape or m.window_size[0] != m.window_size[1]:
v = resize_rel_pos_bias_table(
v,
new_window_size=m.window_size,
new_bias_shape=m.relative_position_bias_table.shape,
)
if old_weights:
k = re.sub(r'layers.(\d+).downsample', lambda x: f'layers.{int(x.group(1)) + 1}.downsample', k)
k = k.replace('head.', 'head.fc.')
out_dict[k] = v
return out_dict
def _create_swin_transformer(variant, pretrained=False, **kwargs):
default_out_indices = tuple(i for i, _ in enumerate(kwargs.get('depths', (1, 1, 3, 1))))
out_indices = kwargs.pop('out_indices', default_out_indices)
model = build_model_with_cfg(
SwinTransformer, variant, pretrained,
pretrained_filter_fn=checkpoint_filter_fn,
feature_cfg=dict(flatten_sequential=True, out_indices=out_indices),
**kwargs)
return model
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': (7, 7),
'crop_pct': .9, 'interpolation': 'bicubic', 'fixed_input_size': True,
'mean': IMAGENET_DEFAULT_MEAN, 'std': IMAGENET_DEFAULT_STD,
'first_conv': 'patch_embed.proj', 'classifier': 'head.fc',
'license': 'mit', **kwargs
}
default_cfgs = generate_default_cfgs({
'swin_small_patch4_window7_224.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.8/swin_small_patch4_window7_224_22kto1k_finetune.pth', ),
'swin_base_patch4_window7_224.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224_22kto1k.pth',),
'swin_base_patch4_window12_384.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window12_384_22kto1k.pth',
input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
'swin_large_patch4_window7_224.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_large_patch4_window7_224_22kto1k.pth',),
'swin_large_patch4_window12_384.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_large_patch4_window12_384_22kto1k.pth',
input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
'swin_tiny_patch4_window7_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_tiny_patch4_window7_224.pth',),
'swin_small_patch4_window7_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_small_patch4_window7_224.pth',),
'swin_base_patch4_window7_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224.pth',),
'swin_base_patch4_window12_384.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window12_384.pth',
input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
# tiny 22k pretrain is worse than 1k, so moved after (untagged priority is based on order)
'swin_tiny_patch4_window7_224.ms_in22k_ft_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.8/swin_tiny_patch4_window7_224_22kto1k_finetune.pth',),
'swin_tiny_patch4_window7_224.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.8/swin_tiny_patch4_window7_224_22k.pth',
num_classes=21841),
'swin_small_patch4_window7_224.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.8/swin_small_patch4_window7_224_22k.pth',
num_classes=21841),
'swin_base_patch4_window7_224.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224_22k.pth',
num_classes=21841),
'swin_base_patch4_window12_384.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window12_384_22k.pth',
input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0, num_classes=21841),
'swin_large_patch4_window7_224.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_large_patch4_window7_224_22k.pth',
num_classes=21841),
'swin_large_patch4_window12_384.ms_in22k': _cfg(
hf_hub_id='timm/',
url='https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_large_patch4_window12_384_22k.pth',
input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0, num_classes=21841),
'swin_s3_tiny_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/s3_t-1d53f6a8.pth'),
'swin_s3_small_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/s3_s-3bb4c69d.pth'),
'swin_s3_base_224.ms_in1k': _cfg(
hf_hub_id='timm/',
url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/s3_b-a1e95db4.pth'),
})
@register_model
def swin_tiny_patch4_window7_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-T @ 224x224, trained ImageNet-1k
"""
model_args = dict(patch_size=4, window_size=7, embed_dim=96, depths=(2, 2, 6, 2), num_heads=(3, 6, 12, 24))
return _create_swin_transformer(
'swin_tiny_patch4_window7_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_small_patch4_window7_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-S @ 224x224
"""
model_args = dict(patch_size=4, window_size=7, embed_dim=96, depths=(2, 2, 18, 2), num_heads=(3, 6, 12, 24))
return _create_swin_transformer(
'swin_small_patch4_window7_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_base_patch4_window7_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-B @ 224x224
"""
model_args = dict(patch_size=4, window_size=7, embed_dim=128, depths=(2, 2, 18, 2), num_heads=(4, 8, 16, 32))
return _create_swin_transformer(
'swin_base_patch4_window7_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_base_patch4_window12_384(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-B @ 384x384
"""
model_args = dict(patch_size=4, window_size=12, embed_dim=128, depths=(2, 2, 18, 2), num_heads=(4, 8, 16, 32))
return _create_swin_transformer(
'swin_base_patch4_window12_384', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_large_patch4_window7_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-L @ 224x224
"""
model_args = dict(patch_size=4, window_size=7, embed_dim=192, depths=(2, 2, 18, 2), num_heads=(6, 12, 24, 48))
return _create_swin_transformer(
'swin_large_patch4_window7_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_large_patch4_window12_384(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-L @ 384x384
"""
model_args = dict(patch_size=4, window_size=12, embed_dim=192, depths=(2, 2, 18, 2), num_heads=(6, 12, 24, 48))
return _create_swin_transformer(
'swin_large_patch4_window12_384', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_s3_tiny_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-S3-T @ 224x224, https://arxiv.org/abs/2111.14725
"""
model_args = dict(
patch_size=4, window_size=(7, 7, 14, 7), embed_dim=96, depths=(2, 2, 6, 2), num_heads=(3, 6, 12, 24))
return _create_swin_transformer('swin_s3_tiny_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_s3_small_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-S3-S @ 224x224, https://arxiv.org/abs/2111.14725
"""
model_args = dict(
patch_size=4, window_size=(14, 14, 14, 7), embed_dim=96, depths=(2, 2, 18, 2), num_heads=(3, 6, 12, 24))
return _create_swin_transformer('swin_s3_small_224', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def swin_s3_base_224(pretrained=False, **kwargs) -> SwinTransformer:
""" Swin-S3-B @ 224x224, https://arxiv.org/abs/2111.14725
"""
model_args = dict(
patch_size=4, window_size=(7, 7, 14, 7), embed_dim=96, depths=(2, 2, 30, 2), num_heads=(3, 6, 12, 24))
return _create_swin_transformer('swin_s3_base_224', pretrained=pretrained, **dict(model_args, **kwargs))
register_model_deprecations(__name__, {
'swin_base_patch4_window7_224_in22k': 'swin_base_patch4_window7_224.ms_in22k',
'swin_base_patch4_window12_384_in22k': 'swin_base_patch4_window12_384.ms_in22k',
'swin_large_patch4_window7_224_in22k': 'swin_large_patch4_window7_224.ms_in22k',
'swin_large_patch4_window12_384_in22k': 'swin_large_patch4_window12_384.ms_in22k',
})
| pytorch-image-models/timm/models/swin_transformer.py/0 | {
"file_path": "pytorch-image-models/timm/models/swin_transformer.py",
"repo_id": "pytorch-image-models",
"token_count": 16908
} | 181 |
"""Pytorch impl of Aligned Xception 41, 65, 71
This is a correct, from scratch impl of Aligned Xception (Deeplab) models compatible with TF weights at
https://github.com/tensorflow/models/blob/master/research/deeplab/g3doc/model_zoo.md
Hacked together by / Copyright 2020 Ross Wightman
"""
from functools import partial
from typing import List, Dict, Type, Optional
import torch
import torch.nn as nn
from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
from timm.layers import ClassifierHead, ConvNormAct, DropPath, PadType, create_conv2d, get_norm_act_layer
from timm.layers.helpers import to_3tuple
from ._builder import build_model_with_cfg
from ._manipulate import checkpoint_seq
from ._registry import register_model, generate_default_cfgs
__all__ = ['XceptionAligned']
class SeparableConv2d(nn.Module):
def __init__(
self,
in_chs: int,
out_chs: int,
kernel_size: int = 3,
stride: int = 1,
dilation: int = 1,
padding: PadType = '',
act_layer: Type[nn.Module] = nn.ReLU,
norm_layer: Type[nn.Module] = nn.BatchNorm2d,
):
super(SeparableConv2d, self).__init__()
self.kernel_size = kernel_size
self.dilation = dilation
# depthwise convolution
self.conv_dw = create_conv2d(
in_chs, in_chs, kernel_size, stride=stride,
padding=padding, dilation=dilation, depthwise=True)
self.bn_dw = norm_layer(in_chs)
self.act_dw = act_layer(inplace=True) if act_layer is not None else nn.Identity()
# pointwise convolution
self.conv_pw = create_conv2d(in_chs, out_chs, kernel_size=1)
self.bn_pw = norm_layer(out_chs)
self.act_pw = act_layer(inplace=True) if act_layer is not None else nn.Identity()
def forward(self, x):
x = self.conv_dw(x)
x = self.bn_dw(x)
x = self.act_dw(x)
x = self.conv_pw(x)
x = self.bn_pw(x)
x = self.act_pw(x)
return x
class PreSeparableConv2d(nn.Module):
def __init__(
self,
in_chs: int,
out_chs: int,
kernel_size: int = 3,
stride: int = 1,
dilation: int = 1,
padding: PadType = '',
act_layer: Type[nn.Module] = nn.ReLU,
norm_layer: Type[nn.Module] = nn.BatchNorm2d,
first_act: bool = True,
):
super(PreSeparableConv2d, self).__init__()
norm_act_layer = get_norm_act_layer(norm_layer, act_layer=act_layer)
self.kernel_size = kernel_size
self.dilation = dilation
self.norm = norm_act_layer(in_chs, inplace=True) if first_act else nn.Identity()
# depthwise convolution
self.conv_dw = create_conv2d(
in_chs, in_chs, kernel_size, stride=stride,
padding=padding, dilation=dilation, depthwise=True)
# pointwise convolution
self.conv_pw = create_conv2d(in_chs, out_chs, kernel_size=1)
def forward(self, x):
x = self.norm(x)
x = self.conv_dw(x)
x = self.conv_pw(x)
return x
class XceptionModule(nn.Module):
def __init__(
self,
in_chs: int,
out_chs: int,
stride: int = 1,
dilation: int = 1,
pad_type: PadType = '',
start_with_relu: bool = True,
no_skip: bool = False,
act_layer: Type[nn.Module] = nn.ReLU,
norm_layer: Optional[Type[nn.Module]] = None,
drop_path: Optional[nn.Module] = None
):
super(XceptionModule, self).__init__()
out_chs = to_3tuple(out_chs)
self.in_channels = in_chs
self.out_channels = out_chs[-1]
self.no_skip = no_skip
if not no_skip and (self.out_channels != self.in_channels or stride != 1):
self.shortcut = ConvNormAct(
in_chs, self.out_channels, 1, stride=stride, norm_layer=norm_layer, apply_act=False)
else:
self.shortcut = None
separable_act_layer = None if start_with_relu else act_layer
self.stack = nn.Sequential()
for i in range(3):
if start_with_relu:
self.stack.add_module(f'act{i + 1}', act_layer(inplace=i > 0))
self.stack.add_module(f'conv{i + 1}', SeparableConv2d(
in_chs, out_chs[i], 3, stride=stride if i == 2 else 1, dilation=dilation, padding=pad_type,
act_layer=separable_act_layer, norm_layer=norm_layer))
in_chs = out_chs[i]
self.drop_path = drop_path
def forward(self, x):
skip = x
x = self.stack(x)
if self.shortcut is not None:
skip = self.shortcut(skip)
if not self.no_skip:
if self.drop_path is not None:
x = self.drop_path(x)
x = x + skip
return x
class PreXceptionModule(nn.Module):
def __init__(
self,
in_chs: int,
out_chs: int,
stride: int = 1,
dilation: int = 1,
pad_type: PadType = '',
no_skip: bool = False,
act_layer: Type[nn.Module] = nn.ReLU,
norm_layer: Optional[Type[nn.Module]] = None,
drop_path: Optional[nn.Module] = None
):
super(PreXceptionModule, self).__init__()
out_chs = to_3tuple(out_chs)
self.in_channels = in_chs
self.out_channels = out_chs[-1]
self.no_skip = no_skip
if not no_skip and (self.out_channels != self.in_channels or stride != 1):
self.shortcut = create_conv2d(in_chs, self.out_channels, 1, stride=stride)
else:
self.shortcut = nn.Identity()
self.norm = get_norm_act_layer(norm_layer, act_layer=act_layer)(in_chs, inplace=True)
self.stack = nn.Sequential()
for i in range(3):
self.stack.add_module(f'conv{i + 1}', PreSeparableConv2d(
in_chs,
out_chs[i],
3,
stride=stride if i == 2 else 1,
dilation=dilation,
padding=pad_type,
act_layer=act_layer,
norm_layer=norm_layer,
first_act=i > 0,
))
in_chs = out_chs[i]
self.drop_path = drop_path
def forward(self, x):
x = self.norm(x)
skip = x
x = self.stack(x)
if not self.no_skip:
if self.drop_path is not None:
x = self.drop_path(x)
x = x + self.shortcut(skip)
return x
class XceptionAligned(nn.Module):
"""Modified Aligned Xception
"""
def __init__(
self,
block_cfg: List[Dict],
num_classes: int = 1000,
in_chans: int = 3,
output_stride: int = 32,
preact: bool = False,
act_layer: Type[nn.Module] = nn.ReLU,
norm_layer: Type[nn.Module] = nn.BatchNorm2d,
drop_rate: float = 0.,
drop_path_rate: float = 0.,
global_pool: str = 'avg',
):
super(XceptionAligned, self).__init__()
assert output_stride in (8, 16, 32)
self.num_classes = num_classes
self.drop_rate = drop_rate
self.grad_checkpointing = False
layer_args = dict(act_layer=act_layer, norm_layer=norm_layer)
self.stem = nn.Sequential(*[
ConvNormAct(in_chans, 32, kernel_size=3, stride=2, **layer_args),
create_conv2d(32, 64, kernel_size=3, stride=1) if preact else
ConvNormAct(32, 64, kernel_size=3, stride=1, **layer_args)
])
curr_dilation = 1
curr_stride = 2
self.feature_info = []
self.blocks = nn.Sequential()
module_fn = PreXceptionModule if preact else XceptionModule
net_num_blocks = len(block_cfg)
net_block_idx = 0
for i, b in enumerate(block_cfg):
block_dpr = drop_path_rate * net_block_idx / (net_num_blocks - 1) # stochastic depth linear decay rule
b['drop_path'] = DropPath(block_dpr) if block_dpr > 0. else None
b['dilation'] = curr_dilation
if b['stride'] > 1:
name = f'blocks.{i}.stack.conv2' if preact else f'blocks.{i}.stack.act3'
self.feature_info += [dict(num_chs=to_3tuple(b['out_chs'])[-2], reduction=curr_stride, module=name)]
next_stride = curr_stride * b['stride']
if next_stride > output_stride:
curr_dilation *= b['stride']
b['stride'] = 1
else:
curr_stride = next_stride
self.blocks.add_module(str(i), module_fn(**b, **layer_args))
self.num_features = self.blocks[-1].out_channels
net_block_idx += 1
self.feature_info += [dict(
num_chs=self.num_features, reduction=curr_stride, module='blocks.' + str(len(self.blocks) - 1))]
self.act = act_layer(inplace=True) if preact else nn.Identity()
self.head = ClassifierHead(
in_features=self.num_features,
num_classes=num_classes,
pool_type=global_pool,
drop_rate=drop_rate,
)
@torch.jit.ignore
def group_matcher(self, coarse=False):
return dict(
stem=r'^stem',
blocks=r'^blocks\.(\d+)',
)
@torch.jit.ignore
def set_grad_checkpointing(self, enable=True):
self.grad_checkpointing = enable
@torch.jit.ignore
def get_classifier(self):
return self.head.fc
def reset_classifier(self, num_classes, global_pool='avg'):
self.head.reset(num_classes, pool_type=global_pool)
def forward_features(self, x):
x = self.stem(x)
if self.grad_checkpointing and not torch.jit.is_scripting():
x = checkpoint_seq(self.blocks, x)
else:
x = self.blocks(x)
x = self.act(x)
return x
def forward_head(self, x, pre_logits: bool = False):
return self.head(x, pre_logits=pre_logits)
def forward(self, x):
x = self.forward_features(x)
x = self.forward_head(x)
return x
def _xception(variant, pretrained=False, **kwargs):
return build_model_with_cfg(
XceptionAligned,
variant,
pretrained,
feature_cfg=dict(flatten_sequential=True, feature_cls='hook'),
**kwargs,
)
def _cfg(url='', **kwargs):
return {
'url': url,
'num_classes': 1000, 'input_size': (3, 299, 299), 'pool_size': (10, 10),
'crop_pct': 0.903, 'interpolation': 'bicubic',
'mean': IMAGENET_INCEPTION_MEAN, 'std': IMAGENET_INCEPTION_STD,
'first_conv': 'stem.0.conv', 'classifier': 'head.fc',
**kwargs
}
default_cfgs = generate_default_cfgs({
'xception65.ra3_in1k': _cfg(
hf_hub_id='timm/',
crop_pct=0.94,
),
'xception41.tf_in1k': _cfg(hf_hub_id='timm/'),
'xception65.tf_in1k': _cfg(hf_hub_id='timm/'),
'xception71.tf_in1k': _cfg(hf_hub_id='timm/'),
'xception41p.ra3_in1k': _cfg(
hf_hub_id='timm/',
crop_pct=0.94,
),
'xception65p.ra3_in1k': _cfg(
hf_hub_id='timm/',
crop_pct=0.94,
),
})
@register_model
def xception41(pretrained=False, **kwargs) -> XceptionAligned:
""" Modified Aligned Xception-41
"""
block_cfg = [
# entry flow
dict(in_chs=64, out_chs=128, stride=2),
dict(in_chs=128, out_chs=256, stride=2),
dict(in_chs=256, out_chs=728, stride=2),
# middle flow
*([dict(in_chs=728, out_chs=728, stride=1)] * 8),
# exit flow
dict(in_chs=728, out_chs=(728, 1024, 1024), stride=2),
dict(in_chs=1024, out_chs=(1536, 1536, 2048), stride=1, no_skip=True, start_with_relu=False),
]
model_args = dict(block_cfg=block_cfg, norm_layer=partial(nn.BatchNorm2d, eps=.001, momentum=.1))
return _xception('xception41', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def xception65(pretrained=False, **kwargs) -> XceptionAligned:
""" Modified Aligned Xception-65
"""
block_cfg = [
# entry flow
dict(in_chs=64, out_chs=128, stride=2),
dict(in_chs=128, out_chs=256, stride=2),
dict(in_chs=256, out_chs=728, stride=2),
# middle flow
*([dict(in_chs=728, out_chs=728, stride=1)] * 16),
# exit flow
dict(in_chs=728, out_chs=(728, 1024, 1024), stride=2),
dict(in_chs=1024, out_chs=(1536, 1536, 2048), stride=1, no_skip=True, start_with_relu=False),
]
model_args = dict(block_cfg=block_cfg, norm_layer=partial(nn.BatchNorm2d, eps=.001, momentum=.1))
return _xception('xception65', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def xception71(pretrained=False, **kwargs) -> XceptionAligned:
""" Modified Aligned Xception-71
"""
block_cfg = [
# entry flow
dict(in_chs=64, out_chs=128, stride=2),
dict(in_chs=128, out_chs=256, stride=1),
dict(in_chs=256, out_chs=256, stride=2),
dict(in_chs=256, out_chs=728, stride=1),
dict(in_chs=728, out_chs=728, stride=2),
# middle flow
*([dict(in_chs=728, out_chs=728, stride=1)] * 16),
# exit flow
dict(in_chs=728, out_chs=(728, 1024, 1024), stride=2),
dict(in_chs=1024, out_chs=(1536, 1536, 2048), stride=1, no_skip=True, start_with_relu=False),
]
model_args = dict(block_cfg=block_cfg, norm_layer=partial(nn.BatchNorm2d, eps=.001, momentum=.1))
return _xception('xception71', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def xception41p(pretrained=False, **kwargs) -> XceptionAligned:
""" Modified Aligned Xception-41 w/ Pre-Act
"""
block_cfg = [
# entry flow
dict(in_chs=64, out_chs=128, stride=2),
dict(in_chs=128, out_chs=256, stride=2),
dict(in_chs=256, out_chs=728, stride=2),
# middle flow
*([dict(in_chs=728, out_chs=728, stride=1)] * 8),
# exit flow
dict(in_chs=728, out_chs=(728, 1024, 1024), stride=2),
dict(in_chs=1024, out_chs=(1536, 1536, 2048), no_skip=True, stride=1),
]
model_args = dict(block_cfg=block_cfg, preact=True, norm_layer=nn.BatchNorm2d)
return _xception('xception41p', pretrained=pretrained, **dict(model_args, **kwargs))
@register_model
def xception65p(pretrained=False, **kwargs) -> XceptionAligned:
""" Modified Aligned Xception-65 w/ Pre-Act
"""
block_cfg = [
# entry flow
dict(in_chs=64, out_chs=128, stride=2),
dict(in_chs=128, out_chs=256, stride=2),
dict(in_chs=256, out_chs=728, stride=2),
# middle flow
*([dict(in_chs=728, out_chs=728, stride=1)] * 16),
# exit flow
dict(in_chs=728, out_chs=(728, 1024, 1024), stride=2),
dict(in_chs=1024, out_chs=(1536, 1536, 2048), stride=1, no_skip=True),
]
model_args = dict(
block_cfg=block_cfg, preact=True, norm_layer=partial(nn.BatchNorm2d, eps=.001, momentum=.1))
return _xception('xception65p', pretrained=pretrained, **dict(model_args, **kwargs))
| pytorch-image-models/timm/models/xception_aligned.py/0 | {
"file_path": "pytorch-image-models/timm/models/xception_aligned.py",
"repo_id": "pytorch-image-models",
"token_count": 7719
} | 182 |
""" Nvidia NovoGrad Optimizer.
Original impl by Nvidia from Jasper example:
- https://github.com/NVIDIA/DeepLearningExamples/blob/master/PyTorch/SpeechRecognition/Jasper
Paper: `Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks`
- https://arxiv.org/abs/1905.11286
"""
import torch
from torch.optim.optimizer import Optimizer
import math
class NvNovoGrad(Optimizer):
"""
Implements Novograd algorithm.
Args:
params (iterable): iterable of parameters to optimize or dicts defining
parameter groups
lr (float, optional): learning rate (default: 1e-3)
betas (Tuple[float, float], optional): coefficients used for computing
running averages of gradient and its square (default: (0.95, 0.98))
eps (float, optional): term added to the denominator to improve
numerical stability (default: 1e-8)
weight_decay (float, optional): weight decay (L2 penalty) (default: 0)
grad_averaging: gradient averaging
amsgrad (boolean, optional): whether to use the AMSGrad variant of this
algorithm from the paper `On the Convergence of Adam and Beyond`_
(default: False)
"""
def __init__(self, params, lr=1e-3, betas=(0.95, 0.98), eps=1e-8,
weight_decay=0, grad_averaging=False, amsgrad=False):
if not 0.0 <= lr:
raise ValueError("Invalid learning rate: {}".format(lr))
if not 0.0 <= eps:
raise ValueError("Invalid epsilon value: {}".format(eps))
if not 0.0 <= betas[0] < 1.0:
raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
if not 0.0 <= betas[1] < 1.0:
raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
defaults = dict(lr=lr, betas=betas, eps=eps,
weight_decay=weight_decay,
grad_averaging=grad_averaging,
amsgrad=amsgrad)
super(NvNovoGrad, self).__init__(params, defaults)
def __setstate__(self, state):
super(NvNovoGrad, self).__setstate__(state)
for group in self.param_groups:
group.setdefault('amsgrad', False)
@torch.no_grad()
def step(self, closure=None):
"""Performs a single optimization step.
Arguments:
closure (callable, optional): A closure that reevaluates the model
and returns the loss.
"""
loss = None
if closure is not None:
with torch.enable_grad():
loss = closure()
for group in self.param_groups:
for p in group['params']:
if p.grad is None:
continue
grad = p.grad
if grad.is_sparse:
raise RuntimeError('Sparse gradients are not supported.')
amsgrad = group['amsgrad']
state = self.state[p]
# State initialization
if len(state) == 0:
state['step'] = 0
# Exponential moving average of gradient values
state['exp_avg'] = torch.zeros_like(p)
# Exponential moving average of squared gradient values
state['exp_avg_sq'] = torch.zeros([]).to(state['exp_avg'].device)
if amsgrad:
# Maintains max of all exp. moving avg. of sq. grad. values
state['max_exp_avg_sq'] = torch.zeros([]).to(state['exp_avg'].device)
exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
if amsgrad:
max_exp_avg_sq = state['max_exp_avg_sq']
beta1, beta2 = group['betas']
state['step'] += 1
norm = torch.sum(torch.pow(grad, 2))
if exp_avg_sq == 0:
exp_avg_sq.copy_(norm)
else:
exp_avg_sq.mul_(beta2).add_(norm, alpha=1 - beta2)
if amsgrad:
# Maintains the maximum of all 2nd moment running avg. till now
torch.max(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
# Use the max. for normalizing running avg. of gradient
denom = max_exp_avg_sq.sqrt().add_(group['eps'])
else:
denom = exp_avg_sq.sqrt().add_(group['eps'])
grad.div_(denom)
if group['weight_decay'] != 0:
grad.add_(p, alpha=group['weight_decay'])
if group['grad_averaging']:
grad.mul_(1 - beta1)
exp_avg.mul_(beta1).add_(grad)
p.add_(exp_avg, alpha=-group['lr'])
return loss
| pytorch-image-models/timm/optim/nvnovograd.py/0 | {
"file_path": "pytorch-image-models/timm/optim/nvnovograd.py",
"repo_id": "pytorch-image-models",
"token_count": 2415
} | 183 |
""" Adaptive Gradient Clipping
An impl of AGC, as per (https://arxiv.org/abs/2102.06171):
@article{brock2021high,
author={Andrew Brock and Soham De and Samuel L. Smith and Karen Simonyan},
title={High-Performance Large-Scale Image Recognition Without Normalization},
journal={arXiv preprint arXiv:},
year={2021}
}
Code references:
* Official JAX impl (paper authors): https://github.com/deepmind/deepmind-research/tree/master/nfnets
* Phil Wang's PyTorch gist: https://gist.github.com/lucidrains/0d6560077edac419ab5d3aa29e674d5c
Hacked together by / Copyright 2021 Ross Wightman
"""
import torch
def unitwise_norm(x, norm_type=2.0):
if x.ndim <= 1:
return x.norm(norm_type)
else:
# works for nn.ConvNd and nn,Linear where output dim is first in the kernel/weight tensor
# might need special cases for other weights (possibly MHA) where this may not be true
return x.norm(norm_type, dim=tuple(range(1, x.ndim)), keepdim=True)
def adaptive_clip_grad(parameters, clip_factor=0.01, eps=1e-3, norm_type=2.0):
if isinstance(parameters, torch.Tensor):
parameters = [parameters]
for p in parameters:
if p.grad is None:
continue
p_data = p.detach()
g_data = p.grad.detach()
max_norm = unitwise_norm(p_data, norm_type=norm_type).clamp_(min=eps).mul_(clip_factor)
grad_norm = unitwise_norm(g_data, norm_type=norm_type)
clipped_grad = g_data * (max_norm / grad_norm.clamp(min=1e-6))
new_grads = torch.where(grad_norm < max_norm, g_data, clipped_grad)
p.grad.detach().copy_(new_grads)
| pytorch-image-models/timm/utils/agc.py/0 | {
"file_path": "pytorch-image-models/timm/utils/agc.py",
"repo_id": "pytorch-image-models",
"token_count": 661
} | 184 |
#!/usr/bin/env python3
""" ImageNet Training Script
This is intended to be a lean and easily modifiable ImageNet training script that reproduces ImageNet
training results with some of the latest networks and training techniques. It favours canonical PyTorch
and standard Python style over trying to be able to 'do it all.' That said, it offers quite a few speed
and training result improvements over the usual PyTorch example scripts. Repurpose as you see fit.
This script was started from an early version of the PyTorch ImageNet example
(https://github.com/pytorch/examples/tree/master/imagenet)
NVIDIA CUDA specific speedups adopted from NVIDIA Apex examples
(https://github.com/NVIDIA/apex/tree/master/examples/imagenet)
Hacked together by / Copyright 2020 Ross Wightman (https://github.com/rwightman)
"""
import argparse
import json
import logging
import os
import time
from collections import OrderedDict
from contextlib import suppress
from datetime import datetime
from functools import partial
import torch
import torch.nn as nn
import torchvision.utils
import yaml
from torch.nn.parallel import DistributedDataParallel as NativeDDP
from timm import utils
from timm.data import create_dataset, create_loader, resolve_data_config, Mixup, FastCollateMixup, AugMixDataset
from timm.layers import convert_splitbn_model, convert_sync_batchnorm, set_fast_norm
from timm.loss import JsdCrossEntropy, SoftTargetCrossEntropy, BinaryCrossEntropy, LabelSmoothingCrossEntropy
from timm.models import create_model, safe_model_name, resume_checkpoint, load_checkpoint, model_parameters
from timm.optim import create_optimizer_v2, optimizer_kwargs
from timm.scheduler import create_scheduler_v2, scheduler_kwargs
from timm.utils import ApexScaler, NativeScaler
try:
from apex import amp
from apex.parallel import DistributedDataParallel as ApexDDP
from apex.parallel import convert_syncbn_model
has_apex = True
except ImportError:
has_apex = False
has_native_amp = False
try:
if getattr(torch.cuda.amp, 'autocast') is not None:
has_native_amp = True
except AttributeError:
pass
try:
import wandb
has_wandb = True
except ImportError:
has_wandb = False
try:
from functorch.compile import memory_efficient_fusion
has_functorch = True
except ImportError as e:
has_functorch = False
has_compile = hasattr(torch, 'compile')
_logger = logging.getLogger('train')
# The first arg parser parses out only the --config argument, this argument is used to
# load a yaml file containing key-values that override the defaults for the main parser below
config_parser = parser = argparse.ArgumentParser(description='Training Config', add_help=False)
parser.add_argument('-c', '--config', default='', type=str, metavar='FILE',
help='YAML config file specifying default arguments')
parser = argparse.ArgumentParser(description='PyTorch ImageNet Training')
# Dataset parameters
group = parser.add_argument_group('Dataset parameters')
# Keep this argument outside the dataset group because it is positional.
parser.add_argument('data', nargs='?', metavar='DIR', const=None,
help='path to dataset (positional is *deprecated*, use --data-dir)')
parser.add_argument('--data-dir', metavar='DIR',
help='path to dataset (root dir)')
parser.add_argument('--dataset', metavar='NAME', default='',
help='dataset type + name ("<type>/<name>") (default: ImageFolder or ImageTar if empty)')
group.add_argument('--train-split', metavar='NAME', default='train',
help='dataset train split (default: train)')
group.add_argument('--val-split', metavar='NAME', default='validation',
help='dataset validation split (default: validation)')
parser.add_argument('--train-num-samples', default=None, type=int,
metavar='N', help='Manually specify num samples in train split, for IterableDatasets.')
parser.add_argument('--val-num-samples', default=None, type=int,
metavar='N', help='Manually specify num samples in validation split, for IterableDatasets.')
group.add_argument('--dataset-download', action='store_true', default=False,
help='Allow download of dataset for torch/ and tfds/ datasets that support it.')
group.add_argument('--class-map', default='', type=str, metavar='FILENAME',
help='path to class to idx mapping file (default: "")')
group.add_argument('--input-img-mode', default=None, type=str,
help='Dataset image conversion mode for input images.')
group.add_argument('--input-key', default=None, type=str,
help='Dataset key for input images.')
group.add_argument('--target-key', default=None, type=str,
help='Dataset key for target labels.')
# Model parameters
group = parser.add_argument_group('Model parameters')
group.add_argument('--model', default='resnet50', type=str, metavar='MODEL',
help='Name of model to train (default: "resnet50")')
group.add_argument('--pretrained', action='store_true', default=False,
help='Start with pretrained version of specified network (if avail)')
group.add_argument('--pretrained-path', default=None, type=str,
help='Load this checkpoint as if they were the pretrained weights (with adaptation).')
group.add_argument('--initial-checkpoint', default='', type=str, metavar='PATH',
help='Load this checkpoint into model after initialization (default: none)')
group.add_argument('--resume', default='', type=str, metavar='PATH',
help='Resume full model and optimizer state from checkpoint (default: none)')
group.add_argument('--no-resume-opt', action='store_true', default=False,
help='prevent resume of optimizer state when resuming model')
group.add_argument('--num-classes', type=int, default=None, metavar='N',
help='number of label classes (Model default if None)')
group.add_argument('--gp', default=None, type=str, metavar='POOL',
help='Global pool type, one of (fast, avg, max, avgmax, avgmaxc). Model default if None.')
group.add_argument('--img-size', type=int, default=None, metavar='N',
help='Image size (default: None => model default)')
group.add_argument('--in-chans', type=int, default=None, metavar='N',
help='Image input channels (default: None => 3)')
group.add_argument('--input-size', default=None, nargs=3, type=int,
metavar='N N N',
help='Input all image dimensions (d h w, e.g. --input-size 3 224 224), uses model default if empty')
group.add_argument('--crop-pct', default=None, type=float,
metavar='N', help='Input image center crop percent (for validation only)')
group.add_argument('--mean', type=float, nargs='+', default=None, metavar='MEAN',
help='Override mean pixel value of dataset')
group.add_argument('--std', type=float, nargs='+', default=None, metavar='STD',
help='Override std deviation of dataset')
group.add_argument('--interpolation', default='', type=str, metavar='NAME',
help='Image resize interpolation type (overrides model)')
group.add_argument('-b', '--batch-size', type=int, default=128, metavar='N',
help='Input batch size for training (default: 128)')
group.add_argument('-vb', '--validation-batch-size', type=int, default=None, metavar='N',
help='Validation batch size override (default: None)')
group.add_argument('--channels-last', action='store_true', default=False,
help='Use channels_last memory layout')
group.add_argument('--fuser', default='', type=str,
help="Select jit fuser. One of ('', 'te', 'old', 'nvfuser')")
group.add_argument('--grad-accum-steps', type=int, default=1, metavar='N',
help='The number of steps to accumulate gradients (default: 1)')
group.add_argument('--grad-checkpointing', action='store_true', default=False,
help='Enable gradient checkpointing through model blocks/stages')
group.add_argument('--fast-norm', default=False, action='store_true',
help='enable experimental fast-norm')
group.add_argument('--model-kwargs', nargs='*', default={}, action=utils.ParseKwargs)
group.add_argument('--head-init-scale', default=None, type=float,
help='Head initialization scale')
group.add_argument('--head-init-bias', default=None, type=float,
help='Head initialization bias value')
# scripting / codegen
scripting_group = group.add_mutually_exclusive_group()
scripting_group.add_argument('--torchscript', dest='torchscript', action='store_true',
help='torch.jit.script the full model')
scripting_group.add_argument('--torchcompile', nargs='?', type=str, default=None, const='inductor',
help="Enable compilation w/ specified backend (default: inductor).")
# Optimizer parameters
group = parser.add_argument_group('Optimizer parameters')
group.add_argument('--opt', default='sgd', type=str, metavar='OPTIMIZER',
help='Optimizer (default: "sgd")')
group.add_argument('--opt-eps', default=None, type=float, metavar='EPSILON',
help='Optimizer Epsilon (default: None, use opt default)')
group.add_argument('--opt-betas', default=None, type=float, nargs='+', metavar='BETA',
help='Optimizer Betas (default: None, use opt default)')
group.add_argument('--momentum', type=float, default=0.9, metavar='M',
help='Optimizer momentum (default: 0.9)')
group.add_argument('--weight-decay', type=float, default=2e-5,
help='weight decay (default: 2e-5)')
group.add_argument('--clip-grad', type=float, default=None, metavar='NORM',
help='Clip gradient norm (default: None, no clipping)')
group.add_argument('--clip-mode', type=str, default='norm',
help='Gradient clipping mode. One of ("norm", "value", "agc")')
group.add_argument('--layer-decay', type=float, default=None,
help='layer-wise learning rate decay (default: None)')
group.add_argument('--opt-kwargs', nargs='*', default={}, action=utils.ParseKwargs)
# Learning rate schedule parameters
group = parser.add_argument_group('Learning rate schedule parameters')
group.add_argument('--sched', type=str, default='cosine', metavar='SCHEDULER',
help='LR scheduler (default: "step"')
group.add_argument('--sched-on-updates', action='store_true', default=False,
help='Apply LR scheduler step on update instead of epoch end.')
group.add_argument('--lr', type=float, default=None, metavar='LR',
help='learning rate, overrides lr-base if set (default: None)')
group.add_argument('--lr-base', type=float, default=0.1, metavar='LR',
help='base learning rate: lr = lr_base * global_batch_size / base_size')
group.add_argument('--lr-base-size', type=int, default=256, metavar='DIV',
help='base learning rate batch size (divisor, default: 256).')
group.add_argument('--lr-base-scale', type=str, default='', metavar='SCALE',
help='base learning rate vs batch_size scaling ("linear", "sqrt", based on opt if empty)')
group.add_argument('--lr-noise', type=float, nargs='+', default=None, metavar='pct, pct',
help='learning rate noise on/off epoch percentages')
group.add_argument('--lr-noise-pct', type=float, default=0.67, metavar='PERCENT',
help='learning rate noise limit percent (default: 0.67)')
group.add_argument('--lr-noise-std', type=float, default=1.0, metavar='STDDEV',
help='learning rate noise std-dev (default: 1.0)')
group.add_argument('--lr-cycle-mul', type=float, default=1.0, metavar='MULT',
help='learning rate cycle len multiplier (default: 1.0)')
group.add_argument('--lr-cycle-decay', type=float, default=0.5, metavar='MULT',
help='amount to decay each learning rate cycle (default: 0.5)')
group.add_argument('--lr-cycle-limit', type=int, default=1, metavar='N',
help='learning rate cycle limit, cycles enabled if > 1')
group.add_argument('--lr-k-decay', type=float, default=1.0,
help='learning rate k-decay for cosine/poly (default: 1.0)')
group.add_argument('--warmup-lr', type=float, default=1e-5, metavar='LR',
help='warmup learning rate (default: 1e-5)')
group.add_argument('--min-lr', type=float, default=0, metavar='LR',
help='lower lr bound for cyclic schedulers that hit 0 (default: 0)')
group.add_argument('--epochs', type=int, default=300, metavar='N',
help='number of epochs to train (default: 300)')
group.add_argument('--epoch-repeats', type=float, default=0., metavar='N',
help='epoch repeat multiplier (number of times to repeat dataset epoch per train epoch).')
group.add_argument('--start-epoch', default=None, type=int, metavar='N',
help='manual epoch number (useful on restarts)')
group.add_argument('--decay-milestones', default=[90, 180, 270], type=int, nargs='+', metavar="MILESTONES",
help='list of decay epoch indices for multistep lr. must be increasing')
group.add_argument('--decay-epochs', type=float, default=90, metavar='N',
help='epoch interval to decay LR')
group.add_argument('--warmup-epochs', type=int, default=5, metavar='N',
help='epochs to warmup LR, if scheduler supports')
group.add_argument('--warmup-prefix', action='store_true', default=False,
help='Exclude warmup period from decay schedule.'),
group.add_argument('--cooldown-epochs', type=int, default=0, metavar='N',
help='epochs to cooldown LR at min_lr, after cyclic schedule ends')
group.add_argument('--patience-epochs', type=int, default=10, metavar='N',
help='patience epochs for Plateau LR scheduler (default: 10)')
group.add_argument('--decay-rate', '--dr', type=float, default=0.1, metavar='RATE',
help='LR decay rate (default: 0.1)')
# Augmentation & regularization parameters
group = parser.add_argument_group('Augmentation and regularization parameters')
group.add_argument('--no-aug', action='store_true', default=False,
help='Disable all training augmentation, override other train aug args')
group.add_argument('--train-crop-mode', type=str, default=None,
help='Crop-mode in train'),
group.add_argument('--scale', type=float, nargs='+', default=[0.08, 1.0], metavar='PCT',
help='Random resize scale (default: 0.08 1.0)')
group.add_argument('--ratio', type=float, nargs='+', default=[3. / 4., 4. / 3.], metavar='RATIO',
help='Random resize aspect ratio (default: 0.75 1.33)')
group.add_argument('--hflip', type=float, default=0.5,
help='Horizontal flip training aug probability')
group.add_argument('--vflip', type=float, default=0.,
help='Vertical flip training aug probability')
group.add_argument('--color-jitter', type=float, default=0.4, metavar='PCT',
help='Color jitter factor (default: 0.4)')
group.add_argument('--color-jitter-prob', type=float, default=None, metavar='PCT',
help='Probability of applying any color jitter.')
group.add_argument('--grayscale-prob', type=float, default=None, metavar='PCT',
help='Probability of applying random grayscale conversion.')
group.add_argument('--gaussian-blur-prob', type=float, default=None, metavar='PCT',
help='Probability of applying gaussian blur.')
group.add_argument('--aa', type=str, default=None, metavar='NAME',
help='Use AutoAugment policy. "v0" or "original". (default: None)'),
group.add_argument('--aug-repeats', type=float, default=0,
help='Number of augmentation repetitions (distributed training only) (default: 0)')
group.add_argument('--aug-splits', type=int, default=0,
help='Number of augmentation splits (default: 0, valid: 0 or >=2)')
group.add_argument('--jsd-loss', action='store_true', default=False,
help='Enable Jensen-Shannon Divergence + CE loss. Use with `--aug-splits`.')
group.add_argument('--bce-loss', action='store_true', default=False,
help='Enable BCE loss w/ Mixup/CutMix use.')
group.add_argument('--bce-sum', action='store_true', default=False,
help='Sum over classes when using BCE loss.')
group.add_argument('--bce-target-thresh', type=float, default=None,
help='Threshold for binarizing softened BCE targets (default: None, disabled).')
group.add_argument('--bce-pos-weight', type=float, default=None,
help='Positive weighting for BCE loss.')
group.add_argument('--reprob', type=float, default=0., metavar='PCT',
help='Random erase prob (default: 0.)')
group.add_argument('--remode', type=str, default='pixel',
help='Random erase mode (default: "pixel")')
group.add_argument('--recount', type=int, default=1,
help='Random erase count (default: 1)')
group.add_argument('--resplit', action='store_true', default=False,
help='Do not random erase first (clean) augmentation split')
group.add_argument('--mixup', type=float, default=0.0,
help='mixup alpha, mixup enabled if > 0. (default: 0.)')
group.add_argument('--cutmix', type=float, default=0.0,
help='cutmix alpha, cutmix enabled if > 0. (default: 0.)')
group.add_argument('--cutmix-minmax', type=float, nargs='+', default=None,
help='cutmix min/max ratio, overrides alpha and enables cutmix if set (default: None)')
group.add_argument('--mixup-prob', type=float, default=1.0,
help='Probability of performing mixup or cutmix when either/both is enabled')
group.add_argument('--mixup-switch-prob', type=float, default=0.5,
help='Probability of switching to cutmix when both mixup and cutmix enabled')
group.add_argument('--mixup-mode', type=str, default='batch',
help='How to apply mixup/cutmix params. Per "batch", "pair", or "elem"')
group.add_argument('--mixup-off-epoch', default=0, type=int, metavar='N',
help='Turn off mixup after this epoch, disabled if 0 (default: 0)')
group.add_argument('--smoothing', type=float, default=0.1,
help='Label smoothing (default: 0.1)')
group.add_argument('--train-interpolation', type=str, default='random',
help='Training interpolation (random, bilinear, bicubic default: "random")')
group.add_argument('--drop', type=float, default=0.0, metavar='PCT',
help='Dropout rate (default: 0.)')
group.add_argument('--drop-connect', type=float, default=None, metavar='PCT',
help='Drop connect rate, DEPRECATED, use drop-path (default: None)')
group.add_argument('--drop-path', type=float, default=None, metavar='PCT',
help='Drop path rate (default: None)')
group.add_argument('--drop-block', type=float, default=None, metavar='PCT',
help='Drop block rate (default: None)')
# Batch norm parameters (only works with gen_efficientnet based models currently)
group = parser.add_argument_group('Batch norm parameters', 'Only works with gen_efficientnet based models currently.')
group.add_argument('--bn-momentum', type=float, default=None,
help='BatchNorm momentum override (if not None)')
group.add_argument('--bn-eps', type=float, default=None,
help='BatchNorm epsilon override (if not None)')
group.add_argument('--sync-bn', action='store_true',
help='Enable NVIDIA Apex or Torch synchronized BatchNorm.')
group.add_argument('--dist-bn', type=str, default='reduce',
help='Distribute BatchNorm stats between nodes after each epoch ("broadcast", "reduce", or "")')
group.add_argument('--split-bn', action='store_true',
help='Enable separate BN layers per augmentation split.')
# Model Exponential Moving Average
group = parser.add_argument_group('Model exponential moving average parameters')
group.add_argument('--model-ema', action='store_true', default=False,
help='Enable tracking moving average of model weights')
group.add_argument('--model-ema-force-cpu', action='store_true', default=False,
help='Force ema to be tracked on CPU, rank=0 node only. Disables EMA validation.')
group.add_argument('--model-ema-decay', type=float, default=0.9998,
help='decay factor for model weights moving average (default: 0.9998)')
# Misc
group = parser.add_argument_group('Miscellaneous parameters')
group.add_argument('--seed', type=int, default=42, metavar='S',
help='random seed (default: 42)')
group.add_argument('--worker-seeding', type=str, default='all',
help='worker seed mode (default: all)')
group.add_argument('--log-interval', type=int, default=50, metavar='N',
help='how many batches to wait before logging training status')
group.add_argument('--recovery-interval', type=int, default=0, metavar='N',
help='how many batches to wait before writing recovery checkpoint')
group.add_argument('--checkpoint-hist', type=int, default=10, metavar='N',
help='number of checkpoints to keep (default: 10)')
group.add_argument('-j', '--workers', type=int, default=4, metavar='N',
help='how many training processes to use (default: 4)')
group.add_argument('--save-images', action='store_true', default=False,
help='save images of input bathes every log interval for debugging')
group.add_argument('--amp', action='store_true', default=False,
help='use NVIDIA Apex AMP or Native AMP for mixed precision training')
group.add_argument('--amp-dtype', default='float16', type=str,
help='lower precision AMP dtype (default: float16)')
group.add_argument('--amp-impl', default='native', type=str,
help='AMP impl to use, "native" or "apex" (default: native)')
group.add_argument('--no-ddp-bb', action='store_true', default=False,
help='Force broadcast buffers for native DDP to off.')
group.add_argument('--synchronize-step', action='store_true', default=False,
help='torch.cuda.synchronize() end of each step')
group.add_argument('--pin-mem', action='store_true', default=False,
help='Pin CPU memory in DataLoader for more efficient (sometimes) transfer to GPU.')
group.add_argument('--no-prefetcher', action='store_true', default=False,
help='disable fast prefetcher')
group.add_argument('--output', default='', type=str, metavar='PATH',
help='path to output folder (default: none, current dir)')
group.add_argument('--experiment', default='', type=str, metavar='NAME',
help='name of train experiment, name of sub-folder for output')
group.add_argument('--eval-metric', default='top1', type=str, metavar='EVAL_METRIC',
help='Best metric (default: "top1"')
group.add_argument('--tta', type=int, default=0, metavar='N',
help='Test/inference time augmentation (oversampling) factor. 0=None (default: 0)')
group.add_argument("--local_rank", default=0, type=int)
group.add_argument('--use-multi-epochs-loader', action='store_true', default=False,
help='use the multi-epochs-loader to save time at the beginning of every epoch')
group.add_argument('--log-wandb', action='store_true', default=False,
help='log training and validation metrics to wandb')
def _parse_args():
# Do we have a config file to parse?
args_config, remaining = config_parser.parse_known_args()
if args_config.config:
with open(args_config.config, 'r') as f:
cfg = yaml.safe_load(f)
parser.set_defaults(**cfg)
# The main arg parser parses the rest of the args, the usual
# defaults will have been overridden if config file specified.
args = parser.parse_args(remaining)
# Cache the args as a text string to save them in the output dir later
args_text = yaml.safe_dump(args.__dict__, default_flow_style=False)
return args, args_text
def main():
utils.setup_default_logging()
args, args_text = _parse_args()
if torch.cuda.is_available():
torch.backends.cuda.matmul.allow_tf32 = True
torch.backends.cudnn.benchmark = True
args.prefetcher = not args.no_prefetcher
args.grad_accum_steps = max(1, args.grad_accum_steps)
device = utils.init_distributed_device(args)
if args.distributed:
_logger.info(
'Training in distributed mode with multiple processes, 1 device per process.'
f'Process {args.rank}, total {args.world_size}, device {args.device}.')
else:
_logger.info(f'Training with a single process on 1 device ({args.device}).')
assert args.rank >= 0
# resolve AMP arguments based on PyTorch / Apex availability
use_amp = None
amp_dtype = torch.float16
if args.amp:
if args.amp_impl == 'apex':
assert has_apex, 'AMP impl specified as APEX but APEX is not installed.'
use_amp = 'apex'
assert args.amp_dtype == 'float16'
else:
assert has_native_amp, 'Please update PyTorch to a version with native AMP (or use APEX).'
use_amp = 'native'
assert args.amp_dtype in ('float16', 'bfloat16')
if args.amp_dtype == 'bfloat16':
amp_dtype = torch.bfloat16
utils.random_seed(args.seed, args.rank)
if args.fuser:
utils.set_jit_fuser(args.fuser)
if args.fast_norm:
set_fast_norm()
in_chans = 3
if args.in_chans is not None:
in_chans = args.in_chans
elif args.input_size is not None:
in_chans = args.input_size[0]
factory_kwargs = {}
if args.pretrained_path:
# merge with pretrained_cfg of model, 'file' has priority over 'url' and 'hf_hub'.
factory_kwargs['pretrained_cfg_overlay'] = dict(
file=args.pretrained_path,
num_classes=-1, # force head adaptation
)
model = create_model(
args.model,
pretrained=args.pretrained,
in_chans=in_chans,
num_classes=args.num_classes,
drop_rate=args.drop,
drop_path_rate=args.drop_path,
drop_block_rate=args.drop_block,
global_pool=args.gp,
bn_momentum=args.bn_momentum,
bn_eps=args.bn_eps,
scriptable=args.torchscript,
checkpoint_path=args.initial_checkpoint,
**factory_kwargs,
**args.model_kwargs,
)
if args.head_init_scale is not None:
with torch.no_grad():
model.get_classifier().weight.mul_(args.head_init_scale)
model.get_classifier().bias.mul_(args.head_init_scale)
if args.head_init_bias is not None:
nn.init.constant_(model.get_classifier().bias, args.head_init_bias)
if args.num_classes is None:
assert hasattr(model, 'num_classes'), 'Model must have `num_classes` attr if not set on cmd line/config.'
args.num_classes = model.num_classes # FIXME handle model default vs config num_classes more elegantly
if args.grad_checkpointing:
model.set_grad_checkpointing(enable=True)
if utils.is_primary(args):
_logger.info(
f'Model {safe_model_name(args.model)} created, param count:{sum([m.numel() for m in model.parameters()])}')
data_config = resolve_data_config(vars(args), model=model, verbose=utils.is_primary(args))
# setup augmentation batch splits for contrastive loss or split bn
num_aug_splits = 0
if args.aug_splits > 0:
assert args.aug_splits > 1, 'A split of 1 makes no sense'
num_aug_splits = args.aug_splits
# enable split bn (separate bn stats per batch-portion)
if args.split_bn:
assert num_aug_splits > 1 or args.resplit
model = convert_splitbn_model(model, max(num_aug_splits, 2))
# move model to GPU, enable channels last layout if set
model.to(device=device)
if args.channels_last:
model.to(memory_format=torch.channels_last)
# setup synchronized BatchNorm for distributed training
if args.distributed and args.sync_bn:
args.dist_bn = '' # disable dist_bn when sync BN active
assert not args.split_bn
if has_apex and use_amp == 'apex':
# Apex SyncBN used with Apex AMP
# WARNING this won't currently work with models using BatchNormAct2d
model = convert_syncbn_model(model)
else:
model = convert_sync_batchnorm(model)
if utils.is_primary(args):
_logger.info(
'Converted model to use Synchronized BatchNorm. WARNING: You may have issues if using '
'zero initialized BN layers (enabled by default for ResNets) while sync-bn enabled.')
if args.torchscript:
assert not args.torchcompile
assert not use_amp == 'apex', 'Cannot use APEX AMP with torchscripted model'
assert not args.sync_bn, 'Cannot use SyncBatchNorm with torchscripted model'
model = torch.jit.script(model)
if not args.lr:
global_batch_size = args.batch_size * args.world_size * args.grad_accum_steps
batch_ratio = global_batch_size / args.lr_base_size
if not args.lr_base_scale:
on = args.opt.lower()
args.lr_base_scale = 'sqrt' if any([o in on for o in ('ada', 'lamb')]) else 'linear'
if args.lr_base_scale == 'sqrt':
batch_ratio = batch_ratio ** 0.5
args.lr = args.lr_base * batch_ratio
if utils.is_primary(args):
_logger.info(
f'Learning rate ({args.lr}) calculated from base learning rate ({args.lr_base}) '
f'and effective global batch size ({global_batch_size}) with {args.lr_base_scale} scaling.')
optimizer = create_optimizer_v2(
model,
**optimizer_kwargs(cfg=args),
**args.opt_kwargs,
)
# setup automatic mixed-precision (AMP) loss scaling and op casting
amp_autocast = suppress # do nothing
loss_scaler = None
if use_amp == 'apex':
assert device.type == 'cuda'
model, optimizer = amp.initialize(model, optimizer, opt_level='O1')
loss_scaler = ApexScaler()
if utils.is_primary(args):
_logger.info('Using NVIDIA APEX AMP. Training in mixed precision.')
elif use_amp == 'native':
try:
amp_autocast = partial(torch.autocast, device_type=device.type, dtype=amp_dtype)
except (AttributeError, TypeError):
# fallback to CUDA only AMP for PyTorch < 1.10
assert device.type == 'cuda'
amp_autocast = torch.cuda.amp.autocast
if device.type == 'cuda' and amp_dtype == torch.float16:
# loss scaler only used for float16 (half) dtype, bfloat16 does not need it
loss_scaler = NativeScaler()
if utils.is_primary(args):
_logger.info('Using native Torch AMP. Training in mixed precision.')
else:
if utils.is_primary(args):
_logger.info('AMP not enabled. Training in float32.')
# optionally resume from a checkpoint
resume_epoch = None
if args.resume:
resume_epoch = resume_checkpoint(
model,
args.resume,
optimizer=None if args.no_resume_opt else optimizer,
loss_scaler=None if args.no_resume_opt else loss_scaler,
log_info=utils.is_primary(args),
)
# setup exponential moving average of model weights, SWA could be used here too
model_ema = None
if args.model_ema:
# Important to create EMA model after cuda(), DP wrapper, and AMP but before DDP wrapper
model_ema = utils.ModelEmaV2(
model, decay=args.model_ema_decay, device='cpu' if args.model_ema_force_cpu else None)
if args.resume:
load_checkpoint(model_ema.module, args.resume, use_ema=True)
# setup distributed training
if args.distributed:
if has_apex and use_amp == 'apex':
# Apex DDP preferred unless native amp is activated
if utils.is_primary(args):
_logger.info("Using NVIDIA APEX DistributedDataParallel.")
model = ApexDDP(model, delay_allreduce=True)
else:
if utils.is_primary(args):
_logger.info("Using native Torch DistributedDataParallel.")
model = NativeDDP(model, device_ids=[device], broadcast_buffers=not args.no_ddp_bb)
# NOTE: EMA model does not need to be wrapped by DDP
if args.torchcompile:
# torch compile should be done after DDP
assert has_compile, 'A version of torch w/ torch.compile() is required for --compile, possibly a nightly.'
model = torch.compile(model, backend=args.torchcompile)
# create the train and eval datasets
if args.data and not args.data_dir:
args.data_dir = args.data
if args.input_img_mode is None:
input_img_mode = 'RGB' if data_config['input_size'][0] == 3 else 'L'
else:
input_img_mode = args.input_img_mode
dataset_train = create_dataset(
args.dataset,
root=args.data_dir,
split=args.train_split,
is_training=True,
class_map=args.class_map,
download=args.dataset_download,
batch_size=args.batch_size,
seed=args.seed,
repeats=args.epoch_repeats,
input_img_mode=input_img_mode,
input_key=args.input_key,
target_key=args.target_key,
num_samples=args.train_num_samples,
)
if args.val_split:
dataset_eval = create_dataset(
args.dataset,
root=args.data_dir,
split=args.val_split,
is_training=False,
class_map=args.class_map,
download=args.dataset_download,
batch_size=args.batch_size,
input_img_mode=input_img_mode,
input_key=args.input_key,
target_key=args.target_key,
num_samples=args.val_num_samples,
)
# setup mixup / cutmix
collate_fn = None
mixup_fn = None
mixup_active = args.mixup > 0 or args.cutmix > 0. or args.cutmix_minmax is not None
if mixup_active:
mixup_args = dict(
mixup_alpha=args.mixup,
cutmix_alpha=args.cutmix,
cutmix_minmax=args.cutmix_minmax,
prob=args.mixup_prob,
switch_prob=args.mixup_switch_prob,
mode=args.mixup_mode,
label_smoothing=args.smoothing,
num_classes=args.num_classes
)
if args.prefetcher:
assert not num_aug_splits # collate conflict (need to support deinterleaving in collate mixup)
collate_fn = FastCollateMixup(**mixup_args)
else:
mixup_fn = Mixup(**mixup_args)
# wrap dataset in AugMix helper
if num_aug_splits > 1:
dataset_train = AugMixDataset(dataset_train, num_splits=num_aug_splits)
# create data loaders w/ augmentation pipeiine
train_interpolation = args.train_interpolation
if args.no_aug or not train_interpolation:
train_interpolation = data_config['interpolation']
loader_train = create_loader(
dataset_train,
input_size=data_config['input_size'],
batch_size=args.batch_size,
is_training=True,
no_aug=args.no_aug,
re_prob=args.reprob,
re_mode=args.remode,
re_count=args.recount,
re_split=args.resplit,
train_crop_mode=args.train_crop_mode,
scale=args.scale,
ratio=args.ratio,
hflip=args.hflip,
vflip=args.vflip,
color_jitter=args.color_jitter,
color_jitter_prob=args.color_jitter_prob,
grayscale_prob=args.grayscale_prob,
gaussian_blur_prob=args.gaussian_blur_prob,
auto_augment=args.aa,
num_aug_repeats=args.aug_repeats,
num_aug_splits=num_aug_splits,
interpolation=train_interpolation,
mean=data_config['mean'],
std=data_config['std'],
num_workers=args.workers,
distributed=args.distributed,
collate_fn=collate_fn,
pin_memory=args.pin_mem,
device=device,
use_prefetcher=args.prefetcher,
use_multi_epochs_loader=args.use_multi_epochs_loader,
worker_seeding=args.worker_seeding,
)
loader_eval = None
if args.val_split:
eval_workers = args.workers
if args.distributed and ('tfds' in args.dataset or 'wds' in args.dataset):
# FIXME reduces validation padding issues when using TFDS, WDS w/ workers and distributed training
eval_workers = min(2, args.workers)
loader_eval = create_loader(
dataset_eval,
input_size=data_config['input_size'],
batch_size=args.validation_batch_size or args.batch_size,
is_training=False,
interpolation=data_config['interpolation'],
mean=data_config['mean'],
std=data_config['std'],
num_workers=eval_workers,
distributed=args.distributed,
crop_pct=data_config['crop_pct'],
pin_memory=args.pin_mem,
device=device,
use_prefetcher=args.prefetcher,
)
# setup loss function
if args.jsd_loss:
assert num_aug_splits > 1 # JSD only valid with aug splits set
train_loss_fn = JsdCrossEntropy(num_splits=num_aug_splits, smoothing=args.smoothing)
elif mixup_active:
# smoothing is handled with mixup target transform which outputs sparse, soft targets
if args.bce_loss:
train_loss_fn = BinaryCrossEntropy(
target_threshold=args.bce_target_thresh,
sum_classes=args.bce_sum,
pos_weight=args.bce_pos_weight,
)
else:
train_loss_fn = SoftTargetCrossEntropy()
elif args.smoothing:
if args.bce_loss:
train_loss_fn = BinaryCrossEntropy(
smoothing=args.smoothing,
target_threshold=args.bce_target_thresh,
sum_classes=args.bce_sum,
pos_weight=args.bce_pos_weight,
)
else:
train_loss_fn = LabelSmoothingCrossEntropy(smoothing=args.smoothing)
else:
train_loss_fn = nn.CrossEntropyLoss()
train_loss_fn = train_loss_fn.to(device=device)
validate_loss_fn = nn.CrossEntropyLoss().to(device=device)
# setup checkpoint saver and eval metric tracking
eval_metric = args.eval_metric if loader_eval is not None else 'loss'
decreasing_metric = eval_metric == 'loss'
best_metric = None
best_epoch = None
saver = None
output_dir = None
if utils.is_primary(args):
if args.experiment:
exp_name = args.experiment
else:
exp_name = '-'.join([
datetime.now().strftime("%Y%m%d-%H%M%S"),
safe_model_name(args.model),
str(data_config['input_size'][-1])
])
output_dir = utils.get_outdir(args.output if args.output else './output/train', exp_name)
saver = utils.CheckpointSaver(
model=model,
optimizer=optimizer,
args=args,
model_ema=model_ema,
amp_scaler=loss_scaler,
checkpoint_dir=output_dir,
recovery_dir=output_dir,
decreasing=decreasing_metric,
max_history=args.checkpoint_hist
)
with open(os.path.join(output_dir, 'args.yaml'), 'w') as f:
f.write(args_text)
if utils.is_primary(args) and args.log_wandb:
if has_wandb:
wandb.init(project=args.experiment, config=args)
else:
_logger.warning(
"You've requested to log metrics to wandb but package not found. "
"Metrics not being logged to wandb, try `pip install wandb`")
# setup learning rate schedule and starting epoch
updates_per_epoch = (len(loader_train) + args.grad_accum_steps - 1) // args.grad_accum_steps
lr_scheduler, num_epochs = create_scheduler_v2(
optimizer,
**scheduler_kwargs(args, decreasing_metric=decreasing_metric),
updates_per_epoch=updates_per_epoch,
)
start_epoch = 0
if args.start_epoch is not None:
# a specified start_epoch will always override the resume epoch
start_epoch = args.start_epoch
elif resume_epoch is not None:
start_epoch = resume_epoch
if lr_scheduler is not None and start_epoch > 0:
if args.sched_on_updates:
lr_scheduler.step_update(start_epoch * updates_per_epoch)
else:
lr_scheduler.step(start_epoch)
if utils.is_primary(args):
_logger.info(
f'Scheduled epochs: {num_epochs}. LR stepped per {"epoch" if lr_scheduler.t_in_epochs else "update"}.')
results = []
try:
for epoch in range(start_epoch, num_epochs):
if hasattr(dataset_train, 'set_epoch'):
dataset_train.set_epoch(epoch)
elif args.distributed and hasattr(loader_train.sampler, 'set_epoch'):
loader_train.sampler.set_epoch(epoch)
train_metrics = train_one_epoch(
epoch,
model,
loader_train,
optimizer,
train_loss_fn,
args,
lr_scheduler=lr_scheduler,
saver=saver,
output_dir=output_dir,
amp_autocast=amp_autocast,
loss_scaler=loss_scaler,
model_ema=model_ema,
mixup_fn=mixup_fn,
)
if args.distributed and args.dist_bn in ('broadcast', 'reduce'):
if utils.is_primary(args):
_logger.info("Distributing BatchNorm running means and vars")
utils.distribute_bn(model, args.world_size, args.dist_bn == 'reduce')
if loader_eval is not None:
eval_metrics = validate(
model,
loader_eval,
validate_loss_fn,
args,
amp_autocast=amp_autocast,
)
if model_ema is not None and not args.model_ema_force_cpu:
if args.distributed and args.dist_bn in ('broadcast', 'reduce'):
utils.distribute_bn(model_ema, args.world_size, args.dist_bn == 'reduce')
ema_eval_metrics = validate(
model_ema.module,
loader_eval,
validate_loss_fn,
args,
amp_autocast=amp_autocast,
log_suffix=' (EMA)',
)
eval_metrics = ema_eval_metrics
else:
eval_metrics = None
if output_dir is not None:
lrs = [param_group['lr'] for param_group in optimizer.param_groups]
utils.update_summary(
epoch,
train_metrics,
eval_metrics,
filename=os.path.join(output_dir, 'summary.csv'),
lr=sum(lrs) / len(lrs),
write_header=best_metric is None,
log_wandb=args.log_wandb and has_wandb,
)
if eval_metrics is not None:
latest_metric = eval_metrics[eval_metric]
else:
latest_metric = train_metrics[eval_metric]
if saver is not None:
# save proper checkpoint with eval metric
best_metric, best_epoch = saver.save_checkpoint(epoch, metric=latest_metric)
if lr_scheduler is not None:
# step LR for next epoch
lr_scheduler.step(epoch + 1, latest_metric)
results.append({
'epoch': epoch,
'train': train_metrics,
'validation': eval_metrics,
})
except KeyboardInterrupt:
pass
results = {'all': results}
if best_metric is not None:
results['best'] = results['all'][best_epoch - start_epoch]
_logger.info('*** Best metric: {0} (epoch {1})'.format(best_metric, best_epoch))
print(f'--result\n{json.dumps(results, indent=4)}')
def train_one_epoch(
epoch,
model,
loader,
optimizer,
loss_fn,
args,
device=torch.device('cuda'),
lr_scheduler=None,
saver=None,
output_dir=None,
amp_autocast=suppress,
loss_scaler=None,
model_ema=None,
mixup_fn=None,
):
if args.mixup_off_epoch and epoch >= args.mixup_off_epoch:
if args.prefetcher and loader.mixup_enabled:
loader.mixup_enabled = False
elif mixup_fn is not None:
mixup_fn.mixup_enabled = False
second_order = hasattr(optimizer, 'is_second_order') and optimizer.is_second_order
has_no_sync = hasattr(model, "no_sync")
update_time_m = utils.AverageMeter()
data_time_m = utils.AverageMeter()
losses_m = utils.AverageMeter()
model.train()
accum_steps = args.grad_accum_steps
last_accum_steps = len(loader) % accum_steps
updates_per_epoch = (len(loader) + accum_steps - 1) // accum_steps
num_updates = epoch * updates_per_epoch
last_batch_idx = len(loader) - 1
last_batch_idx_to_accum = len(loader) - last_accum_steps
data_start_time = update_start_time = time.time()
optimizer.zero_grad()
update_sample_count = 0
for batch_idx, (input, target) in enumerate(loader):
last_batch = batch_idx == last_batch_idx
need_update = last_batch or (batch_idx + 1) % accum_steps == 0
update_idx = batch_idx // accum_steps
if batch_idx >= last_batch_idx_to_accum:
accum_steps = last_accum_steps
if not args.prefetcher:
input, target = input.to(device), target.to(device)
if mixup_fn is not None:
input, target = mixup_fn(input, target)
if args.channels_last:
input = input.contiguous(memory_format=torch.channels_last)
# multiply by accum steps to get equivalent for full update
data_time_m.update(accum_steps * (time.time() - data_start_time))
def _forward():
with amp_autocast():
output = model(input)
loss = loss_fn(output, target)
if accum_steps > 1:
loss /= accum_steps
return loss
def _backward(_loss):
if loss_scaler is not None:
loss_scaler(
_loss,
optimizer,
clip_grad=args.clip_grad,
clip_mode=args.clip_mode,
parameters=model_parameters(model, exclude_head='agc' in args.clip_mode),
create_graph=second_order,
need_update=need_update,
)
else:
_loss.backward(create_graph=second_order)
if need_update:
if args.clip_grad is not None:
utils.dispatch_clip_grad(
model_parameters(model, exclude_head='agc' in args.clip_mode),
value=args.clip_grad,
mode=args.clip_mode,
)
optimizer.step()
if has_no_sync and not need_update:
with model.no_sync():
loss = _forward()
_backward(loss)
else:
loss = _forward()
_backward(loss)
if not args.distributed:
losses_m.update(loss.item() * accum_steps, input.size(0))
update_sample_count += input.size(0)
if not need_update:
data_start_time = time.time()
continue
num_updates += 1
optimizer.zero_grad()
if model_ema is not None:
model_ema.update(model)
if args.synchronize_step and device.type == 'cuda':
torch.cuda.synchronize()
time_now = time.time()
update_time_m.update(time.time() - update_start_time)
update_start_time = time_now
if update_idx % args.log_interval == 0:
lrl = [param_group['lr'] for param_group in optimizer.param_groups]
lr = sum(lrl) / len(lrl)
if args.distributed:
reduced_loss = utils.reduce_tensor(loss.data, args.world_size)
losses_m.update(reduced_loss.item() * accum_steps, input.size(0))
update_sample_count *= args.world_size
if utils.is_primary(args):
_logger.info(
f'Train: {epoch} [{update_idx:>4d}/{updates_per_epoch} '
f'({100. * update_idx / (updates_per_epoch - 1):>3.0f}%)] '
f'Loss: {losses_m.val:#.3g} ({losses_m.avg:#.3g}) '
f'Time: {update_time_m.val:.3f}s, {update_sample_count / update_time_m.val:>7.2f}/s '
f'({update_time_m.avg:.3f}s, {update_sample_count / update_time_m.avg:>7.2f}/s) '
f'LR: {lr:.3e} '
f'Data: {data_time_m.val:.3f} ({data_time_m.avg:.3f})'
)
if args.save_images and output_dir:
torchvision.utils.save_image(
input,
os.path.join(output_dir, 'train-batch-%d.jpg' % batch_idx),
padding=0,
normalize=True
)
if saver is not None and args.recovery_interval and (
(update_idx + 1) % args.recovery_interval == 0):
saver.save_recovery(epoch, batch_idx=update_idx)
if lr_scheduler is not None:
lr_scheduler.step_update(num_updates=num_updates, metric=losses_m.avg)
update_sample_count = 0
data_start_time = time.time()
# end for
if hasattr(optimizer, 'sync_lookahead'):
optimizer.sync_lookahead()
return OrderedDict([('loss', losses_m.avg)])
def validate(
model,
loader,
loss_fn,
args,
device=torch.device('cuda'),
amp_autocast=suppress,
log_suffix=''
):
batch_time_m = utils.AverageMeter()
losses_m = utils.AverageMeter()
top1_m = utils.AverageMeter()
top5_m = utils.AverageMeter()
model.eval()
end = time.time()
last_idx = len(loader) - 1
with torch.no_grad():
for batch_idx, (input, target) in enumerate(loader):
last_batch = batch_idx == last_idx
if not args.prefetcher:
input = input.to(device)
target = target.to(device)
if args.channels_last:
input = input.contiguous(memory_format=torch.channels_last)
with amp_autocast():
output = model(input)
if isinstance(output, (tuple, list)):
output = output[0]
# augmentation reduction
reduce_factor = args.tta
if reduce_factor > 1:
output = output.unfold(0, reduce_factor, reduce_factor).mean(dim=2)
target = target[0:target.size(0):reduce_factor]
loss = loss_fn(output, target)
acc1, acc5 = utils.accuracy(output, target, topk=(1, 5))
if args.distributed:
reduced_loss = utils.reduce_tensor(loss.data, args.world_size)
acc1 = utils.reduce_tensor(acc1, args.world_size)
acc5 = utils.reduce_tensor(acc5, args.world_size)
else:
reduced_loss = loss.data
if device.type == 'cuda':
torch.cuda.synchronize()
losses_m.update(reduced_loss.item(), input.size(0))
top1_m.update(acc1.item(), output.size(0))
top5_m.update(acc5.item(), output.size(0))
batch_time_m.update(time.time() - end)
end = time.time()
if utils.is_primary(args) and (last_batch or batch_idx % args.log_interval == 0):
log_name = 'Test' + log_suffix
_logger.info(
f'{log_name}: [{batch_idx:>4d}/{last_idx}] '
f'Time: {batch_time_m.val:.3f} ({batch_time_m.avg:.3f}) '
f'Loss: {losses_m.val:>7.3f} ({losses_m.avg:>6.3f}) '
f'Acc@1: {top1_m.val:>7.3f} ({top1_m.avg:>7.3f}) '
f'Acc@5: {top5_m.val:>7.3f} ({top5_m.avg:>7.3f})'
)
metrics = OrderedDict([('loss', losses_m.avg), ('top1', top1_m.avg), ('top5', top5_m.avg)])
return metrics
if __name__ == '__main__':
main()
| pytorch-image-models/train.py/0 | {
"file_path": "pytorch-image-models/train.py",
"repo_id": "pytorch-image-models",
"token_count": 24011
} | 185 |
install-server:
cd server && make install
install-custom-kernels:
if [ "$$BUILD_EXTENSIONS" = "True" ]; then cd server/custom_kernels && python setup.py install; else echo "Custom kernels are disabled, you need to set the BUILD_EXTENSIONS environment variable to 'True' in order to build them. (Please read the docs, kernels might not work on all hardware)"; fi
install-integration-tests:
cd integration-tests && pip install -r requirements.txt
cd clients/python && pip install .
install-router:
cd router && cargo install --path .
install-launcher:
cd launcher && cargo install --path .
install-benchmark:
cd benchmark && cargo install --path .
install: install-server install-router install-launcher install-custom-kernels
server-dev:
cd server && make run-dev
router-dev:
cd router && cargo run -- --port 8080
rust-tests: install-router install-launcher
cargo test
integration-tests: install-integration-tests
pytest -s -vv -m "not private" integration-tests
update-integration-tests: install-integration-tests
pytest -s -vv --snapshot-update integration-tests
python-server-tests:
HF_HUB_ENABLE_HF_TRANSFER=1 pytest -s -vv -m "not private" server/tests
python-client-tests:
pytest clients/python/tests
python-tests: python-server-tests python-client-tests
run-falcon-7b-instruct:
text-generation-launcher --model-id tiiuae/falcon-7b-instruct --port 8080
run-falcon-7b-instruct-quantize:
text-generation-launcher --model-id tiiuae/falcon-7b-instruct --quantize bitsandbytes --port 8080
clean:
rm -rf target aml
| text-generation-inference/Makefile/0 | {
"file_path": "text-generation-inference/Makefile",
"repo_id": "text-generation-inference",
"token_count": 498
} | 186 |
# Serving Private & Gated Models
If the model you wish to serve is behind gated access or the model repository on Hugging Face Hub is private, and you have access to the model, you can provide your Hugging Face Hub access token. You can generate and copy a read token from [Hugging Face Hub tokens page](https://huggingface.co/settings/tokens)
If you're using the CLI, set the `HUGGING_FACE_HUB_TOKEN` environment variable. For example:
```
export HUGGING_FACE_HUB_TOKEN=<YOUR READ TOKEN>
```
If you would like to do it through Docker, you can provide your token by specifying `HUGGING_FACE_HUB_TOKEN` as shown below.
```bash
model=meta-llama/Llama-2-7b-chat-hf
volume=$PWD/data
token=<your READ token>
docker run --gpus all \
--shm-size 1g \
-e HUGGING_FACE_HUB_TOKEN=$token \
-p 8080:80 \
-v $volume:/data ghcr.io/huggingface/text-generation-inference:1.4 \
--model-id $model
```
| text-generation-inference/docs/source/basic_tutorials/gated_model_access.md/0 | {
"file_path": "text-generation-inference/docs/source/basic_tutorials/gated_model_access.md",
"repo_id": "text-generation-inference",
"token_count": 320
} | 187 |
import sys
import subprocess
import contextlib
import pytest
import asyncio
import os
import docker
import json
import math
import time
import random
from docker.errors import NotFound
from typing import Optional, List, Dict
from syrupy.extensions.json import JSONSnapshotExtension
from aiohttp import ClientConnectorError, ClientOSError, ServerDisconnectedError
from text_generation import AsyncClient
from text_generation.types import Response, Details, InputToken, Token, BestOfSequence
DOCKER_IMAGE = os.getenv("DOCKER_IMAGE", None)
HUGGING_FACE_HUB_TOKEN = os.getenv("HUGGING_FACE_HUB_TOKEN", None)
DOCKER_VOLUME = os.getenv("DOCKER_VOLUME", "/data")
class ResponseComparator(JSONSnapshotExtension):
rtol = 0.2
def serialize(
self,
data,
*,
exclude=None,
matcher=None,
):
if isinstance(data, List):
data = [d.dict() for d in data]
data = self._filter(
data=data, depth=0, path=(), exclude=exclude, matcher=matcher
)
return json.dumps(data, indent=2, ensure_ascii=False, sort_keys=False) + "\n"
def matches(
self,
*,
serialized_data,
snapshot_data,
) -> bool:
def convert_data(data):
data = json.loads(data)
if isinstance(data, Dict):
return Response(**data)
if isinstance(data, List):
return [Response(**d) for d in data]
raise NotImplementedError
def eq_token(token: Token, other: Token) -> bool:
return (
token.id == other.id
and token.text == other.text
and math.isclose(token.logprob, other.logprob, rel_tol=self.rtol)
and token.special == other.special
)
def eq_prefill_token(prefill_token: InputToken, other: InputToken) -> bool:
try:
return (
prefill_token.id == other.id
and prefill_token.text == other.text
and (
math.isclose(
prefill_token.logprob, other.logprob, rel_tol=self.rtol
)
if prefill_token.logprob is not None
else prefill_token.logprob == other.logprob
)
)
except TypeError:
return False
def eq_best_of(details: BestOfSequence, other: BestOfSequence) -> bool:
return (
details.finish_reason == other.finish_reason
and details.generated_tokens == other.generated_tokens
and details.seed == other.seed
and len(details.prefill) == len(other.prefill)
and all(
[
eq_prefill_token(d, o)
for d, o in zip(details.prefill, other.prefill)
]
)
and len(details.tokens) == len(other.tokens)
and all([eq_token(d, o) for d, o in zip(details.tokens, other.tokens)])
)
def eq_details(details: Details, other: Details) -> bool:
return (
details.finish_reason == other.finish_reason
and details.generated_tokens == other.generated_tokens
and details.seed == other.seed
and len(details.prefill) == len(other.prefill)
and all(
[
eq_prefill_token(d, o)
for d, o in zip(details.prefill, other.prefill)
]
)
and len(details.tokens) == len(other.tokens)
and all([eq_token(d, o) for d, o in zip(details.tokens, other.tokens)])
and (
len(details.best_of_sequences)
if details.best_of_sequences is not None
else 0
)
== (
len(other.best_of_sequences)
if other.best_of_sequences is not None
else 0
)
and (
all(
[
eq_best_of(d, o)
for d, o in zip(
details.best_of_sequences, other.best_of_sequences
)
]
)
if details.best_of_sequences is not None
else details.best_of_sequences == other.best_of_sequences
)
)
def eq_response(response: Response, other: Response) -> bool:
return response.generated_text == other.generated_text and eq_details(
response.details, other.details
)
serialized_data = convert_data(serialized_data)
snapshot_data = convert_data(snapshot_data)
if not isinstance(serialized_data, List):
serialized_data = [serialized_data]
if not isinstance(snapshot_data, List):
snapshot_data = [snapshot_data]
return len(snapshot_data) == len(serialized_data) and all(
[eq_response(r, o) for r, o in zip(serialized_data, snapshot_data)]
)
class GenerousResponseComparator(ResponseComparator):
# Needed for GPTQ with exllama which has serious numerical fluctuations.
rtol = 0.75
class LauncherHandle:
def __init__(self, port: int):
self.client = AsyncClient(f"http://localhost:{port}")
def _inner_health(self):
raise NotImplementedError
async def health(self, timeout: int = 60):
assert timeout > 0
for _ in range(timeout):
if not self._inner_health():
raise RuntimeError("Launcher crashed")
try:
await self.client.generate("test")
return
except (ClientConnectorError, ClientOSError, ServerDisconnectedError) as e:
time.sleep(1)
raise RuntimeError("Health check failed")
class ContainerLauncherHandle(LauncherHandle):
def __init__(self, docker_client, container_name, port: int):
super(ContainerLauncherHandle, self).__init__(port)
self.docker_client = docker_client
self.container_name = container_name
def _inner_health(self) -> bool:
container = self.docker_client.containers.get(self.container_name)
return container.status in ["running", "created"]
class ProcessLauncherHandle(LauncherHandle):
def __init__(self, process, port: int):
super(ProcessLauncherHandle, self).__init__(port)
self.process = process
def _inner_health(self) -> bool:
return self.process.poll() is None
@pytest.fixture
def response_snapshot(snapshot):
return snapshot.use_extension(ResponseComparator)
@pytest.fixture
def generous_response_snapshot(snapshot):
return snapshot.use_extension(GenerousResponseComparator)
@pytest.fixture(scope="module")
def event_loop():
loop = asyncio.get_event_loop()
yield loop
loop.close()
@pytest.fixture(scope="module")
def launcher(event_loop):
@contextlib.contextmanager
def local_launcher(
model_id: str,
num_shard: Optional[int] = None,
quantize: Optional[str] = None,
trust_remote_code: bool = False,
use_flash_attention: bool = True,
dtype: Optional[str] = None,
):
port = random.randint(8000, 10_000)
master_port = random.randint(10_000, 20_000)
shard_uds_path = (
f"/tmp/tgi-tests-{model_id.split('/')[-1]}-{num_shard}-{quantize}-server"
)
args = [
"text-generation-launcher",
"--model-id",
model_id,
"--port",
str(port),
"--master-port",
str(master_port),
"--shard-uds-path",
shard_uds_path,
]
env = os.environ
if num_shard is not None:
args.extend(["--num-shard", str(num_shard)])
if quantize is not None:
args.append("--quantize")
args.append(quantize)
if dtype is not None:
args.append("--dtype")
args.append(dtype)
if trust_remote_code:
args.append("--trust-remote-code")
env["LOG_LEVEL"] = "info,text_generation_router=debug"
if not use_flash_attention:
env["USE_FLASH_ATTENTION"] = "false"
with subprocess.Popen(
args, stdout=subprocess.PIPE, stderr=subprocess.PIPE, env=env
) as process:
yield ProcessLauncherHandle(process, port)
process.terminate()
process.wait(60)
launcher_output = process.stdout.read().decode("utf-8")
print(launcher_output, file=sys.stderr)
process.stdout.close()
process.stderr.close()
if not use_flash_attention:
del env["USE_FLASH_ATTENTION"]
@contextlib.contextmanager
def docker_launcher(
model_id: str,
num_shard: Optional[int] = None,
quantize: Optional[str] = None,
trust_remote_code: bool = False,
use_flash_attention: bool = True,
dtype: Optional[str] = None,
):
port = random.randint(8000, 10_000)
args = ["--model-id", model_id, "--env"]
if num_shard is not None:
args.extend(["--num-shard", str(num_shard)])
if quantize is not None:
args.append("--quantize")
args.append(quantize)
if dtype is not None:
args.append("--dtype")
args.append(dtype)
if trust_remote_code:
args.append("--trust-remote-code")
client = docker.from_env()
container_name = f"tgi-tests-{model_id.split('/')[-1]}-{num_shard}-{quantize}"
try:
container = client.containers.get(container_name)
container.stop()
container.wait()
except NotFound:
pass
gpu_count = num_shard if num_shard is not None else 1
env = {"LOG_LEVEL": "info,text_generation_router=debug"}
if not use_flash_attention:
env["USE_FLASH_ATTENTION"] = "false"
if HUGGING_FACE_HUB_TOKEN is not None:
env["HUGGING_FACE_HUB_TOKEN"] = HUGGING_FACE_HUB_TOKEN
volumes = []
if DOCKER_VOLUME:
volumes = [f"{DOCKER_VOLUME}:/data"]
container = client.containers.run(
DOCKER_IMAGE,
command=args,
name=container_name,
environment=env,
auto_remove=False,
detach=True,
device_requests=[
docker.types.DeviceRequest(count=gpu_count, capabilities=[["gpu"]])
],
volumes=volumes,
ports={"80/tcp": port},
shm_size="1G",
)
yield ContainerLauncherHandle(client, container.name, port)
if not use_flash_attention:
del env["USE_FLASH_ATTENTION"]
try:
container.stop()
container.wait()
except NotFound:
pass
container_output = container.logs().decode("utf-8")
print(container_output, file=sys.stderr)
container.remove()
if DOCKER_IMAGE is not None:
return docker_launcher
return local_launcher
@pytest.fixture(scope="module")
def generate_load():
async def generate_load_inner(
client: AsyncClient, prompt: str, max_new_tokens: int, n: int
) -> List[Response]:
futures = [
client.generate(
prompt, max_new_tokens=max_new_tokens, decoder_input_details=True
)
for _ in range(n)
]
return await asyncio.gather(*futures)
return generate_load_inner
| text-generation-inference/integration-tests/conftest.py/0 | {
"file_path": "text-generation-inference/integration-tests/conftest.py",
"repo_id": "text-generation-inference",
"token_count": 6015
} | 188 |
[
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 1,
"logprob": null,
"text": "<s>"
},
{
"id": 4321,
"logprob": -8.6875,
"text": "Test"
},
{
"id": 2009,
"logprob": -11.546875,
"text": "request"
}
],
"seed": null,
"tokens": [
{
"id": 363,
"logprob": -1.5351562,
"special": false,
"text": " for"
},
{
"id": 847,
"logprob": -2.5566406,
"special": false,
"text": " /"
},
{
"id": 2754,
"logprob": -2.2519531,
"special": false,
"text": "api"
},
{
"id": 29914,
"logprob": -0.03414917,
"special": false,
"text": "/"
},
{
"id": 29894,
"logprob": -0.96240234,
"special": false,
"text": "v"
},
{
"id": 29896,
"logprob": -0.3647461,
"special": false,
"text": "1"
},
{
"id": 29914,
"logprob": -0.012901306,
"special": false,
"text": "/"
},
{
"id": 16418,
"logprob": -3.1542969,
"special": false,
"text": "projects"
},
{
"id": 29914,
"logprob": -0.4362793,
"special": false,
"text": "/"
},
{
"id": 29896,
"logprob": -1.9394531,
"special": false,
"text": "1"
}
],
"top_tokens": null
},
"generated_text": " for /api/v1/projects/1"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 1,
"logprob": null,
"text": "<s>"
},
{
"id": 4321,
"logprob": -8.6875,
"text": "Test"
},
{
"id": 2009,
"logprob": -11.546875,
"text": "request"
}
],
"seed": null,
"tokens": [
{
"id": 363,
"logprob": -1.5332031,
"special": false,
"text": " for"
},
{
"id": 847,
"logprob": -2.5625,
"special": false,
"text": " /"
},
{
"id": 2754,
"logprob": -2.2617188,
"special": false,
"text": "api"
},
{
"id": 29914,
"logprob": -0.033996582,
"special": false,
"text": "/"
},
{
"id": 29894,
"logprob": -0.9609375,
"special": false,
"text": "v"
},
{
"id": 29896,
"logprob": -0.36572266,
"special": false,
"text": "1"
},
{
"id": 29914,
"logprob": -0.0129776,
"special": false,
"text": "/"
},
{
"id": 16418,
"logprob": -3.15625,
"special": false,
"text": "projects"
},
{
"id": 29914,
"logprob": -0.4362793,
"special": false,
"text": "/"
},
{
"id": 29896,
"logprob": -1.9394531,
"special": false,
"text": "1"
}
],
"top_tokens": null
},
"generated_text": " for /api/v1/projects/1"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 1,
"logprob": null,
"text": "<s>"
},
{
"id": 4321,
"logprob": -8.6875,
"text": "Test"
},
{
"id": 2009,
"logprob": -11.546875,
"text": "request"
}
],
"seed": null,
"tokens": [
{
"id": 363,
"logprob": -1.5332031,
"special": false,
"text": " for"
},
{
"id": 847,
"logprob": -2.5625,
"special": false,
"text": " /"
},
{
"id": 2754,
"logprob": -2.2617188,
"special": false,
"text": "api"
},
{
"id": 29914,
"logprob": -0.033996582,
"special": false,
"text": "/"
},
{
"id": 29894,
"logprob": -0.9609375,
"special": false,
"text": "v"
},
{
"id": 29896,
"logprob": -0.36572266,
"special": false,
"text": "1"
},
{
"id": 29914,
"logprob": -0.0129776,
"special": false,
"text": "/"
},
{
"id": 16418,
"logprob": -3.15625,
"special": false,
"text": "projects"
},
{
"id": 29914,
"logprob": -0.4362793,
"special": false,
"text": "/"
},
{
"id": 29896,
"logprob": -1.9394531,
"special": false,
"text": "1"
}
],
"top_tokens": null
},
"generated_text": " for /api/v1/projects/1"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 1,
"logprob": null,
"text": "<s>"
},
{
"id": 4321,
"logprob": -8.6875,
"text": "Test"
},
{
"id": 2009,
"logprob": -11.546875,
"text": "request"
}
],
"seed": null,
"tokens": [
{
"id": 363,
"logprob": -1.5332031,
"special": false,
"text": " for"
},
{
"id": 847,
"logprob": -2.5625,
"special": false,
"text": " /"
},
{
"id": 2754,
"logprob": -2.2617188,
"special": false,
"text": "api"
},
{
"id": 29914,
"logprob": -0.033996582,
"special": false,
"text": "/"
},
{
"id": 29894,
"logprob": -0.9609375,
"special": false,
"text": "v"
},
{
"id": 29896,
"logprob": -0.36572266,
"special": false,
"text": "1"
},
{
"id": 29914,
"logprob": -0.0129776,
"special": false,
"text": "/"
},
{
"id": 16418,
"logprob": -3.15625,
"special": false,
"text": "projects"
},
{
"id": 29914,
"logprob": -0.4362793,
"special": false,
"text": "/"
},
{
"id": 29896,
"logprob": -1.9394531,
"special": false,
"text": "1"
}
],
"top_tokens": null
},
"generated_text": " for /api/v1/projects/1"
}
]
| text-generation-inference/integration-tests/models/__snapshots__/test_flash_llama/test_flash_llama_load.json/0 | {
"file_path": "text-generation-inference/integration-tests/models/__snapshots__/test_flash_llama/test_flash_llama_load.json",
"repo_id": "text-generation-inference",
"token_count": 4901
} | 189 |
[
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 14402,
"logprob": null,
"text": "Test"
},
{
"id": 2581,
"logprob": -11.6171875,
"text": " request"
}
],
"seed": null,
"tokens": [
{
"id": 25,
"logprob": -2.3203125,
"special": false,
"text": ":"
},
{
"id": 1391,
"logprob": -0.98779297,
"special": false,
"text": " {"
},
{
"id": 25927,
"logprob": -0.7729492,
"special": false,
"text": "request"
},
{
"id": 92,
"logprob": -0.7241211,
"special": false,
"text": "}"
},
{
"id": 4943,
"logprob": -0.4091797,
"special": false,
"text": "\")"
},
{
"id": 198,
"logprob": -0.119018555,
"special": false,
"text": "\n"
},
{
"id": 50280,
"logprob": -0.9707031,
"special": false,
"text": " "
},
{
"id": 26209,
"logprob": -1.4414062,
"special": false,
"text": "response"
},
{
"id": 796,
"logprob": -0.056854248,
"special": false,
"text": " ="
},
{
"id": 2116,
"logprob": -1.1533203,
"special": false,
"text": " self"
}
],
"top_tokens": null
},
"generated_text": ": {request}\")\n response = self"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 14402,
"logprob": null,
"text": "Test"
},
{
"id": 2581,
"logprob": -11.6171875,
"text": " request"
}
],
"seed": null,
"tokens": [
{
"id": 25,
"logprob": -2.3203125,
"special": false,
"text": ":"
},
{
"id": 1391,
"logprob": -0.98779297,
"special": false,
"text": " {"
},
{
"id": 25927,
"logprob": -0.7729492,
"special": false,
"text": "request"
},
{
"id": 92,
"logprob": -0.7241211,
"special": false,
"text": "}"
},
{
"id": 4943,
"logprob": -0.4091797,
"special": false,
"text": "\")"
},
{
"id": 198,
"logprob": -0.119018555,
"special": false,
"text": "\n"
},
{
"id": 50280,
"logprob": -0.9707031,
"special": false,
"text": " "
},
{
"id": 26209,
"logprob": -1.4414062,
"special": false,
"text": "response"
},
{
"id": 796,
"logprob": -0.056854248,
"special": false,
"text": " ="
},
{
"id": 2116,
"logprob": -1.1533203,
"special": false,
"text": " self"
}
],
"top_tokens": null
},
"generated_text": ": {request}\")\n response = self"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 14402,
"logprob": null,
"text": "Test"
},
{
"id": 2581,
"logprob": -11.6171875,
"text": " request"
}
],
"seed": null,
"tokens": [
{
"id": 25,
"logprob": -2.3203125,
"special": false,
"text": ":"
},
{
"id": 1391,
"logprob": -0.98779297,
"special": false,
"text": " {"
},
{
"id": 25927,
"logprob": -0.7729492,
"special": false,
"text": "request"
},
{
"id": 92,
"logprob": -0.7241211,
"special": false,
"text": "}"
},
{
"id": 4943,
"logprob": -0.4091797,
"special": false,
"text": "\")"
},
{
"id": 198,
"logprob": -0.119018555,
"special": false,
"text": "\n"
},
{
"id": 50280,
"logprob": -0.9707031,
"special": false,
"text": " "
},
{
"id": 26209,
"logprob": -1.4414062,
"special": false,
"text": "response"
},
{
"id": 796,
"logprob": -0.056854248,
"special": false,
"text": " ="
},
{
"id": 2116,
"logprob": -1.1533203,
"special": false,
"text": " self"
}
],
"top_tokens": null
},
"generated_text": ": {request}\")\n response = self"
},
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 14402,
"logprob": null,
"text": "Test"
},
{
"id": 2581,
"logprob": -11.6171875,
"text": " request"
}
],
"seed": null,
"tokens": [
{
"id": 25,
"logprob": -2.3203125,
"special": false,
"text": ":"
},
{
"id": 1391,
"logprob": -0.98779297,
"special": false,
"text": " {"
},
{
"id": 25927,
"logprob": -0.7729492,
"special": false,
"text": "request"
},
{
"id": 92,
"logprob": -0.7241211,
"special": false,
"text": "}"
},
{
"id": 4943,
"logprob": -0.4091797,
"special": false,
"text": "\")"
},
{
"id": 198,
"logprob": -0.119018555,
"special": false,
"text": "\n"
},
{
"id": 50280,
"logprob": -0.9707031,
"special": false,
"text": " "
},
{
"id": 26209,
"logprob": -1.4414062,
"special": false,
"text": "response"
},
{
"id": 796,
"logprob": -0.056854248,
"special": false,
"text": " ="
},
{
"id": 2116,
"logprob": -1.1533203,
"special": false,
"text": " self"
}
],
"top_tokens": null
},
"generated_text": ": {request}\")\n response = self"
}
]
| text-generation-inference/integration-tests/models/__snapshots__/test_flash_phi/test_flash_phi_load.json/0 | {
"file_path": "text-generation-inference/integration-tests/models/__snapshots__/test_flash_phi/test_flash_phi_load.json",
"repo_id": "text-generation-inference",
"token_count": 4672
} | 190 |
{
"details": {
"best_of_sequences": null,
"finish_reason": "length",
"generated_tokens": 10,
"prefill": [
{
"id": 50278,
"logprob": null,
"text": "<|USER|>"
},
{
"id": 1276,
"logprob": -4.5546875,
"text": "What"
},
{
"id": 434,
"logprob": -4.1992188,
"text": "'s"
},
{
"id": 634,
"logprob": -5.125,
"text": " your"
},
{
"id": 12315,
"logprob": -9.8984375,
"text": " mood"
},
{
"id": 3063,
"logprob": -4.0976562,
"text": " today"
},
{
"id": 32,
"logprob": -0.14562988,
"text": "?"
},
{
"id": 50279,
"logprob": -0.26733398,
"text": "<|ASSISTANT|>"
}
],
"seed": null,
"tokens": [
{
"id": 42,
"logprob": -0.86279297,
"special": false,
"text": "I"
},
{
"id": 1353,
"logprob": -0.94921875,
"special": false,
"text": "'m"
},
{
"id": 7016,
"logprob": -2.1835938,
"special": false,
"text": " sorry"
},
{
"id": 13,
"logprob": -0.074035645,
"special": false,
"text": ","
},
{
"id": 1394,
"logprob": -0.86376953,
"special": false,
"text": "You"
},
{
"id": 452,
"logprob": -1.2070312,
"special": false,
"text": " have"
},
{
"id": 247,
"logprob": -1.4365234,
"special": false,
"text": " a"
},
{
"id": 4327,
"logprob": -1.109375,
"special": false,
"text": " choice"
},
{
"id": 273,
"logprob": -0.93408203,
"special": false,
"text": " of"
},
{
"id": 752,
"logprob": -1.8808594,
"special": false,
"text": " what"
}
]
},
"generated_text": "I'm sorry,You have a choice of what"
}
| text-generation-inference/integration-tests/models/__snapshots__/test_neox/test_neox.json/0 | {
"file_path": "text-generation-inference/integration-tests/models/__snapshots__/test_neox/test_neox.json",
"repo_id": "text-generation-inference",
"token_count": 1351
} | 191 |
import pytest
@pytest.fixture(scope="module")
def flash_neox_sharded_handle(launcher):
with launcher("OpenAssistant/oasst-sft-1-pythia-12b", num_shard=2) as handle:
yield handle
@pytest.fixture(scope="module")
async def flash_neox_sharded(flash_neox_sharded_handle):
await flash_neox_sharded_handle.health(300)
return flash_neox_sharded_handle.client
@pytest.mark.asyncio
async def test_flash_neox(flash_neox_sharded, response_snapshot):
response = await flash_neox_sharded.generate(
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
max_new_tokens=10,
decoder_input_details=True,
)
assert response.details.generated_tokens == 10
assert response == response_snapshot
@pytest.mark.asyncio
async def test_flash_neox_load(flash_neox_sharded, generate_load, response_snapshot):
responses = await generate_load(
flash_neox_sharded,
"<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>",
max_new_tokens=10,
n=4,
)
assert len(responses) == 4
assert all([r.generated_text == responses[0].generated_text for r in responses])
assert responses == response_snapshot
| text-generation-inference/integration-tests/models/test_flash_neox_sharded.py/0 | {
"file_path": "text-generation-inference/integration-tests/models/test_flash_neox_sharded.py",
"repo_id": "text-generation-inference",
"token_count": 491
} | 192 |
use std::error::Error;
use vergen::EmitBuilder;
fn main() -> Result<(), Box<dyn Error>> {
// Emit cargo and rustc compile time values
EmitBuilder::builder().all_cargo().all_rustc().emit()?;
// Try to get the git sha from the local git repository
if EmitBuilder::builder()
.fail_on_error()
.git_sha(false)
.emit()
.is_err()
{
// Unable to get the git sha
if let Ok(sha) = std::env::var("GIT_SHA") {
// Set it from an env var
println!("cargo:rustc-env=VERGEN_GIT_SHA={sha}");
}
}
// Set docker label if present
if let Ok(label) = std::env::var("DOCKER_LABEL") {
// Set it from an env var
println!("cargo:rustc-env=DOCKER_LABEL={label}");
}
Ok(())
}
| text-generation-inference/launcher/build.rs/0 | {
"file_path": "text-generation-inference/launcher/build.rs",
"repo_id": "text-generation-inference",
"token_count": 363
} | 193 |
use crate::client::{DecodeTimings, PrefillTimings};
/// Multi shard Client
use crate::{Batch, CachedBatch, Client, Generation, HealthResponse, ShardInfo};
use crate::{ClientError, Result};
use futures::future::join_all;
use tonic::transport::Uri;
use tracing::instrument;
#[derive(Debug, Clone)]
/// Text Generation Inference gRPC multi client
pub struct ShardedClient {
clients: Vec<Client>,
}
impl ShardedClient {
fn new(clients: Vec<Client>) -> Self {
Self { clients }
}
/// Create a new ShardedClient from a master client. The master client will communicate with
/// the other shards and returns all uris/unix sockets with the `service_discovery` gRPC method.
async fn from_master_client(mut master_client: Client) -> Result<Self> {
// Get all uris/unix sockets from the master client
let uris = master_client.service_discovery().await?;
let futures = uris.into_iter().map(Client::connect_uds);
let clients: Result<Vec<Client>> = join_all(futures).await.into_iter().collect();
Ok(Self::new(clients?))
}
/// Returns a client connected to the given uri
pub async fn connect(uri: Uri) -> Result<Self> {
let master_client = Client::connect(uri).await?;
Self::from_master_client(master_client).await
}
/// Returns a client connected to the given unix socket
pub async fn connect_uds(path: String) -> Result<Self> {
let master_client = Client::connect_uds(path).await?;
Self::from_master_client(master_client).await
}
/// Get the model info
#[instrument(skip(self))]
pub async fn info(&mut self) -> Result<ShardInfo> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| client.info())
.collect();
join_all(futures).await.pop().unwrap()
}
/// GRPC health check
#[instrument(skip(self))]
pub async fn health(&mut self) -> Result<HealthResponse> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| client.health())
.collect();
join_all(futures).await.pop().unwrap()
}
/// Clear the past generations cache
#[instrument(skip(self))]
pub async fn clear_cache(&mut self, batch_id: Option<u64>) -> Result<()> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| client.clear_cache(batch_id))
.collect();
join_all(futures).await.into_iter().collect()
}
/// Filter a cached batch
#[instrument(skip(self))]
pub async fn filter_batch(
&mut self,
batch_id: u64,
request_ids: Vec<u64>,
) -> Result<Option<CachedBatch>> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| Box::pin(client.filter_batch(batch_id, request_ids.clone())))
.collect();
// all shards return the same message
join_all(futures).await.pop().unwrap()
}
/// Warmup on a max size batch
///
/// Returns the maximum amount of tokens supported by the hardware
#[instrument(skip(self))]
pub async fn warmup(
&mut self,
max_input_length: u32,
max_prefill_tokens: u32,
max_total_tokens: u32,
) -> Result<Option<u32>> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| {
Box::pin(client.warmup(max_input_length, max_prefill_tokens, max_total_tokens))
})
.collect();
// Take the minimum value
let results = join_all(futures)
.await
.into_iter()
.collect::<Result<Vec<Option<u32>>>>()?;
Ok(results.into_iter().flatten().min())
}
/// Generate one token for each request in the given batch
///
/// Returns Generation for each request in batch
/// and the next cached batch
#[instrument(skip_all, fields(id = & batch.id, size = & batch.size))]
pub async fn prefill(
&mut self,
batch: Batch,
) -> Result<(Vec<Generation>, Option<CachedBatch>, PrefillTimings)> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| Box::pin(client.prefill(batch.clone())))
.collect();
#[allow(clippy::type_complexity)]
let results: Result<Vec<(Vec<Generation>, Option<CachedBatch>, PrefillTimings)>> =
join_all(futures).await.into_iter().collect();
let mut results = results?;
let (mut generations, next_batch, mut timings) =
results.pop().ok_or(ClientError::EmptyResults)?;
// Merge generations from different model shards
for (mut shard_generations, _, shard_timings) in results.into_iter() {
generations.append(&mut shard_generations);
// Return the timings of the slowest shard
if shard_timings.total > timings.total {
timings = shard_timings;
}
}
Ok((generations, next_batch, timings))
}
/// Generate one token for each request in the given cached batches
///
/// Returns Generation for each request in batches
/// and the next cached batch
#[instrument(skip_all, fields(size = batches.iter().map(| batch | {batch.size}).sum::< u32 > ()))]
pub async fn decode(
&mut self,
batches: Vec<CachedBatch>,
) -> Result<(Vec<Generation>, Option<CachedBatch>, DecodeTimings)> {
let futures: Vec<_> = self
.clients
.iter_mut()
.map(|client| Box::pin(client.decode(batches.clone())))
.collect();
#[allow(clippy::type_complexity)]
let results: Result<Vec<(Vec<Generation>, Option<CachedBatch>, DecodeTimings)>> =
join_all(futures).await.into_iter().collect();
let mut results = results?;
let (mut generations, next_batch, mut timings) =
results.pop().ok_or(ClientError::EmptyResults)?;
// Merge generations from different model shards
for (mut shard_generations, _, shard_timings) in results.into_iter() {
generations.append(&mut shard_generations);
// Return the timings of the slowest shard
if shard_timings.total > timings.total {
timings = shard_timings;
}
}
Ok((generations, next_batch, timings))
}
}
| text-generation-inference/router/client/src/sharded_client.rs/0 | {
"file_path": "text-generation-inference/router/client/src/sharded_client.rs",
"repo_id": "text-generation-inference",
"token_count": 2837
} | 194 |
flash_att_commit := 3a9bfd076f98746c73362328958dbc68d145fbec
flash-attention:
# Clone flash attention
pip install -U packaging ninja --no-cache-dir
git clone https://github.com/HazyResearch/flash-attention.git
build-flash-attention: flash-attention
cd flash-attention && git fetch && git checkout $(flash_att_commit)
cd flash-attention && python setup.py build
cd flash-attention/csrc/rotary && python setup.py build
cd flash-attention/csrc/layer_norm && python setup.py build
install-flash-attention: build-flash-attention
pip uninstall flash_attn rotary_emb dropout_layer_norm -y || true
cd flash-attention && python setup.py install && cd csrc/layer_norm && python setup.py install && cd ../rotary && python setup.py install | text-generation-inference/server/Makefile-flash-att/0 | {
"file_path": "text-generation-inference/server/Makefile-flash-att",
"repo_id": "text-generation-inference",
"token_count": 242
} | 195 |
// Adapted from turboderp exllama: https://github.com/turboderp/exllama
#include <torch/extension.h>
#include <c10/cuda/CUDAGuard.h>
#include <ATen/cuda/CUDAContext.h>
#include <cuda_runtime.h>
#include <cuda_fp16.h>
#include <cstdint>
#include <cstdio>
#include "util.cuh"
#include "tuning.h"
#include "cuda_buffers.cuh"
#include "cuda_func/q4_matrix.cuh"
#include "cuda_func/q4_matmul.cuh"
#include "cuda_func/column_remap.cuh"
// Check CUDA return code. We don't want to include Torch headers in the .cu files because parsing them adds almost a
// minute to the compile time on a 12900K. Also passing exceptions back to Python is super tricky, so in place of
// exceptions, CUDA functions return with a cudaError_t which we can parse and dump to the console.
void check_cuda(cudaError_t ret)
{
switch (ret)
{
case cudaSuccess:
break;
case cudaUnspecified:
printf(" **** Unspecified error\n");
TORCH_CHECK(false, "CUDA error");
break;
default:
printf(" **** CUDA error\n"); \
printf(" **** %s\n", cudaGetErrorString(ret)); \
TORCH_CHECK(false, "CUDA error"); \
break;
}
}
// Some decluttering macros
#define STRINGIFY_(__x) #__x
#define STRINGIFY(__x) STRINGIFY_(__x)
#define TORCH_CHECK_DTYPE(__x, __dtype) TORCH_CHECK((__x).dtype() == torch::__dtype, #__x " is incorrect datatype, must be " #__dtype)
#define TORCH_CHECK_DTYPE_OPT(__x, __dtype) TORCH_CHECK((__x).device().is_meta() || (__x).dtype() == torch::__dtype, #__x " is incorrect datatype, must be " #__dtype)
#define TORCH_CHECK_SHAPES(__x, __dim_x, __y, __dim_y, __scale_y) TORCH_CHECK((__x).size(__dim_x) == (__y).size(__dim_y) * __scale_y, #__x " and " #__y " have incompatible shapes")
#define TORCH_CHECK_SHAPES_OPT(__x, __dim_x, __y, __dim_y, __scale_y) TORCH_CHECK((__x).device().is_meta() || (__x).size(__dim_x) == (__y).size(__dim_y) * __scale_y, #__x " and " #__y " have incompatible shapes")
#define TORCH_CHECK_SHAPE_MOD(__x, __dim_x, __mod) TORCH_CHECK((__x).size(__dim_x) % __mod == 0, #__x ".shape[" STRINGIFY(__dim_x) "] must be a multiple of " STRINGIFY(__mod))
#define TORCH_CHECK_DEVICE_INDEX(__index) \
do { \
TORCH_CHECK(__index >= 0, "no device index"); \
TORCH_CHECK(__index < CUDA_MAX_DEVICES, "invalid device index"); \
} while(0)
#define TORCH_CHECK_QUANT(__w, __w_scales, __w_zeros, __seq_g_idx, __x_map) \
do { \
TORCH_CHECK_DTYPE(__w, kInt); \
TORCH_CHECK_DTYPE(__w_scales, kHalf); \
TORCH_CHECK_DTYPE(__w_zeros, kInt); \
TORCH_CHECK_DTYPE_OPT(__seq_g_idx, kShort); \
TORCH_CHECK_DTYPE_OPT(__x_map, kInt); \
TORCH_CHECK_SHAPES_OPT(__seq_g_idx, 0, __w, 0, 2 * 8); \
TORCH_CHECK_SHAPES_OPT(__x_map, 0, __w, 0, 8); \
} while(0)
int get_groupsize(torch::Tensor w, torch::Tensor w_zeros)
{
int groupsize = w.size(0) * 8 / w_zeros.size(0);
TORCH_CHECK(groupsize * w_zeros.size(0) == w.size(0) * 8, "w.shape[-2] must be a multiple of zeros.shape[-2]")
return groupsize;
}
// Tuning parameters
ExLlamaTuning tuningParams;
void set_tuning_params
(
int matmul_recons_thd,
bool matmul_fused_remap,
bool matmul_no_half2
)
{
tuningParams.matmul_recons_thd = matmul_recons_thd;
tuningParams.matmul_fused_remap = matmul_fused_remap;
tuningParams.matmul_no_half2 = matmul_no_half2;
}
// Release all unmanaged objects allocated by the extension
void cleanup()
{
cleanup_buffers_cuda();
g_q4_free_matrices();
}
// Prepare buffers for forward pass
void prepare_buffers
(
torch::Device device,
torch::Tensor temp_state,
torch::Tensor temp_dq
)
{
int device_index = device.index();
TORCH_CHECK_DEVICE_INDEX(device_index);
const at::cuda::OptionalCUDAGuard device_guard(device);
prepare_buffers_cuda
(
device_index,
(half*) temp_state.data_ptr(),
(half*) temp_dq.data_ptr()
);
}
// Create Q4Matrix, return handle
uintptr_t make_q4
(
torch::Tensor qweight,
torch::Tensor qzeros,
torch::Tensor scales,
torch::Tensor g_idx,
int device
)
{
TORCH_CHECK_DTYPE(qweight, kInt);
TORCH_CHECK_DTYPE(qzeros, kInt);
TORCH_CHECK_DTYPE(scales, kHalf);
TORCH_CHECK_DTYPE_OPT(g_idx, kInt);
TORCH_CHECK_SHAPES(qweight, 1, qzeros, 1, 8);
TORCH_CHECK_SHAPES(scales, 1, qweight, 1, 1);
TORCH_CHECK_SHAPES(qzeros, 0, scales, 0, 1);
int width = qweight.size(1);
int height = qweight.size(0) * 8;
int groups = qzeros.size(0);
Q4Matrix* m = new Q4Matrix
(
height,
width,
groups,
(uint32_t*) qweight.data_ptr(),
(uint32_t*) qzeros.data_ptr(),
(half*) scales.data_ptr(),
g_idx.device().is_meta() ? NULL : (uint32_t*) g_idx.data_ptr(),
device
);
g_q4_keep_matrix(m);
return reinterpret_cast<uintptr_t> (m);
}
// Matmul half @ quant -> half
void q4_matmul
(
torch::Tensor x,
uintptr_t w,
torch::Tensor out
)
{
Q4Matrix* wm = reinterpret_cast<Q4Matrix*> (w);
TORCH_CHECK_DTYPE(x, kHalf);
TORCH_CHECK_DTYPE(out, kHalf);
TORCH_CHECK_SHAPES(x, 0, out, 0, 1);
TORCH_CHECK(wm->height == x.size(-1), "x and w have incompatible shapes")
const at::cuda::OptionalCUDAGuard device_guard(device_of(x));
int x_height = x.size(0);
if (tuningParams.matmul_recons_thd == 0 || x_height < tuningParams.matmul_recons_thd)
{
q4_matmul_cuda
(
&tuningParams,
(half*) x.data_ptr(),
x_height,
wm,
(half*) out.data_ptr()
);
}
else
{
q4_matmul_recons_cuda
(
&tuningParams,
(half*) x.data_ptr(),
x_height,
wm,
(half*) out.data_ptr(),
at::cuda::getCurrentCUDABlasHandle()
);
}
}
// Remap columns in half tensor
void column_remap
(
torch::Tensor x,
torch::Tensor x_new,
torch::Tensor x_map
)
{
TORCH_CHECK_DTYPE(x, kHalf);
TORCH_CHECK_DTYPE(x_new, kHalf);
TORCH_CHECK_DTYPE(x_map, kInt);
TORCH_CHECK_SHAPES(x_map, 0, x, 1, 1);
int height = x.size(0);
int width = x.size(1);
const at::cuda::OptionalCUDAGuard device_guard(device_of(x));
column_remap_cuda
(
(half*) x.data_ptr(),
(half*) x_new.data_ptr(),
height,
width,
(uint32_t*) x_map.data_ptr()
);
}
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m)
{
m.def("set_tuning_params", &set_tuning_params, "set_tuning_params");
m.def("prepare_buffers", &prepare_buffers, "prepare_buffers");
m.def("cleanup", &cleanup, "cleanup");
m.def("make_q4", &make_q4, "make_q4");
m.def("q4_matmul", &q4_matmul, "q4_matmul");
}
| text-generation-inference/server/exllama_kernels/exllama_kernels/exllama_ext.cpp/0 | {
"file_path": "text-generation-inference/server/exllama_kernels/exllama_kernels/exllama_ext.cpp",
"repo_id": "text-generation-inference",
"token_count": 3215
} | 196 |
#ifndef _qdq_2_cuh
#define _qdq_2_cuh
#include "qdq_util.cuh"
#include "../../config.h"
#if QMODE_2BIT == 1
// Permutation:
//
// ffddbb99 77553311 eeccaa88 66442200
__forceinline__ __device__ void shuffle_2bit_16
(
uint32_t* q,
int stride
)
{
uint32_t qa = q[0];
uint32_t qb = 0;
#pragma unroll
for (int i = 0; i < 8; i++)
{
uint32_t qa0 = qa & 0x03;
uint32_t qa1 = (qa & 0x0c) >> 2;
qa >>= 4;
qb |= (qa1 << (i * 2 + 16));
qb |= (qa0 << (i * 2));
}
q[0] = qb;
}
__forceinline__ __device__ void dequant_2bit_16
(
const uint32_t q_0,
half2 (&dq)[8],
int stride
)
{
const uint32_t c0 = 0x64006400;
const half y4_ = __float2half_rn(1.0f / 4.0f);
const half y16_ = __float2half_rn(1.0f / 16.0f);
const half y64_ = __float2half_rn(1.0f / 64.0f);
const half2 y4 = __halves2half2(y4_, y4_);
const half2 y16 = __halves2half2(y16_, y16_);
const half2 y64 = __halves2half2(y64_, y64_);
const half z1_ = __float2half_rn(-1024.0f - 2.0f);
const half z4_ = __float2half_rn(-1024.0f / 4.0f - 2.0f);
const half z16_ = __float2half_rn(-1024.0f / 16.0f - 2.0f);
const half z64_ = __float2half_rn(-1024.0f / 64.0f - 2.0f);
const half2 z1 = __halves2half2(z1_, z1_);
const half2 z4 = __halves2half2(z4_, z4_);
const half2 z16 = __halves2half2(z16_, z16_);
const half2 z64 = __halves2half2(z64_, z64_);
uint32_t qa = q_0;
half2_uint32 q0((qa & 0x00030003) | c0); // half2(q[ 0], q[ 1]) + 1024
half2_uint32 q1((qa & 0x000c000c) | c0); // half2(q[ 2], q[ 3]) * 4 + 1024
half2_uint32 q2((qa & 0x00300030) | c0); // half2(q[ 4], q[ 5]) * 16 + 1024
half2_uint32 q3((qa & 0x00c000c0) | c0); // half2(q[ 6], q[ 7]) * 64 + 1024
qa >>= 8;
half2_uint32 q4((qa & 0x00030003) | c0); // half2(q[ 8], q[ 8]) + 1024
half2_uint32 q5((qa & 0x000c000c) | c0); // half2(q[10], q[11]) * 4 + 1024
half2_uint32 q6((qa & 0x00300030) | c0); // half2(q[12], q[13]) * 16 + 1024
half2_uint32 q7((qa & 0x00c000c0) | c0); // half2(q[14], q[15]) * 64 + 1024
dq[0] = __hadd2(q0.as_half2, z1);
dq[1] = __hfma2(q1.as_half2, y4, z4);
dq[2] = __hfma2(q2.as_half2, y16, z16);
dq[3] = __hfma2(q3.as_half2, y64, z64);
dq[4] = __hadd2(q4.as_half2, z1);
dq[5] = __hfma2(q5.as_half2, y4, z4);
dq[6] = __hfma2(q6.as_half2, y16, z16);
dq[7] = __hfma2(q7.as_half2, y64, z64);
}
#else
__forceinline__ __device__ void shuffle_2bit_16
(
uint32_t* q,
int stride
)
{
}
__forceinline__ __device__ void dequant_2bit_16
(
const uint32_t q_0,
half2 (&dq)[8],
int stride
)
{
half dqh[16];
for (int i = 0; i < 16; i++) dqh[i] = dq_ns(exb(q_0, i * 2, 0x03), 2);
for (int i = 0; i < 8; i++) dq[i] = __halves2half2(dqh[i * 2], dqh[i * 2 + 1]);
}
#endif
#endif | text-generation-inference/server/exllamav2_kernels/exllamav2_kernels/cuda/quant/qdq_2.cuh/0 | {
"file_path": "text-generation-inference/server/exllamav2_kernels/exllamav2_kernels/cuda/quant/qdq_2.cuh",
"repo_id": "text-generation-inference",
"token_count": 1588
} | 197 |
import pytest
import torch
from copy import copy
from transformers import AutoTokenizer
from text_generation_server.pb import generate_pb2
from text_generation_server.models.causal_lm import CausalLMBatch
from text_generation_server.utils import weight_hub_files, download_weights
from text_generation_server.models.bloom import BloomCausalLMBatch, BLOOMSharded
@pytest.fixture(scope="session")
def default_bloom():
model_id = "bigscience/bloom-560m"
revision = "main"
filenames = weight_hub_files(model_id, revision, ".safetensors")
download_weights(filenames, model_id, revision)
return BLOOMSharded(model_id)
@pytest.fixture(scope="session")
def bloom_560m_tokenizer():
return AutoTokenizer.from_pretrained("bigscience/bloom-560m", padding_side="left")
@pytest.fixture
def default_pb_request(default_pb_parameters, default_pb_stop_parameters):
return generate_pb2.Request(
id=0,
inputs="Test",
prefill_logprobs=True,
truncate=100,
parameters=default_pb_parameters,
stopping_parameters=default_pb_stop_parameters,
)
@pytest.fixture
def default_pb_batch(default_pb_request):
return generate_pb2.Batch(id=0, requests=[default_pb_request], size=1)
@pytest.fixture
def default_bloom_batch(default_pb_batch, bloom_560m_tokenizer):
return BloomCausalLMBatch.from_pb(
default_pb_batch, bloom_560m_tokenizer, torch.float32, torch.device("cpu")
)
@pytest.fixture
def default_multi_requests_bloom_batch(default_pb_request, bloom_560m_tokenizer):
req_0 = copy(default_pb_request)
req_0.id = 1
req_1 = default_pb_request
req_1.id = 2
req_1.stopping_parameters.max_new_tokens = 5
batch_pb = generate_pb2.Batch(id=0, requests=[req_0, req_1], size=2)
return BloomCausalLMBatch.from_pb(
batch_pb, bloom_560m_tokenizer, torch.float32, torch.device("cpu")
)
def test_batch_from_pb(default_pb_batch, default_bloom_batch):
batch = default_bloom_batch
assert batch.batch_id == default_pb_batch.id
assert batch.requests == default_pb_batch.requests
assert len(batch.input_ids) == default_pb_batch.size
assert batch.input_ids[0][-1] == 10264
assert torch.all(batch.input_ids[0][:-1] == 3)
assert batch.attention_mask[0][0] == 1
assert torch.all(batch.attention_mask[0][1:] == 0)
assert batch.past_key_values is None
assert all(
[
torch.equal(input_ids, all_input_ids[:, 0])
for input_ids, all_input_ids in zip(batch.input_ids, batch.all_input_ids)
]
)
assert batch.input_lengths == [1]
assert len(batch) == default_pb_batch.size
assert len(batch.next_token_choosers) == len(batch.stopping_criterias) == len(batch)
assert batch.max_input_length == batch.input_lengths[0]
def test_batch_concatenate_no_prefill(default_bloom_batch):
with pytest.raises(ValueError):
BloomCausalLMBatch.concatenate([default_bloom_batch, default_bloom_batch])
def test_causal_lm_batch_type(default_bloom):
assert default_bloom.batch_type == BloomCausalLMBatch
def test_causal_lm_generate_token(default_bloom, default_bloom_batch):
sequence_length = len(default_bloom_batch.all_input_ids[0])
generations, next_batch, _ = default_bloom.generate_token(default_bloom_batch)
assert len(generations) == len(default_bloom_batch)
assert isinstance(next_batch, CausalLMBatch)
assert not next_batch.keys_head_dim_last
assert len(next_batch.all_input_ids) == len(next_batch)
assert len(next_batch.all_input_ids[0]) == sequence_length + 1
assert len(next_batch.attention_mask[0]) == 11
assert torch.all(next_batch.all_input_ids[0][-2:] == 10264)
assert torch.all(next_batch.all_input_ids[0][:-2] == 3)
assert torch.all(next_batch.attention_mask[0][:2] == 1)
assert torch.all(next_batch.attention_mask[0][2:] == 0)
assert next_batch.input_ids.shape == (len(next_batch), 1)
assert next_batch.input_ids[0, 0] == 10264
assert next_batch.input_lengths == [2]
assert next_batch.max_input_length == next_batch.input_lengths[0]
assert next_batch.past_key_values is not None
assert all(
[p[0].shape == (16, 64, sequence_length) for p in next_batch.past_key_values]
)
assert all(
[p[1].shape == (16, sequence_length, 64) for p in next_batch.past_key_values]
)
assert all([generation.generated_text is None for generation in generations])
assert all([len(generation.prefill_tokens) == 1 for generation in generations])
assert all(
[
token_id.item() == 10264
for generation in generations
for token_id in generation.tokens.token_ids
]
)
assert all(
[
token_text == "Test"
for generation in generations
for token_text in generation.tokens.texts
]
)
assert generations[0].request_id == 0
def test_causal_lm_generate_token_completion(default_bloom, default_bloom_batch):
next_batch = default_bloom_batch
for _ in range(default_bloom_batch.stopping_criterias[0].max_new_tokens - 1):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(default_bloom_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is None
assert len(generations) == 1
assert (
generations[0].generated_text.text == "TestTestTestTestTestTestTestTestTestTest"
)
assert generations[0].request_id == default_bloom_batch.requests[0].id
assert (
generations[0].generated_text.generated_tokens
== default_bloom_batch.stopping_criterias[0].max_new_tokens
)
def test_causal_lm_generate_token_completion_multi(
default_bloom, default_multi_requests_bloom_batch
):
next_batch = default_multi_requests_bloom_batch
for i in range(
default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens - 1
):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(default_multi_requests_bloom_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is not None
assert len(generations) == 2
assert generations[1].generated_text.text == "TestTestTestTestTest"
assert (
generations[1].request_id == default_multi_requests_bloom_batch.requests[1].id
)
assert (
generations[1].generated_text.generated_tokens
== default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens
)
# Copy stopping_criterias before filtering
stopping_criterias = default_multi_requests_bloom_batch.stopping_criterias.copy()
next_batch = next_batch.filter([next_batch.requests[0].id])
for _ in range(
stopping_criterias[0].max_new_tokens - stopping_criterias[1].max_new_tokens - 1
):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(next_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is None
assert len(generations) == 1
assert (
generations[0].generated_text.text == "TestTestTestTestTestTestTestTestTestTest"
)
assert (
generations[0].request_id == default_multi_requests_bloom_batch.requests[0].id
)
assert (
generations[0].generated_text.generated_tokens
== default_multi_requests_bloom_batch.stopping_criterias[0].max_new_tokens
)
def test_batch_concatenate(
default_bloom, default_bloom_batch, default_multi_requests_bloom_batch
):
next_batch_0 = default_bloom_batch
_, next_batch_0, _ = default_bloom.generate_token(next_batch_0)
_, next_batch_0, _ = default_bloom.generate_token(next_batch_0)
next_batch_1 = default_multi_requests_bloom_batch
_, next_batch_1, _ = default_bloom.generate_token(next_batch_1)
# Clone past_key_values before concatenating to compare after,
# because they are removed from the concatenated batches
next_batch_0_past_key_values = [
(k.clone(), v.clone()) for (k, v) in next_batch_0.past_key_values
]
next_batch_1_past_key_values = [
(k.clone(), v.clone()) for (k, v) in next_batch_1.past_key_values
]
next_batch = BloomCausalLMBatch.concatenate([next_batch_0, next_batch_1])
assert torch.equal(next_batch.all_input_ids[0], next_batch_0.all_input_ids[0])
assert torch.equal(next_batch.all_input_ids[1], next_batch_1.all_input_ids[0])
assert torch.equal(next_batch.all_input_ids[2], next_batch_1.all_input_ids[1])
assert torch.all(
next_batch.attention_mask[0, : -next_batch.padding_right_offset] == 1
)
assert torch.all(
next_batch.attention_mask[1:, 1 : -next_batch.padding_right_offset] == 1
)
assert torch.all(next_batch.attention_mask[1:, 3:] == 0)
assert next_batch.batch_id == 0
assert torch.all(next_batch.input_ids == 10264)
assert next_batch.input_lengths == [3, 2, 2]
assert next_batch.max_input_length == 3
assert next_batch.requests[0] == next_batch_0.requests[0]
assert next_batch.requests[1:] == next_batch_1.requests
assert next_batch.next_token_choosers[0] == next_batch_0.next_token_choosers[0]
assert next_batch.next_token_choosers[1:] == next_batch_1.next_token_choosers
assert next_batch.stopping_criterias[0] == next_batch_0.stopping_criterias[0]
assert next_batch.stopping_criterias[1:] == next_batch_1.stopping_criterias
assert next_batch.past_key_values is not None
assert all([p[0].shape == (3, 16, 64, 2) for p in next_batch.past_key_values])
assert all([p[1].shape == (3, 16, 2, 64) for p in next_batch.past_key_values])
for i, past in enumerate(next_batch.past_key_values):
assert torch.equal(next_batch_0_past_key_values[i][0][:, :, -2:], past[0][0])
assert torch.equal(
next_batch_1_past_key_values[i][0][:, :, -1:],
past[0][1:, :, :, -1].reshape(-1, 64, 1),
)
assert torch.equal(next_batch_0_past_key_values[i][1][:, -2:, :], past[1][0])
assert torch.equal(
next_batch_1_past_key_values[i][1][:, -1:, :],
past[1][1:, :, -1, :].reshape(-1, 1, 64),
)
for _ in range(
default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens - 2
):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(next_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is not None
assert len(generations) == 3
assert generations[2].generated_text.text == "TestTestTestTestTest"
assert (
generations[2].request_id == default_multi_requests_bloom_batch.requests[1].id
)
assert (
generations[2].generated_text.generated_tokens
== default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens
)
next_batch = next_batch.filter(
[next_batch.requests[0].id, next_batch.requests[1].id]
)
for _ in range(
default_bloom_batch.stopping_criterias[0].max_new_tokens
- default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens
- 2
):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(next_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is not None
assert len(generations) == 2
assert (
generations[0].generated_text.text == "TestTestTestTestTestTestTestTestTestTest"
)
assert generations[0].request_id == default_bloom_batch.requests[0].id
assert (
generations[0].generated_text.generated_tokens
== default_bloom_batch.stopping_criterias[0].max_new_tokens
)
next_batch = next_batch.filter([next_batch.requests[1].id])
for _ in range(
default_multi_requests_bloom_batch.stopping_criterias[0].max_new_tokens
- default_bloom_batch.stopping_criterias[0].max_new_tokens
- default_multi_requests_bloom_batch.stopping_criterias[1].max_new_tokens
- 4
):
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert len(generations) == len(next_batch)
generations, next_batch, _ = default_bloom.generate_token(next_batch)
assert next_batch is None
assert len(generations) == 1
assert (
generations[0].generated_text.text == "TestTestTestTestTestTestTestTestTestTest"
)
assert (
generations[0].request_id == default_multi_requests_bloom_batch.requests[0].id
)
assert (
generations[0].generated_text.generated_tokens
== default_multi_requests_bloom_batch.stopping_criterias[0].max_new_tokens
)
| text-generation-inference/server/tests/models/test_bloom.py/0 | {
"file_path": "text-generation-inference/server/tests/models/test_bloom.py",
"repo_id": "text-generation-inference",
"token_count": 5296
} | 198 |
import math
import torch
from typing import Optional, List, Tuple
BLOCK_SIZE: int = 16
# Will be set in warmup
CACHE_MANAGER: Optional["CacheManager"] = None
class CacheManager:
def __init__(
self,
num_blocks: int,
num_layers: int,
num_heads: int,
head_size: int,
repeat_slots: bool,
dtype: torch.dtype,
device: torch.device,
):
self.block_size = BLOCK_SIZE
self.num_blocks = num_blocks
self.repeat_slots = repeat_slots
element_size = torch.tensor([], dtype=dtype).element_size()
x = self.block_size // element_size
self.kv_cache = [
(
torch.empty(
(num_blocks, num_heads, head_size // x, self.block_size, x),
dtype=dtype,
device=device,
),
torch.empty(
(num_blocks, num_heads, head_size, self.block_size),
dtype=dtype,
device=device,
),
)
for _ in range(num_layers)
]
self.free_block_mask = torch.ones(num_blocks, dtype=torch.int32, device="cpu")
self.slots = torch.arange(
0, num_blocks * self.block_size, dtype=torch.int32
).view(num_blocks, self.block_size)
def allocate(
self,
needed_blocks_slots: List[Tuple[int, int]],
blocks: int,
max_blocks: int,
device: torch.device,
):
# Get free blocks indices by finding values in mask that are not set to 0
free_block_indices = self.free_block_mask.nonzero()
assert (
len(free_block_indices) >= blocks
), f"Out of available cache blocks: asked {blocks}, only {len(free_block_indices)} free blocks"
# Slice by the number of required blocks
block_indices = free_block_indices[:blocks]
block_indices = block_indices.flatten()
# Padded block tables
block_tables_tensor = torch.zeros(
(len(needed_blocks_slots), max_blocks), dtype=torch.int32
)
# Allocate paged attention blocks
cumulative_blocks = 0
slots = []
block_tables = []
for i, (needed_blocks, needed_slots) in enumerate(needed_blocks_slots):
# Get allocated blocks for this sequence
allocated_blocks = block_indices[
cumulative_blocks : cumulative_blocks + needed_blocks
]
# Get slots for the allocated blocks
all_slots = self.slots[allocated_blocks].flatten()
# Repeat slots in the case of context sliding window
if needed_slots > len(all_slots) and self.repeat_slots:
repeats = math.ceil(needed_slots / len(all_slots))
all_slots = all_slots.repeat(repeats)
allocated_slots = all_slots[:needed_slots]
slots.append(allocated_slots)
block_tables.append(allocated_blocks.tolist())
block_tables_tensor[i, :needed_blocks] = allocated_blocks
cumulative_blocks += needed_blocks
block_tables = block_tables
block_tables_tensor = block_tables_tensor.to(device)
slots = torch.concat(slots).to(device)
# Allocate the required number of blocks by setting the mask to 0
self.free_block_mask[block_indices] = 0
return block_tables, block_tables_tensor, slots
def free(self, block_indices: Optional[List[int]]):
if block_indices is not None and block_indices:
# Reset mask
self.free_block_mask[block_indices] = 1
def set_cache_manager(
num_blocks: int,
num_layers: int,
num_heads: int,
head_size: int,
repeat_slots: bool,
dtype: torch.dtype,
device: torch.device,
) -> CacheManager:
global CACHE_MANAGER
if CACHE_MANAGER is not None:
del CACHE_MANAGER
torch.cuda.empty_cache()
CACHE_MANAGER = CacheManager(
num_blocks, num_layers, num_heads, head_size, repeat_slots, dtype, device
)
return CACHE_MANAGER
def get_cache_manager() -> CacheManager:
global CACHE_MANAGER
if CACHE_MANAGER is None:
raise RuntimeError("cache manager was not initialized")
return CACHE_MANAGER
| text-generation-inference/server/text_generation_server/models/cache_manager.py/0 | {
"file_path": "text-generation-inference/server/text_generation_server/models/cache_manager.py",
"repo_id": "text-generation-inference",
"token_count": 2033
} | 199 |
# coding=utf-8
# Copyright 2021 The OpenAI Team Authors and The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" PyTorch IdeficsVision model: a copy of CLIPVisionModel using a simpler config object"""
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
import torch.utils.checkpoint
from torch import nn
from transformers.activations import ACT2FN
from transformers.modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling
from transformers.utils import (
ModelOutput,
logging,
)
from text_generation_server.utils.layers import (
TensorParallelColumnLinear,
TensorParallelRowLinear,
TensorParallelEmbedding,
)
logger = logging.get_logger(__name__)
@dataclass
class IdeficsVisionModelOutput(ModelOutput):
"""
Base class for vision model's outputs that also contains image embeddings of the pooling of the last hidden states.
Args:
image_embeds (`torch.FloatTensor` of shape `(batch_size, output_dim)` *optional* returned when model is initialized with `with_projection=True`):
The image embeddings obtained by applying the projection layer to the pooler_output.
last_hidden_state (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`):
Sequence of hidden-states at the output of the last layer of the model.
hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.
Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
attentions (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`):
Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
sequence_length)`.
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
heads.
"""
image_embeds: Optional[torch.FloatTensor] = None
last_hidden_state: torch.FloatTensor = None
hidden_states: Optional[Tuple[torch.FloatTensor]] = None
attentions: Optional[Tuple[torch.FloatTensor]] = None
# Copied from transformers.models.clip.modeling_clip.CLIPVisionEmbeddings with CLIP->Idefics
class IdeficsVisionEmbeddings(nn.Module):
def __init__(self, prefix, config, weights):
super().__init__()
self.config = config
self.embed_dim = config.hidden_size
self.image_size = config.image_size
self.patch_size = config.patch_size
self.class_embedding = nn.Parameter(
weights.get_tensor(f"{prefix}.class_embedding")
)
self.patch_embedding = nn.Conv2d.load_no_bias(
prefix=f"{prefix}.patch_embedding",
weights=weights,
in_channels=config.num_channels,
out_channels=self.embed_dim,
kernel_size=self.patch_size,
stride=self.patch_size,
)
self.num_patches = (self.image_size // self.patch_size) ** 2
self.num_positions = self.num_patches + 1
self.position_embedding = TensorParallelEmbedding(
prefix="model.vision_model.embeddings.position_embedding", weights=weights
)
self.position_ids = (
torch.arange(self.num_positions).expand((1, -1)).to(device=weights.device)
)
def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
batch_size = pixel_values.shape[0]
target_dtype = self.patch_embedding.weight.dtype
patch_embeds = self.patch_embedding(
pixel_values.to(dtype=target_dtype)
) # shape = [*, width, grid, grid]
patch_embeds = patch_embeds.flatten(2).transpose(1, 2)
class_embeds = self.class_embedding.expand(batch_size, 1, -1)
embeddings = torch.cat([class_embeds, patch_embeds], dim=1)
embeddings = embeddings + self.position_embedding(self.position_ids)
return embeddings
# Copied from transformers.models.clip.modeling_clip.CLIPAttention with CLIP->IdeficsVision
class IdeficsVisionAttention(nn.Module):
"""Multi-headed attention from 'Attention Is All You Need' paper"""
def __init__(self, prefix, config, weights):
super().__init__()
self.config = config
self.embed_dim = config.hidden_size
self.num_heads = config.num_attention_heads
self.head_dim = self.embed_dim // self.num_heads
if self.head_dim * self.num_heads != self.embed_dim:
raise ValueError(
f"embed_dim must be divisible by num_heads (got `embed_dim`: {self.embed_dim} and `num_heads`:"
f" {self.num_heads})."
)
self.scale = self.head_dim**-0.5
self.dropout = config.attention_dropout
process_group = weights.process_group
if self.num_heads % weights.process_group.size() != 0:
raise ValueError(
f"`num_heads` must be divisible by `num_shards` (got `num_heads`: {self.num_heads} "
f"and `num_shards`: {weights.process_group.size()}"
)
self.num_heads = self.num_heads // weights.process_group.size()
self.embed_dim = self.embed_dim // weights.process_group.size()
self.k_proj = TensorParallelColumnLinear.load(
config, prefix=f"{prefix}.k_proj", weights=weights, bias=True
)
self.v_proj = TensorParallelColumnLinear.load(
config, prefix=f"{prefix}.v_proj", weights=weights, bias=True
)
self.q_proj = TensorParallelColumnLinear.load(
config, prefix=f"{prefix}.q_proj", weights=weights, bias=True
)
self.out_proj = TensorParallelRowLinear.load(
config, prefix=f"{prefix}.out_proj", weights=weights, bias=True
)
def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
return (
tensor.view(bsz, seq_len, self.num_heads, self.head_dim)
.transpose(1, 2)
.contiguous()
)
def forward(
self,
hidden_states: torch.Tensor,
attention_mask: Optional[torch.Tensor] = None,
causal_attention_mask: Optional[torch.Tensor] = None,
output_attentions: Optional[bool] = False,
) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
"""Input shape: Batch x Time x Channel"""
bsz, tgt_len, _ = hidden_states.size()
# get query proj
query_states = self.q_proj(hidden_states) * self.scale
key_states = self._shape(self.k_proj(hidden_states), -1, bsz)
value_states = self._shape(self.v_proj(hidden_states), -1, bsz)
proj_shape = (bsz * self.num_heads, -1, self.head_dim)
query_states = self._shape(query_states, tgt_len, bsz).view(*proj_shape)
key_states = key_states.view(*proj_shape)
value_states = value_states.view(*proj_shape)
src_len = key_states.size(1)
attn_weights = torch.bmm(query_states, key_states.transpose(1, 2))
if attn_weights.size() != (bsz * self.num_heads, tgt_len, src_len):
raise ValueError(
f"Attention weights should be of size {(bsz * self.num_heads, tgt_len, src_len)}, but is"
f" {attn_weights.size()}"
)
# apply the causal_attention_mask first
if causal_attention_mask is not None:
if causal_attention_mask.size() != (bsz, 1, tgt_len, src_len):
raise ValueError(
f"Attention mask should be of size {(bsz, 1, tgt_len, src_len)}, but is"
f" {causal_attention_mask.size()}"
)
attn_weights = (
attn_weights.view(bsz, self.num_heads, tgt_len, src_len)
+ causal_attention_mask
)
attn_weights = attn_weights.view(bsz * self.num_heads, tgt_len, src_len)
if attention_mask is not None:
if attention_mask.size() != (bsz, 1, tgt_len, src_len):
raise ValueError(
f"Attention mask should be of size {(bsz, 1, tgt_len, src_len)}, but is {attention_mask.size()}"
)
attn_weights = (
attn_weights.view(bsz, self.num_heads, tgt_len, src_len)
+ attention_mask
)
attn_weights = attn_weights.view(bsz * self.num_heads, tgt_len, src_len)
attn_weights = nn.functional.softmax(attn_weights, dim=-1)
if output_attentions:
# this operation is a bit akward, but it's required to
# make sure that attn_weights keeps its gradient.
# In order to do so, attn_weights have to reshaped
# twice and have to be reused in the following
attn_weights_reshaped = attn_weights.view(
bsz, self.num_heads, tgt_len, src_len
)
attn_weights = attn_weights_reshaped.view(
bsz * self.num_heads, tgt_len, src_len
)
else:
attn_weights_reshaped = None
attn_probs = nn.functional.dropout(
attn_weights, p=self.dropout, training=self.training
)
attn_output = torch.bmm(attn_probs, value_states)
if attn_output.size() != (bsz * self.num_heads, tgt_len, self.head_dim):
raise ValueError(
f"`attn_output` should be of size {(bsz, self.num_heads, tgt_len, self.head_dim)}, but is"
f" {attn_output.size()}"
)
attn_output = attn_output.view(bsz, self.num_heads, tgt_len, self.head_dim)
attn_output = attn_output.transpose(1, 2)
attn_output = attn_output.reshape(bsz, tgt_len, self.embed_dim)
attn_output = self.out_proj(attn_output)
return attn_output, attn_weights_reshaped
# Copied from transformers.models.clip.modeling_clip.CLIPMLP with CLIP->IdeficsVision
class IdeficsVisionMLP(nn.Module):
def __init__(self, prefix, config, weights):
super().__init__()
self.config = config
self.activation_fn = ACT2FN[config.hidden_act]
self.fc1 = TensorParallelColumnLinear.load(
config, prefix=f"{prefix}.fc1", weights=weights, bias=True
)
self.fc2 = TensorParallelRowLinear.load(
config, prefix=f"{prefix}.fc2", weights=weights, bias=True
)
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
hidden_states = self.fc1(hidden_states)
hidden_states = self.activation_fn(hidden_states)
hidden_states = self.fc2(hidden_states)
return hidden_states
# Copied from transformers.models.clip.modeling_clip.CLIPEncoderLayer with CLIP->IdeficsVision
class IdeficsVisionEncoderLayer(nn.Module):
def __init__(self, prefix, config, weights):
super().__init__()
self.embed_dim = config.hidden_size
self.self_attn = IdeficsVisionAttention(
prefix=f"{prefix}.self_attn", config=config, weights=weights
)
self.layer_norm1 = nn.LayerNorm.load(
prefix=f"{prefix}.layer_norm1", weights=weights, eps=config.layer_norm_eps
)
self.mlp = IdeficsVisionMLP(
prefix=f"{prefix}.mlp", config=config, weights=weights
)
self.layer_norm2 = nn.LayerNorm.load(
prefix=f"{prefix}.layer_norm2", weights=weights, eps=config.layer_norm_eps
)
def forward(
self,
hidden_states: torch.Tensor,
attention_mask: torch.Tensor,
causal_attention_mask: torch.Tensor,
output_attentions: Optional[bool] = False,
) -> Tuple[torch.FloatTensor]:
"""
Args:
hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
attention_mask (`torch.FloatTensor`): attention mask of size
`(batch, 1, tgt_len, src_len)` where padding elements are indicated by very large negative values.
`(config.encoder_attention_heads,)`.
output_attentions (`bool`, *optional*):
Whether or not to return the attentions tensors of all attention layers. See `attentions` under
returned tensors for more detail.
"""
residual = hidden_states
hidden_states = self.layer_norm1(hidden_states)
hidden_states, attn_weights = self.self_attn(
hidden_states=hidden_states,
attention_mask=attention_mask,
causal_attention_mask=causal_attention_mask,
output_attentions=output_attentions,
)
hidden_states = residual + hidden_states
residual = hidden_states
hidden_states = self.layer_norm2(hidden_states)
hidden_states = self.mlp(hidden_states)
hidden_states = residual + hidden_states
outputs = (hidden_states,)
if output_attentions:
outputs += (attn_weights,)
return outputs
# Copied from transformers.models.clip.modeling_clip.CLIPEncoder with CLIP->IdeficsVision
class IdeficsVisionEncoder(nn.Module):
"""
Transformer encoder consisting of `config.num_hidden_layers` self attention layers. Each layer is a
[`IdeficsVisionEncoderLayer`].
Args:
config: IdeficsVisionConfig
"""
def __init__(self, prefix, config, weights):
super().__init__()
self.config = config
self.layers = nn.ModuleList(
[
IdeficsVisionEncoderLayer(
prefix=f"{prefix}.encoder.layers.{layer_id}",
config=config,
weights=weights,
)
for layer_id in range(config.num_hidden_layers)
]
)
# self.gradient_checkpointing = False
def forward(
self,
inputs_embeds,
attention_mask: Optional[torch.Tensor] = None,
causal_attention_mask: Optional[torch.Tensor] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
return_dict: Optional[bool] = None,
) -> Union[Tuple, BaseModelOutput]:
r"""
Args:
inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`):
Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation.
This is useful if you want more control over how to convert `input_ids` indices into associated vectors
than the model's internal embedding lookup matrix.
attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
- 1 for tokens that are **not masked**,
- 0 for tokens that are **masked**.
[What are attention masks?](../glossary#attention-mask)
causal_attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
Causal mask for the text model. Mask values selected in `[0, 1]`:
- 1 for tokens that are **not masked**,
- 0 for tokens that are **masked**.
[What are attention masks?](../glossary#attention-mask)
output_attentions (`bool`, *optional*):
Whether or not to return the attentions tensors of all attention layers. See `attentions` under
returned tensors for more detail.
output_hidden_states (`bool`, *optional*):
Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors
for more detail.
return_dict (`bool`, *optional*):
Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
"""
output_attentions = (
output_attentions
if output_attentions is not None
else self.config.output_attentions
)
output_hidden_states = (
output_hidden_states
if output_hidden_states is not None
else self.config.output_hidden_states
)
return_dict = (
return_dict if return_dict is not None else self.config.use_return_dict
)
encoder_states = () if output_hidden_states else None
all_attentions = () if output_attentions else None
hidden_states = inputs_embeds
for idx, encoder_layer in enumerate(self.layers):
if output_hidden_states:
encoder_states = encoder_states + (hidden_states,)
# if self.gradient_checkpointing and self.training:
# def create_custom_forward(module):
# def custom_forward(*inputs):
# return module(*inputs, output_attentions)
# return custom_forward
# layer_outputs = torch.utils.checkpoint.checkpoint(
# create_custom_forward(encoder_layer),
# hidden_states,
# attention_mask,
# causal_attention_mask,
# )
# else:
layer_outputs = encoder_layer(
hidden_states,
attention_mask,
causal_attention_mask,
output_attentions=output_attentions,
)
hidden_states = layer_outputs[0]
if output_attentions:
all_attentions = all_attentions + (layer_outputs[1],)
if output_hidden_states:
encoder_states = encoder_states + (hidden_states,)
if not return_dict:
return tuple(
v
for v in [hidden_states, encoder_states, all_attentions]
if v is not None
)
return BaseModelOutput(
last_hidden_state=hidden_states,
hidden_states=encoder_states,
attentions=all_attentions,
)
# Adapted from transformers.models.clip.modeling_clip.CLIPVisionTransformer
class IdeficsVisionTransformer(nn.Module):
def __init__(self, prefix, config, weights):
super().__init__()
self.config = config
embed_dim = config.hidden_size
self.embeddings = IdeficsVisionEmbeddings(
prefix=f"{prefix}.embeddings", config=config, weights=weights
)
self.pre_layrnorm = nn.LayerNorm.load(
prefix=f"{prefix}.pre_layrnorm", weights=weights, eps=config.layer_norm_eps
)
self.encoder = IdeficsVisionEncoder(
prefix=prefix, config=config, weights=weights
)
self.post_layernorm = nn.LayerNorm.load(
prefix=f"{prefix}.post_layernorm",
weights=weights,
eps=config.layer_norm_eps,
)
# copied from transformers.models.clip.modeling_clip.CLIPVisionTransformer.forward
def forward(
self,
pixel_values: Optional[torch.FloatTensor] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
return_dict: Optional[bool] = None,
) -> Union[Tuple, BaseModelOutputWithPooling]:
r"""
Returns:
"""
output_attentions = (
output_attentions
if output_attentions is not None
else self.config.output_attentions
)
output_hidden_states = (
output_hidden_states
if output_hidden_states is not None
else self.config.output_hidden_states
)
return_dict = (
return_dict if return_dict is not None else self.config.use_return_dict
)
if pixel_values is None:
raise ValueError("You have to specify pixel_values")
hidden_states = self.embeddings(pixel_values)
hidden_states = self.pre_layrnorm(hidden_states)
encoder_outputs = self.encoder(
inputs_embeds=hidden_states,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
return_dict=return_dict,
)
last_hidden_state = encoder_outputs[0]
pooled_output = last_hidden_state[:, 0, :]
pooled_output = self.post_layernorm(pooled_output)
if not return_dict:
return (last_hidden_state, pooled_output) + encoder_outputs[1:]
return BaseModelOutputWithPooling(
last_hidden_state=last_hidden_state,
pooler_output=pooled_output,
hidden_states=encoder_outputs.hidden_states,
attentions=encoder_outputs.attentions,
)
| text-generation-inference/server/text_generation_server/models/custom_modeling/idefics_vision.py/0 | {
"file_path": "text-generation-inference/server/text_generation_server/models/custom_modeling/idefics_vision.py",
"repo_id": "text-generation-inference",
"token_count": 9661
} | 200 |
import torch
import torch.distributed
from typing import List, Optional, Tuple
from transformers import (
AutoTokenizer,
AutoConfig,
AutoProcessor,
)
from text_generation_server.models.custom_modeling.idefics_config import IdeficsConfig
from text_generation_server.models.custom_modeling.idefics_processing import (
IdeficsProcessor,
)
from transformers import LlamaTokenizerFast
from text_generation_server.models.custom_modeling.idefics_modeling import (
IdeficsForVisionText2Text,
)
from text_generation_server.models.idefics_causal_lm import IdeficsCausalLM
from text_generation_server.utils import (
initialize_torch_distributed,
weight_files,
Weights,
)
class IDEFICSSharded(IdeficsCausalLM):
def __init__(
self,
model_id: str,
revision: Optional[str] = None,
quantize: Optional[str] = None,
dtype: Optional[torch.dtype] = None,
trust_remote_code: bool = False,
):
self.process_group, rank, world_size = initialize_torch_distributed()
if torch.cuda.is_available():
device = torch.device(f"cuda:{rank}")
# 9b seems to work correctly enough in float16, but 80b seems
# to be really saturating for f16.
dtype = torch.bfloat16 if dtype is None else dtype
else:
device = torch.device("cpu")
dtype = torch.float32 if dtype is None else dtype
self.device, self.dtype = device, dtype
config = IdeficsConfig.from_pretrained(
model_id,
revision=revision,
trust_remote_code=trust_remote_code,
)
config.quantize = quantize
config.vision_config.quantize = quantize
tokenizer = LlamaTokenizerFast.from_pretrained(
model_id,
revision=revision,
padding_side="left",
truncation_side="left",
trust_remote_code=trust_remote_code,
)
self.processor = IdeficsProcessor.from_pretrained(
model_id,
revision=revision,
padding_side="left",
truncation_side="left",
trust_remote_code=trust_remote_code,
)
torch.distributed.barrier(group=self.process_group)
filenames = weight_files(model_id, revision=revision, extension=".safetensors")
weights = Weights(
filenames,
device=device,
dtype=dtype,
process_group=self.process_group,
)
model = IdeficsForVisionText2Text(config, weights)
torch.distributed.barrier(group=self.process_group)
super(IdeficsCausalLM, self).__init__(
model=model,
tokenizer=tokenizer,
requires_padding=True,
dtype=dtype,
device=device,
rank=rank,
world_size=world_size,
)
| text-generation-inference/server/text_generation_server/models/idefics.py/0 | {
"file_path": "text-generation-inference/server/text_generation_server/models/idefics.py",
"repo_id": "text-generation-inference",
"token_count": 1306
} | 201 |
import datetime
import torch
import os
from loguru import logger
from pathlib import Path
from safetensors.torch import save_file, load_file, _find_shared_tensors, _is_complete
from typing import List, Dict
from collections import defaultdict
def _remove_duplicate_names(
state_dict: Dict[str, torch.Tensor],
*,
preferred_names: List[str] = None,
discard_names: List[str] = None,
) -> Dict[str, List[str]]:
if preferred_names is None:
preferred_names = []
preferred_names = set(preferred_names)
if discard_names is None:
discard_names = []
discard_names = set(discard_names)
shareds = _find_shared_tensors(state_dict)
to_remove = defaultdict(list)
for shared in shareds:
complete_names = set(
[name for name in shared if _is_complete(state_dict[name])]
)
if not complete_names:
if len(shared) == 1:
# Force contiguous
name = list(shared)[0]
state_dict[name] = state_dict[name].clone()
complete_names = {name}
else:
raise RuntimeError(
f"Error while trying to find names to remove to save state dict, but found no suitable name to keep for saving amongst: {shared}. None is covering the entire storage.Refusing to save/load the model since you could be storing much more memory than needed. Please refer to https://huggingface.co/docs/safetensors/torch_shared_tensors for more information. Or open an issue."
)
keep_name = sorted(list(complete_names))[0]
# Mecanism to preferentially select keys to keep
# coming from the on-disk file to allow
# loading models saved with a different choice
# of keep_name
preferred = complete_names.difference(discard_names)
if preferred:
keep_name = sorted(list(preferred))[0]
if preferred_names:
preferred = preferred_names.intersection(complete_names)
if preferred:
keep_name = sorted(list(preferred))[0]
for name in sorted(shared):
if name != keep_name:
to_remove[keep_name].append(name)
return to_remove
def convert_file(pt_file: Path, sf_file: Path, discard_names: List[str]):
"""
Convert a pytorch file to a safetensors file
This will remove duplicate tensors from the file.
Unfortunately, this might not respect *transformers* convention.
Forcing us to check for potentially different keys during load when looking
for specific tensors (making tensor sharing explicit).
"""
loaded = torch.load(pt_file, map_location="cpu")
if "state_dict" in loaded:
loaded = loaded["state_dict"]
to_removes = _remove_duplicate_names(loaded, discard_names=discard_names)
metadata = {"format": "pt"}
for kept_name, to_remove_group in to_removes.items():
for to_remove in to_remove_group:
if to_remove not in metadata:
metadata[to_remove] = kept_name
del loaded[to_remove]
# Force tensors to be contiguous
loaded = {k: v.contiguous() for k, v in loaded.items()}
dirname = os.path.dirname(sf_file)
os.makedirs(dirname, exist_ok=True)
save_file(loaded, sf_file, metadata=metadata)
reloaded = load_file(sf_file)
for k in loaded:
pt_tensor = loaded[k]
sf_tensor = reloaded[k]
if not torch.equal(pt_tensor, sf_tensor):
raise RuntimeError(f"The output tensors do not match for key {k}")
def convert_files(pt_files: List[Path], sf_files: List[Path], discard_names: List[str]):
assert len(pt_files) == len(sf_files)
N = len(pt_files)
# We do this instead of using tqdm because we want to parse the logs with the launcher
for i, (pt_file, sf_file) in enumerate(zip(pt_files, sf_files)):
# Skip blacklisted files
if (
"arguments" in pt_file.name
or "args" in pt_file.name
or "training" in pt_file.name
):
continue
start = datetime.datetime.now()
convert_file(pt_file, sf_file, discard_names)
elapsed = datetime.datetime.now() - start
logger.info(f"Convert: [{i + 1}/{N}] -- Took: {elapsed}")
| text-generation-inference/server/text_generation_server/utils/convert.py/0 | {
"file_path": "text-generation-inference/server/text_generation_server/utils/convert.py",
"repo_id": "text-generation-inference",
"token_count": 1769
} | 202 |
SPECULATE = None
def get_speculate() -> int:
global SPECULATE
return SPECULATE
def set_speculate(speculate: int):
global SPECULATE
SPECULATE = speculate
| text-generation-inference/server/text_generation_server/utils/speculate.py/0 | {
"file_path": "text-generation-inference/server/text_generation_server/utils/speculate.py",
"repo_id": "text-generation-inference",
"token_count": 66
} | 203 |
.PHONY: style check-style test
DATA_DIR = data
dir_guard=@mkdir -p $(@D)
# Format source code automatically
style:
npm run lint
# Check the source code is formatted correctly
check-style:
npm run lint-check
TESTS_RESOURCES = $(DATA_DIR)/small.txt $(DATA_DIR)/roberta.json $(DATA_DIR)/tokenizer-wiki.json $(DATA_DIR)/bert-wiki.json
# Launch the test suite
test: $(TESTS_RESOURCES)
npm run test
$(DATA_DIR)/big.txt :
$(dir_guard)
wget https://norvig.com/big.txt -O $@
$(DATA_DIR)/small.txt : $(DATA_DIR)/big.txt
head -100 $(DATA_DIR)/big.txt > $@
$(DATA_DIR)/roberta.json :
$(dir_guard)
wget https://huggingface.co/roberta-large/raw/main/tokenizer.json -O $@
$(DATA_DIR)/tokenizer-wiki.json :
$(dir_guard)
wget https://s3.amazonaws.com/models.huggingface.co/bert/anthony/doc-quicktour/tokenizer.json -O $@
$(DATA_DIR)/bert-wiki.json :
$(dir_guard)
wget https://s3.amazonaws.com/models.huggingface.co/bert/anthony/doc-pipeline/tokenizer.json -O $@
| tokenizers/bindings/node/Makefile/0 | {
"file_path": "tokenizers/bindings/node/Makefile",
"repo_id": "tokenizers",
"token_count": 406
} | 204 |
import {
byteLevelPreTokenizer,
metaspacePreTokenizer,
punctuationPreTokenizer,
sequencePreTokenizer,
splitPreTokenizer,
whitespaceSplitPreTokenizer,
} from '../../'
describe('byteLevelPreTokenizer', () => {
it('instantiates correctly', () => {
const processor = byteLevelPreTokenizer()
expect(processor.constructor.name).toEqual('PreTokenizer')
})
})
describe('metaspacePreTokenizer', () => {
it('instantiates correctly without any parameter', () => {
const processor = metaspacePreTokenizer()
expect(processor.constructor.name).toEqual('PreTokenizer')
})
it('accepts `undefined` as first parameter', () => {
expect(metaspacePreTokenizer(undefined)).toBeDefined()
})
it('accepts `undefined` as second parameter', () => {
expect(metaspacePreTokenizer('t', undefined)).toBeDefined()
})
it('can pre-tokenize strings', () => {
const pretok = metaspacePreTokenizer()
expect(pretok.preTokenizeString('Hello there friend')).toEqual([
['▁Hello', [0, 5]],
['▁there', [5, 11]],
['▁friend', [11, 18]],
])
})
})
describe('punctuationPreTokenizer', () => {
it('instantiates correctly without any parameter', () => {
const processor = punctuationPreTokenizer()
expect(processor.constructor.name).toEqual('PreTokenizer')
})
it('instantiates correctly with non-default split delimeter', () => {
const processor = punctuationPreTokenizer('removed')
expect(processor.constructor.name).toEqual('PreTokenizer')
})
})
describe('splitPreTokenizer', () => {
it('instantiates correctly with invert parameter', () => {
const processor = splitPreTokenizer(' ', 'mergedWithPrevious', false)
expect(processor.constructor.name).toEqual('PreTokenizer')
})
})
describe('sequencePreTokenizer', () => {
it('instantiates correctly', () => {
const punctuation = punctuationPreTokenizer()
const whitespace = whitespaceSplitPreTokenizer()
const sequence2 = sequencePreTokenizer([])
expect(sequence2.constructor.name).toEqual('PreTokenizer')
const sequence3 = sequencePreTokenizer([punctuation, whitespace])
expect(sequence3.constructor.name).toEqual('PreTokenizer')
})
})
| tokenizers/bindings/node/lib/bindings/pre-tokenizers.test.ts/0 | {
"file_path": "tokenizers/bindings/node/lib/bindings/pre-tokenizers.test.ts",
"repo_id": "tokenizers",
"token_count": 728
} | 205 |
{
"name": "tokenizers-linux-arm64-gnu",
"version": "0.13.4-rc1",
"os": [
"linux"
],
"cpu": [
"arm64"
],
"main": "tokenizers.linux-arm64-gnu.node",
"files": [
"tokenizers.linux-arm64-gnu.node"
],
"description": "Tokenizers platform specific bindings",
"keywords": [
"napi-rs",
"NAPI",
"N-API",
"Rust",
"node-addon",
"node-addon-api"
],
"license": "MIT",
"engines": {
"node": ">= 10"
},
"publishConfig": {
"registry": "https://registry.npmjs.org/",
"access": "public"
},
"repository": "tokenizers",
"libc": [
"glibc"
]
} | tokenizers/bindings/node/npm/linux-arm64-gnu/package.json/0 | {
"file_path": "tokenizers/bindings/node/npm/linux-arm64-gnu/package.json",
"repo_id": "tokenizers",
"token_count": 289
} | 206 |
use crate::arc_rwlock_serde;
use serde::{Deserialize, Serialize};
extern crate tokenizers as tk;
use napi::bindgen_prelude::*;
use napi_derive::napi;
use std::sync::{Arc, RwLock};
use tk::decoders::DecoderWrapper;
/// Decoder
#[derive(Clone, Serialize, Deserialize)]
#[napi]
pub struct Decoder {
#[serde(flatten, with = "arc_rwlock_serde")]
decoder: Option<Arc<RwLock<DecoderWrapper>>>,
}
#[napi]
impl Decoder {
#[napi]
pub fn decode(&self, tokens: Vec<String>) -> Result<String> {
use tk::Decoder;
self
.decoder
.as_ref()
.unwrap()
.read()
.unwrap()
.decode(tokens)
.map_err(|e| Error::from_reason(format!("{}", e)))
}
}
impl tk::Decoder for Decoder {
fn decode_chain(&self, tokens: Vec<String>) -> tk::Result<Vec<String>> {
self
.decoder
.as_ref()
.ok_or("Uninitialized Decoder")?
.read()
.unwrap()
.decode_chain(tokens)
}
}
#[napi]
pub fn bpe_decoder(suffix: Option<String>) -> Decoder {
let suffix = suffix.unwrap_or("</w>".to_string());
let decoder = Some(Arc::new(RwLock::new(
tk::decoders::bpe::BPEDecoder::new(suffix).into(),
)));
Decoder { decoder }
}
#[napi]
pub fn byte_fallback_decoder() -> Decoder {
Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::decoders::byte_fallback::ByteFallback::new().into(),
))),
}
}
#[napi]
pub fn ctc_decoder(
#[napi(ts_arg_type = "string = '<pad>'")] pad_token: Option<String>,
word_delimiter_token: Option<String>,
cleanup: Option<bool>,
) -> Decoder {
let pad_token = pad_token.unwrap_or("<pad>".to_string());
let word_delimiter_token = word_delimiter_token.unwrap_or("|".to_string());
let cleanup = cleanup.unwrap_or(true);
let decoder = Some(Arc::new(RwLock::new(
tk::decoders::ctc::CTC::new(pad_token, word_delimiter_token, cleanup).into(),
)));
Decoder { decoder }
}
#[napi]
pub fn fuse_decoder() -> Decoder {
Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::decoders::fuse::Fuse::new().into(),
))),
}
}
#[napi]
pub fn metaspace_decoder(
#[napi(ts_arg_type = "string = '▁'")] replacement: Option<String>,
#[napi(ts_arg_type = "bool = true")] add_prefix_space: Option<bool>,
) -> Result<Decoder> {
let add_prefix_space = add_prefix_space.unwrap_or(true);
let replacement = replacement.unwrap_or("▁".to_string());
if replacement.chars().count() != 1 {
return Err(Error::from_reason(
"replacement is supposed to be a single char",
));
}
let replacement = replacement.chars().next().unwrap();
Ok(Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::decoders::metaspace::Metaspace::new(replacement, add_prefix_space).into(),
))),
})
}
#[napi]
pub fn replace_decoder(pattern: String, content: String) -> Result<Decoder> {
Ok(Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::normalizers::replace::Replace::new(pattern, content)
.map_err(|e| Error::from_reason(e.to_string()))?
.into(),
))),
})
}
#[napi]
pub fn sequence_decoder(decoders: Vec<&Decoder>) -> Decoder {
let sequence: Vec<tk::DecoderWrapper> = decoders
.into_iter()
.filter_map(|decoder| {
decoder
.decoder
.as_ref()
.map(|decoder| (**decoder).read().unwrap().clone())
})
.clone()
.collect();
Decoder {
decoder: Some(Arc::new(RwLock::new(tk::DecoderWrapper::Sequence(
tk::decoders::sequence::Sequence::new(sequence),
)))),
}
}
#[napi]
pub fn strip_decoder(content: String, left: u32, right: u32) -> Result<Decoder> {
let content: char = content.chars().next().ok_or(Error::from_reason(
"Expected non empty string for strip pattern",
))?;
Ok(Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::decoders::strip::Strip::new(content, left as usize, right as usize).into(),
))),
})
}
#[napi]
pub fn word_piece_decoder(
#[napi(ts_arg_type = "string = '##'")] prefix: Option<String>,
#[napi(ts_arg_type = "bool = true")] cleanup: Option<bool>,
) -> Decoder {
let prefix = prefix.unwrap_or("##".to_string());
let cleanup = cleanup.unwrap_or(true);
Decoder {
decoder: Some(Arc::new(RwLock::new(
tk::decoders::wordpiece::WordPiece::new(prefix, cleanup).into(),
))),
}
}
| tokenizers/bindings/node/src/decoders.rs/0 | {
"file_path": "tokenizers/bindings/node/src/decoders.rs",
"repo_id": "tokenizers",
"token_count": 1821
} | 207 |
[target.x86_64-apple-darwin]
rustflags = [
"-C", "link-arg=-undefined",
"-C", "link-arg=dynamic_lookup",
"-C", "link-arg=-mmacosx-version-min=10.11",
]
[target.aarch64-apple-darwin]
rustflags = [
"-C", "link-arg=-undefined",
"-C", "link-arg=dynamic_lookup",
"-C", "link-arg=-mmacosx-version-min=10.11",
]
| tokenizers/bindings/python/.cargo/config.toml/0 | {
"file_path": "tokenizers/bindings/python/.cargo/config.toml",
"repo_id": "tokenizers",
"token_count": 146
} | 208 |
from .. import decoders
Decoder = decoders.Decoder
ByteLevel = decoders.ByteLevel
Replace = decoders.Replace
WordPiece = decoders.WordPiece
ByteFallback = decoders.ByteFallback
Fuse = decoders.Fuse
Strip = decoders.Strip
Metaspace = decoders.Metaspace
BPEDecoder = decoders.BPEDecoder
CTC = decoders.CTC
Sequence = decoders.Sequence
| tokenizers/bindings/python/py_src/tokenizers/decoders/__init__.py/0 | {
"file_path": "tokenizers/bindings/python/py_src/tokenizers/decoders/__init__.py",
"repo_id": "tokenizers",
"token_count": 128
} | 209 |
# Generated content DO NOT EDIT
class PostProcessor:
"""
Base class for all post-processors
This class is not supposed to be instantiated directly. Instead, any implementation of
a PostProcessor will return an instance of this class when instantiated.
"""
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
class BertProcessing(PostProcessor):
"""
This post-processor takes care of adding the special tokens needed by
a Bert model:
- a SEP token
- a CLS token
Args:
sep (:obj:`Tuple[str, int]`):
A tuple with the string representation of the SEP token, and its id
cls (:obj:`Tuple[str, int]`):
A tuple with the string representation of the CLS token, and its id
"""
def __init__(self, sep, cls):
pass
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
class ByteLevel(PostProcessor):
"""
This post-processor takes care of trimming the offsets.
By default, the ByteLevel BPE might include whitespaces in the produced tokens. If you don't
want the offsets to include these whitespaces, then this PostProcessor must be used.
Args:
trim_offsets (:obj:`bool`):
Whether to trim the whitespaces from the produced offsets.
"""
def __init__(self, trim_offsets=True):
pass
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
class RobertaProcessing(PostProcessor):
"""
This post-processor takes care of adding the special tokens needed by
a Roberta model:
- a SEP token
- a CLS token
It also takes care of trimming the offsets.
By default, the ByteLevel BPE might include whitespaces in the produced tokens. If you don't
want the offsets to include these whitespaces, then this PostProcessor should be initialized
with :obj:`trim_offsets=True`
Args:
sep (:obj:`Tuple[str, int]`):
A tuple with the string representation of the SEP token, and its id
cls (:obj:`Tuple[str, int]`):
A tuple with the string representation of the CLS token, and its id
trim_offsets (:obj:`bool`, `optional`, defaults to :obj:`True`):
Whether to trim the whitespaces from the produced offsets.
add_prefix_space (:obj:`bool`, `optional`, defaults to :obj:`True`):
Whether the add_prefix_space option was enabled during pre-tokenization. This
is relevant because it defines the way the offsets are trimmed out.
"""
def __init__(self, sep, cls, trim_offsets=True, add_prefix_space=True):
pass
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
class Sequence(PostProcessor):
"""
Sequence Processor
Args:
processors (:obj:`List[PostProcessor]`)
The processors that need to be chained
"""
def __init__(self, processors):
pass
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
class TemplateProcessing(PostProcessor):
"""
Provides a way to specify templates in order to add the special tokens to each
input sequence as relevant.
Let's take :obj:`BERT` tokenizer as an example. It uses two special tokens, used to
delimitate each sequence. :obj:`[CLS]` is always used at the beginning of the first
sequence, and :obj:`[SEP]` is added at the end of both the first, and the pair
sequences. The final result looks like this:
- Single sequence: :obj:`[CLS] Hello there [SEP]`
- Pair sequences: :obj:`[CLS] My name is Anthony [SEP] What is my name? [SEP]`
With the type ids as following::
[CLS] ... [SEP] ... [SEP]
0 0 0 1 1
You can achieve such behavior using a TemplateProcessing::
TemplateProcessing(
single="[CLS] $0 [SEP]",
pair="[CLS] $A [SEP] $B:1 [SEP]:1",
special_tokens=[("[CLS]", 1), ("[SEP]", 0)],
)
In this example, each input sequence is identified using a ``$`` construct. This identifier
lets us specify each input sequence, and the type_id to use. When nothing is specified,
it uses the default values. Here are the different ways to specify it:
- Specifying the sequence, with default ``type_id == 0``: ``$A`` or ``$B``
- Specifying the `type_id` with default ``sequence == A``: ``$0``, ``$1``, ``$2``, ...
- Specifying both: ``$A:0``, ``$B:1``, ...
The same construct is used for special tokens: ``<identifier>(:<type_id>)?``.
**Warning**: You must ensure that you are giving the correct tokens/ids as these
will be added to the Encoding without any further check. If the given ids correspond
to something totally different in a `Tokenizer` using this `PostProcessor`, it
might lead to unexpected results.
Args:
single (:obj:`Template`):
The template used for single sequences
pair (:obj:`Template`):
The template used when both sequences are specified
special_tokens (:obj:`Tokens`):
The list of special tokens used in each sequences
Types:
Template (:obj:`str` or :obj:`List`):
- If a :obj:`str` is provided, the whitespace is used as delimiter between tokens
- If a :obj:`List[str]` is provided, a list of tokens
Tokens (:obj:`List[Union[Tuple[int, str], Tuple[str, int], dict]]`):
- A :obj:`Tuple` with both a token and its associated ID, in any order
- A :obj:`dict` with the following keys:
- "id": :obj:`str` => The special token id, as specified in the Template
- "ids": :obj:`List[int]` => The associated IDs
- "tokens": :obj:`List[str]` => The associated tokens
The given dict expects the provided :obj:`ids` and :obj:`tokens` lists to have
the same length.
"""
def __init__(self, single, pair, special_tokens):
pass
def num_special_tokens_to_add(self, is_pair):
"""
Return the number of special tokens that would be added for single/pair sentences.
Args:
is_pair (:obj:`bool`):
Whether the input would be a pair of sequences
Returns:
:obj:`int`: The number of tokens to add
"""
pass
def process(self, encoding, pair=None, add_special_tokens=True):
"""
Post-process the given encodings, generating the final one
Args:
encoding (:class:`~tokenizers.Encoding`):
The encoding for the first sequence
pair (:class:`~tokenizers.Encoding`, `optional`):
The encoding for the pair sequence
add_special_tokens (:obj:`bool`):
Whether to add the special tokens
Return:
:class:`~tokenizers.Encoding`: The final encoding
"""
pass
| tokenizers/bindings/python/py_src/tokenizers/processors/__init__.pyi/0 | {
"file_path": "tokenizers/bindings/python/py_src/tokenizers/processors/__init__.pyi",
"repo_id": "tokenizers",
"token_count": 4779
} | 210 |
use std::collections::HashMap;
use std::path::{Path, PathBuf};
use std::sync::{Arc, RwLock};
use crate::token::PyToken;
use crate::trainers::PyTrainer;
use pyo3::exceptions;
use pyo3::prelude::*;
use pyo3::types::*;
use serde::{Deserialize, Serialize};
use tk::models::bpe::{BpeBuilder, Merges, Vocab, BPE};
use tk::models::unigram::Unigram;
use tk::models::wordlevel::WordLevel;
use tk::models::wordpiece::{WordPiece, WordPieceBuilder};
use tk::models::ModelWrapper;
use tk::{Model, Token};
use tokenizers as tk;
use super::error::{deprecation_warning, ToPyResult};
/// Base class for all models
///
/// The model represents the actual tokenization algorithm. This is the part that
/// will contain and manage the learned vocabulary.
///
/// This class cannot be constructed directly. Please use one of the concrete models.
#[pyclass(module = "tokenizers.models", name = "Model", subclass)]
#[derive(Clone, Serialize, Deserialize)]
pub struct PyModel {
#[serde(flatten)]
pub model: Arc<RwLock<ModelWrapper>>,
}
impl PyModel {
pub(crate) fn get_as_subtype(&self, py: Python<'_>) -> PyResult<PyObject> {
let base = self.clone();
Ok(match *self.model.as_ref().read().unwrap() {
ModelWrapper::BPE(_) => Py::new(py, (PyBPE {}, base))?.into_py(py),
ModelWrapper::WordPiece(_) => Py::new(py, (PyWordPiece {}, base))?.into_py(py),
ModelWrapper::WordLevel(_) => Py::new(py, (PyWordLevel {}, base))?.into_py(py),
ModelWrapper::Unigram(_) => Py::new(py, (PyUnigram {}, base))?.into_py(py),
})
}
}
impl Model for PyModel {
type Trainer = PyTrainer;
fn tokenize(&self, tokens: &str) -> tk::Result<Vec<Token>> {
self.model.read().unwrap().tokenize(tokens)
}
fn token_to_id(&self, token: &str) -> Option<u32> {
self.model.read().unwrap().token_to_id(token)
}
fn id_to_token(&self, id: u32) -> Option<String> {
self.model.read().unwrap().id_to_token(id)
}
fn get_vocab(&self) -> HashMap<String, u32> {
self.model.read().unwrap().get_vocab()
}
fn get_vocab_size(&self) -> usize {
self.model.read().unwrap().get_vocab_size()
}
fn save(&self, folder: &Path, name: Option<&str>) -> tk::Result<Vec<PathBuf>> {
self.model.read().unwrap().save(folder, name)
}
fn get_trainer(&self) -> Self::Trainer {
self.model.read().unwrap().get_trainer().into()
}
}
impl<I> From<I> for PyModel
where
I: Into<ModelWrapper>,
{
fn from(model: I) -> Self {
Self {
model: Arc::new(RwLock::new(model.into())),
}
}
}
#[pymethods]
impl PyModel {
#[new]
#[pyo3(text_signature = None)]
fn __new__() -> Self {
// Instantiate a default empty model. This doesn't really make sense, but we need
// to be able to instantiate an empty model for pickle capabilities.
PyModel {
model: Arc::new(RwLock::new(BPE::default().into())),
}
}
fn __getstate__(&self, py: Python) -> PyResult<PyObject> {
let data = serde_json::to_string(&self.model).map_err(|e| {
exceptions::PyException::new_err(format!(
"Error while attempting to pickle Model: {}",
e
))
})?;
Ok(PyBytes::new(py, data.as_bytes()).to_object(py))
}
fn __setstate__(&mut self, py: Python, state: PyObject) -> PyResult<()> {
match state.extract::<&PyBytes>(py) {
Ok(s) => {
self.model = serde_json::from_slice(s.as_bytes()).map_err(|e| {
exceptions::PyException::new_err(format!(
"Error while attempting to unpickle Model: {}",
e
))
})?;
Ok(())
}
Err(e) => Err(e),
}
}
/// Tokenize a sequence
///
/// Args:
/// sequence (:obj:`str`):
/// A sequence to tokenize
///
/// Returns:
/// A :obj:`List` of :class:`~tokenizers.Token`: The generated tokens
#[pyo3(text_signature = "(self, sequence)")]
fn tokenize(&self, sequence: &str) -> PyResult<Vec<PyToken>> {
Ok(ToPyResult(self.model.read().unwrap().tokenize(sequence))
.into_py()?
.into_iter()
.map(|t| t.into())
.collect())
}
/// Get the ID associated to a token
///
/// Args:
/// token (:obj:`str`):
/// A token to convert to an ID
///
/// Returns:
/// :obj:`int`: The ID associated to the token
#[pyo3(text_signature = "(self, tokens)")]
fn token_to_id(&self, token: &str) -> Option<u32> {
self.model.read().unwrap().token_to_id(token)
}
/// Get the token associated to an ID
///
/// Args:
/// id (:obj:`int`):
/// An ID to convert to a token
///
/// Returns:
/// :obj:`str`: The token associated to the ID
#[pyo3(text_signature = "(self, id)")]
fn id_to_token(&self, id: u32) -> Option<String> {
self.model.read().unwrap().id_to_token(id)
}
/// Save the current model
///
/// Save the current model in the given folder, using the given prefix for the various
/// files that will get created.
/// Any file with the same name that already exists in this folder will be overwritten.
///
/// Args:
/// folder (:obj:`str`):
/// The path to the target folder in which to save the various files
///
/// prefix (:obj:`str`, `optional`):
/// An optional prefix, used to prefix each file name
///
/// Returns:
/// :obj:`List[str]`: The list of saved files
#[pyo3(text_signature = "(self, folder, prefix)")]
fn save<'a>(
&self,
py: Python<'_>,
folder: &str,
mut prefix: Option<&'a str>,
name: Option<&'a str>,
) -> PyResult<Vec<String>> {
if name.is_some() {
deprecation_warning(
py,
"0.10.0",
"Parameter `name` of Model.save has been renamed `prefix`",
)?;
if prefix.is_none() {
prefix = name;
}
}
let saved: PyResult<Vec<_>> =
ToPyResult(self.model.read().unwrap().save(Path::new(folder), prefix)).into();
Ok(saved?
.into_iter()
.map(|path| path.to_string_lossy().into_owned())
.collect())
}
/// Get the associated :class:`~tokenizers.trainers.Trainer`
///
/// Retrieve the :class:`~tokenizers.trainers.Trainer` associated to this
/// :class:`~tokenizers.models.Model`.
///
/// Returns:
/// :class:`~tokenizers.trainers.Trainer`: The Trainer used to train this model
#[pyo3(text_signature = "(self)")]
fn get_trainer(&self, py: Python<'_>) -> PyResult<PyObject> {
PyTrainer::from(self.model.read().unwrap().get_trainer()).get_as_subtype(py)
}
}
/// An implementation of the BPE (Byte-Pair Encoding) algorithm
///
/// Args:
/// vocab (:obj:`Dict[str, int]`, `optional`):
/// A dictionnary of string keys and their ids :obj:`{"am": 0,...}`
///
/// merges (:obj:`List[Tuple[str, str]]`, `optional`):
/// A list of pairs of tokens (:obj:`Tuple[str, str]`) :obj:`[("a", "b"),...]`
///
/// cache_capacity (:obj:`int`, `optional`):
/// The number of words that the BPE cache can contain. The cache allows
/// to speed-up the process by keeping the result of the merge operations
/// for a number of words.
///
/// dropout (:obj:`float`, `optional`):
/// A float between 0 and 1 that represents the BPE dropout to use.
///
/// unk_token (:obj:`str`, `optional`):
/// The unknown token to be used by the model.
///
/// continuing_subword_prefix (:obj:`str`, `optional`):
/// The prefix to attach to subword units that don't represent a beginning of word.
///
/// end_of_word_suffix (:obj:`str`, `optional`):
/// The suffix to attach to subword units that represent an end of word.
///
/// fuse_unk (:obj:`bool`, `optional`):
/// Whether to fuse any subsequent unknown tokens into a single one
///
/// byte_fallback (:obj:`bool`, `optional`):
/// Whether to use spm byte-fallback trick (defaults to False)
#[pyclass(extends=PyModel, module = "tokenizers.models", name = "BPE")]
pub struct PyBPE {}
impl PyBPE {
fn with_builder(mut builder: BpeBuilder, kwargs: Option<&PyDict>) -> PyResult<(Self, PyModel)> {
if let Some(kwargs) = kwargs {
for (key, value) in kwargs {
let key: &str = key.extract()?;
match key {
"cache_capacity" => builder = builder.cache_capacity(value.extract()?),
"dropout" => {
if let Some(dropout) = value.extract()? {
builder = builder.dropout(dropout);
}
}
"unk_token" => {
if let Some(unk) = value.extract()? {
builder = builder.unk_token(unk);
}
}
"continuing_subword_prefix" => {
builder = builder.continuing_subword_prefix(value.extract()?)
}
"end_of_word_suffix" => builder = builder.end_of_word_suffix(value.extract()?),
"fuse_unk" => builder = builder.fuse_unk(value.extract()?),
"byte_fallback" => builder = builder.byte_fallback(value.extract()?),
_ => println!("Ignored unknown kwarg option {}", key),
};
}
}
match builder.build() {
Err(e) => Err(exceptions::PyException::new_err(format!(
"Error while initializing BPE: {}",
e
))),
Ok(bpe) => Ok((PyBPE {}, bpe.into())),
}
}
}
macro_rules! getter {
($self: ident, $variant: ident, $($name: tt)+) => {{
let super_ = $self.as_ref();
let model = super_.model.read().unwrap();
if let ModelWrapper::$variant(ref mo) = *model {
mo.$($name)+
} else {
unreachable!()
}
}};
}
macro_rules! setter {
($self: ident, $variant: ident, $name: ident, $value: expr) => {{
let super_ = $self.as_ref();
let mut model = super_.model.write().unwrap();
if let ModelWrapper::$variant(ref mut mo) = *model {
mo.$name = $value;
}
}};
}
#[derive(FromPyObject)]
enum PyVocab<'a> {
Vocab(Vocab),
Filename(&'a str),
}
#[derive(FromPyObject)]
enum PyMerges<'a> {
Merges(Merges),
Filename(&'a str),
}
#[pymethods]
impl PyBPE {
#[getter]
fn get_dropout(self_: PyRef<Self>) -> Option<f32> {
getter!(self_, BPE, dropout)
}
#[setter]
fn set_dropout(self_: PyRef<Self>, dropout: Option<f32>) {
setter!(self_, BPE, dropout, dropout);
}
#[getter]
fn get_unk_token(self_: PyRef<Self>) -> Option<String> {
getter!(self_, BPE, unk_token.clone())
}
#[setter]
fn set_unk_token(self_: PyRef<Self>, unk_token: Option<String>) {
setter!(self_, BPE, unk_token, unk_token);
}
#[getter]
fn get_continuing_subword_prefix(self_: PyRef<Self>) -> Option<String> {
getter!(self_, BPE, continuing_subword_prefix.clone())
}
#[setter]
fn set_continuing_subword_prefix(
self_: PyRef<Self>,
continuing_subword_prefix: Option<String>,
) {
setter!(
self_,
BPE,
continuing_subword_prefix,
continuing_subword_prefix
);
}
#[getter]
fn get_end_of_word_suffix(self_: PyRef<Self>) -> Option<String> {
getter!(self_, BPE, end_of_word_suffix.clone())
}
#[setter]
fn set_end_of_word_suffix(self_: PyRef<Self>, end_of_word_suffix: Option<String>) {
setter!(self_, BPE, end_of_word_suffix, end_of_word_suffix);
}
#[getter]
fn get_fuse_unk(self_: PyRef<Self>) -> bool {
getter!(self_, BPE, fuse_unk)
}
#[setter]
fn set_fuse_unk(self_: PyRef<Self>, fuse_unk: bool) {
setter!(self_, BPE, fuse_unk, fuse_unk);
}
#[getter]
fn get_byte_fallback(self_: PyRef<Self>) -> bool {
getter!(self_, BPE, byte_fallback)
}
#[setter]
fn set_byte_fallback(self_: PyRef<Self>, byte_fallback: bool) {
setter!(self_, BPE, byte_fallback, byte_fallback);
}
#[new]
#[pyo3(
signature = (vocab=None, merges=None, **kwargs),
text_signature = "(self, vocab=None, merges=None, cache_capacity=None, dropout=None, unk_token=None, continuing_subword_prefix=None, end_of_word_suffix=None, fuse_unk=None, byte_fallback=False)")]
fn new(
py: Python<'_>,
vocab: Option<PyVocab>,
merges: Option<PyMerges>,
kwargs: Option<&PyDict>,
) -> PyResult<(Self, PyModel)> {
if (vocab.is_some() && merges.is_none()) || (vocab.is_none() && merges.is_some()) {
return Err(exceptions::PyValueError::new_err(
"`vocab` and `merges` must be both specified",
));
}
let mut builder = BPE::builder();
if let (Some(vocab), Some(merges)) = (vocab, merges) {
match (vocab, merges) {
(PyVocab::Vocab(vocab), PyMerges::Merges(merges)) => {
builder = builder.vocab_and_merges(vocab, merges);
}
(PyVocab::Filename(vocab_filename), PyMerges::Filename(merges_filename)) => {
deprecation_warning(
py,
"0.9.0",
"BPE.__init__ will not create from files anymore, try `BPE.from_file` instead",
)?;
builder =
builder.files(vocab_filename.to_string(), merges_filename.to_string());
}
_ => {
return Err(exceptions::PyValueError::new_err(
"`vocab` and `merges` must be both be from memory or both filenames",
));
}
}
}
PyBPE::with_builder(builder, kwargs)
}
/// Read a :obj:`vocab.json` and a :obj:`merges.txt` files
///
/// This method provides a way to read and parse the content of these files,
/// returning the relevant data structures. If you want to instantiate some BPE models
/// from memory, this method gives you the expected input from the standard files.
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.json` file
///
/// merges (:obj:`str`):
/// The path to a :obj:`merges.txt` file
///
/// Returns:
/// A :obj:`Tuple` with the vocab and the merges:
/// The vocabulary and merges loaded into memory
#[staticmethod]
#[pyo3(text_signature = "(self, vocab, merges)")]
fn read_file(vocab: &str, merges: &str) -> PyResult<(Vocab, Merges)> {
BPE::read_file(vocab, merges).map_err(|e| {
exceptions::PyException::new_err(format!(
"Error while reading vocab & merges files: {}",
e
))
})
}
/// Instantiate a BPE model from the given files.
///
/// This method is roughly equivalent to doing::
///
/// vocab, merges = BPE.read_file(vocab_filename, merges_filename)
/// bpe = BPE(vocab, merges)
///
/// If you don't need to keep the :obj:`vocab, merges` values lying around,
/// this method is more optimized than manually calling
/// :meth:`~tokenizers.models.BPE.read_file` to initialize a :class:`~tokenizers.models.BPE`
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.json` file
///
/// merges (:obj:`str`):
/// The path to a :obj:`merges.txt` file
///
/// Returns:
/// :class:`~tokenizers.models.BPE`: An instance of BPE loaded from these files
#[classmethod]
#[pyo3(signature = (vocab, merges, **kwargs))]
#[pyo3(text_signature = "(cls, vocab, merge, **kwargs)")]
fn from_file(
_cls: &PyType,
py: Python,
vocab: &str,
merges: &str,
kwargs: Option<&PyDict>,
) -> PyResult<Py<Self>> {
let (vocab, merges) = BPE::read_file(vocab, merges).map_err(|e| {
exceptions::PyException::new_err(format!("Error while reading BPE files: {}", e))
})?;
Py::new(
py,
PyBPE::new(
py,
Some(PyVocab::Vocab(vocab)),
Some(PyMerges::Merges(merges)),
kwargs,
)?,
)
}
}
/// An implementation of the WordPiece algorithm
///
/// Args:
/// vocab (:obj:`Dict[str, int]`, `optional`):
/// A dictionnary of string keys and their ids :obj:`{"am": 0,...}`
///
/// unk_token (:obj:`str`, `optional`):
/// The unknown token to be used by the model.
///
/// max_input_chars_per_word (:obj:`int`, `optional`):
/// The maximum number of characters to authorize in a single word.
#[pyclass(extends=PyModel, module = "tokenizers.models", name = "WordPiece")]
pub struct PyWordPiece {}
impl PyWordPiece {
fn with_builder(
mut builder: WordPieceBuilder,
kwargs: Option<&PyDict>,
) -> PyResult<(Self, PyModel)> {
if let Some(kwargs) = kwargs {
for (key, val) in kwargs {
let key: &str = key.extract()?;
match key {
"unk_token" => {
builder = builder.unk_token(val.extract()?);
}
"max_input_chars_per_word" => {
builder = builder.max_input_chars_per_word(val.extract()?);
}
"continuing_subword_prefix" => {
builder = builder.continuing_subword_prefix(val.extract()?);
}
_ => println!("Ignored unknown kwargs option {}", key),
}
}
}
match builder.build() {
Err(e) => Err(exceptions::PyException::new_err(format!(
"Error while initializing WordPiece: {}",
e
))),
Ok(wordpiece) => Ok((PyWordPiece {}, wordpiece.into())),
}
}
}
#[pymethods]
impl PyWordPiece {
#[getter]
fn get_unk_token(self_: PyRef<Self>) -> String {
getter!(self_, WordPiece, unk_token.clone())
}
#[setter]
fn set_unk_token(self_: PyRef<Self>, unk_token: String) {
setter!(self_, WordPiece, unk_token, unk_token);
}
#[getter]
fn get_continuing_subword_prefix(self_: PyRef<Self>) -> String {
getter!(self_, WordPiece, continuing_subword_prefix.clone())
}
#[setter]
fn set_continuing_subword_prefix(self_: PyRef<Self>, continuing_subword_prefix: String) {
setter!(
self_,
WordPiece,
continuing_subword_prefix,
continuing_subword_prefix
);
}
#[getter]
fn get_max_input_chars_per_word(self_: PyRef<Self>) -> usize {
getter!(self_, WordPiece, max_input_chars_per_word)
}
#[setter]
fn set_max_input_chars_per_word(self_: PyRef<Self>, max: usize) {
setter!(self_, WordPiece, max_input_chars_per_word, max);
}
#[new]
#[pyo3(signature = (vocab=None, **kwargs), text_signature = "(self, vocab, unk_token, max_input_chars_per_word)")]
fn new(
py: Python<'_>,
vocab: Option<PyVocab>,
kwargs: Option<&PyDict>,
) -> PyResult<(Self, PyModel)> {
let mut builder = WordPiece::builder();
if let Some(vocab) = vocab {
match vocab {
PyVocab::Vocab(vocab) => {
builder = builder.vocab(vocab);
}
PyVocab::Filename(vocab_filename) => {
deprecation_warning(
py,
"0.9.0",
"WordPiece.__init__ will not create from files anymore, try `WordPiece.from_file` instead",
)?;
builder = builder.files(vocab_filename.to_string());
}
}
}
PyWordPiece::with_builder(builder, kwargs)
}
/// Read a :obj:`vocab.txt` file
///
/// This method provides a way to read and parse the content of a standard `vocab.txt`
/// file as used by the WordPiece Model, returning the relevant data structures. If you
/// want to instantiate some WordPiece models from memory, this method gives you the
/// expected input from the standard files.
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.txt` file
///
/// Returns:
/// :obj:`Dict[str, int]`: The vocabulary as a :obj:`dict`
#[staticmethod]
#[pyo3(text_signature = "(vocab)")]
fn read_file(vocab: &str) -> PyResult<Vocab> {
WordPiece::read_file(vocab).map_err(|e| {
exceptions::PyException::new_err(format!("Error while reading WordPiece file: {}", e))
})
}
/// Instantiate a WordPiece model from the given file
///
/// This method is roughly equivalent to doing::
///
/// vocab = WordPiece.read_file(vocab_filename)
/// wordpiece = WordPiece(vocab)
///
/// If you don't need to keep the :obj:`vocab` values lying around, this method is
/// more optimized than manually calling :meth:`~tokenizers.models.WordPiece.read_file` to
/// initialize a :class:`~tokenizers.models.WordPiece`
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.txt` file
///
/// Returns:
/// :class:`~tokenizers.models.WordPiece`: An instance of WordPiece loaded from file
#[classmethod]
#[pyo3(signature = (vocab, **kwargs))]
#[pyo3(text_signature = "(vocab, **kwargs)")]
fn from_file(
_cls: &PyType,
py: Python,
vocab: &str,
kwargs: Option<&PyDict>,
) -> PyResult<Py<Self>> {
let vocab = WordPiece::read_file(vocab).map_err(|e| {
exceptions::PyException::new_err(format!("Error while reading WordPiece file: {}", e))
})?;
Py::new(
py,
PyWordPiece::new(py, Some(PyVocab::Vocab(vocab)), kwargs)?,
)
}
}
/// An implementation of the WordLevel algorithm
///
/// Most simple tokenizer model based on mapping tokens to their corresponding id.
///
/// Args:
/// vocab (:obj:`str`, `optional`):
/// A dictionnary of string keys and their ids :obj:`{"am": 0,...}`
///
/// unk_token (:obj:`str`, `optional`):
/// The unknown token to be used by the model.
#[pyclass(extends=PyModel, module = "tokenizers.models", name = "WordLevel")]
pub struct PyWordLevel {}
#[pymethods]
impl PyWordLevel {
#[getter]
fn get_unk_token(self_: PyRef<Self>) -> String {
getter!(self_, WordLevel, unk_token.clone())
}
#[setter]
fn set_unk_token(self_: PyRef<Self>, unk_token: String) {
setter!(self_, WordLevel, unk_token, unk_token);
}
#[new]
#[pyo3(signature = (vocab=None, unk_token = None), text_signature = "(self, vocab, unk_token)")]
fn new(
py: Python<'_>,
vocab: Option<PyVocab>,
unk_token: Option<String>,
) -> PyResult<(Self, PyModel)> {
let mut builder = WordLevel::builder();
if let Some(vocab) = vocab {
match vocab {
PyVocab::Vocab(vocab) => {
builder = builder.vocab(vocab);
}
PyVocab::Filename(vocab_filename) => {
deprecation_warning(
py,
"0.9.0",
"WordLevel.__init__ will not create from files anymore, \
try `WordLevel.from_file` instead",
)?;
builder = builder.files(vocab_filename.to_string());
}
};
}
if let Some(unk_token) = unk_token {
builder = builder.unk_token(unk_token);
}
Ok((
PyWordLevel {},
builder
.build()
.map_err(|e| exceptions::PyException::new_err(e.to_string()))?
.into(),
))
}
/// Read a :obj:`vocab.json`
///
/// This method provides a way to read and parse the content of a vocabulary file,
/// returning the relevant data structures. If you want to instantiate some WordLevel models
/// from memory, this method gives you the expected input from the standard files.
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.json` file
///
/// Returns:
/// :obj:`Dict[str, int]`: The vocabulary as a :obj:`dict`
#[staticmethod]
#[pyo3(text_signature = "(vocab)")]
fn read_file(vocab: &str) -> PyResult<Vocab> {
WordLevel::read_file(vocab).map_err(|e| {
exceptions::PyException::new_err(format!("Error while reading WordLevel file: {}", e))
})
}
/// Instantiate a WordLevel model from the given file
///
/// This method is roughly equivalent to doing::
///
/// vocab = WordLevel.read_file(vocab_filename)
/// wordlevel = WordLevel(vocab)
///
/// If you don't need to keep the :obj:`vocab` values lying around, this method is
/// more optimized than manually calling :meth:`~tokenizers.models.WordLevel.read_file` to
/// initialize a :class:`~tokenizers.models.WordLevel`
///
/// Args:
/// vocab (:obj:`str`):
/// The path to a :obj:`vocab.json` file
///
/// Returns:
/// :class:`~tokenizers.models.WordLevel`: An instance of WordLevel loaded from file
#[classmethod]
#[pyo3(signature = (vocab, unk_token = None))]
#[pyo3(text_signature = "(vocab, unk_token)")]
fn from_file(
_cls: &PyType,
py: Python,
vocab: &str,
unk_token: Option<String>,
) -> PyResult<Py<Self>> {
let vocab = WordLevel::read_file(vocab).map_err(|e| {
exceptions::PyException::new_err(format!("Error while reading WordLevel file: {}", e))
})?;
Py::new(
py,
PyWordLevel::new(py, Some(PyVocab::Vocab(vocab)), unk_token)?,
)
}
}
/// An implementation of the Unigram algorithm
///
/// Args:
/// vocab (:obj:`List[Tuple[str, float]]`, `optional`, `optional`):
/// A list of vocabulary items and their relative score [("am", -0.2442),...]
#[pyclass(extends=PyModel, module = "tokenizers.models", name = "Unigram")]
pub struct PyUnigram {}
#[pymethods]
impl PyUnigram {
#[new]
#[pyo3(text_signature = "(self, vocab, unk_id, byte_fallback)")]
fn new(
vocab: Option<Vec<(String, f64)>>,
unk_id: Option<usize>,
byte_fallback: Option<bool>,
) -> PyResult<(Self, PyModel)> {
match (vocab, unk_id, byte_fallback) {
(Some(vocab), unk_id, byte_fallback) => {
let model =
Unigram::from(vocab, unk_id, byte_fallback.unwrap_or(false)).map_err(|e| {
exceptions::PyException::new_err(format!(
"Error while loading Unigram: {}",
e
))
})?;
Ok((PyUnigram {}, model.into()))
}
(None, None, _) => Ok((PyUnigram {}, Unigram::default().into())),
_ => Err(exceptions::PyValueError::new_err(
"`vocab` and `unk_id` must be both specified",
)),
}
}
}
/// Models Module
#[pymodule]
pub fn models(_py: Python, m: &PyModule) -> PyResult<()> {
m.add_class::<PyModel>()?;
m.add_class::<PyBPE>()?;
m.add_class::<PyWordPiece>()?;
m.add_class::<PyWordLevel>()?;
m.add_class::<PyUnigram>()?;
Ok(())
}
#[cfg(test)]
mod test {
use crate::models::PyModel;
use pyo3::prelude::*;
use tk::models::bpe::BPE;
use tk::models::ModelWrapper;
#[test]
fn get_subtype() {
Python::with_gil(|py| {
let py_model = PyModel::from(BPE::default());
let py_bpe = py_model.get_as_subtype(py).unwrap();
assert_eq!("BPE", py_bpe.as_ref(py).get_type().name().unwrap());
})
}
#[test]
fn serialize() {
let rs_bpe = BPE::default();
let rs_bpe_ser = serde_json::to_string(&rs_bpe).unwrap();
let rs_wrapper: ModelWrapper = rs_bpe.into();
let rs_wrapper_ser = serde_json::to_string(&rs_wrapper).unwrap();
let py_model = PyModel::from(rs_wrapper);
let py_ser = serde_json::to_string(&py_model).unwrap();
assert_eq!(py_ser, rs_bpe_ser);
assert_eq!(py_ser, rs_wrapper_ser);
let py_model: PyModel = serde_json::from_str(&rs_bpe_ser).unwrap();
match *py_model.model.as_ref().read().unwrap() {
ModelWrapper::BPE(_) => (),
_ => panic!("Expected Bert postprocessor."),
};
let py_model: PyModel = serde_json::from_str(&rs_wrapper_ser).unwrap();
match *py_model.model.as_ref().read().unwrap() {
ModelWrapper::BPE(_) => (),
_ => panic!("Expected Bert postprocessor."),
};
}
}
| tokenizers/bindings/python/src/models.rs/0 | {
"file_path": "tokenizers/bindings/python/src/models.rs",
"repo_id": "tokenizers",
"token_count": 14445
} | 211 |
import json
import pickle
import pytest
from tokenizers.decoders import (
CTC,
BPEDecoder,
ByteLevel,
Decoder,
Metaspace,
Sequence,
WordPiece,
ByteFallback,
Replace,
Strip,
Fuse,
)
class TestByteLevel:
def test_instantiate(self):
assert ByteLevel() is not None
assert isinstance(ByteLevel(), Decoder)
assert isinstance(ByteLevel(), ByteLevel)
assert isinstance(pickle.loads(pickle.dumps(ByteLevel())), ByteLevel)
def test_decoding(self):
decoder = ByteLevel()
assert decoder.decode(["My", "Ġname", "Ġis", "ĠJohn"]) == "My name is John"
def test_manual_reload(self):
byte_level = ByteLevel()
state = json.loads(byte_level.__getstate__())
reloaded = ByteLevel(**state)
assert isinstance(reloaded, ByteLevel)
class TestReplace:
def test_instantiate(self):
assert Replace("_", " ") is not None
assert isinstance(Replace("_", " "), Decoder)
assert isinstance(Replace("_", " "), Replace)
# assert isinstance(pickle.loads(pickle.dumps(Replace("_", " "))), Replace)
def test_decoding(self):
decoder = Replace("_", " ")
assert decoder.decode(["My", "_name", "_is", "_John"]) == "My name is John"
class TestWordPiece:
def test_instantiate(self):
assert WordPiece() is not None
assert WordPiece(prefix="__") is not None
assert WordPiece(cleanup=True) is not None
assert isinstance(WordPiece(), Decoder)
assert isinstance(WordPiece(), WordPiece)
assert isinstance(pickle.loads(pickle.dumps(WordPiece())), WordPiece)
def test_decoding(self):
decoder = WordPiece()
assert decoder.decode(["My", "na", "##me", "is", "Jo", "##hn"]) == "My name is John"
assert decoder.decode(["I", "'m", "Jo", "##hn"]) == "I'm John"
decoder = WordPiece(prefix="__", cleanup=False)
assert decoder.decode(["My", "na", "__me", "is", "Jo", "__hn"]) == "My name is John"
assert decoder.decode(["I", "'m", "Jo", "__hn"]) == "I 'm John"
def test_can_modify(self):
decoder = WordPiece(prefix="$$", cleanup=False)
assert decoder.prefix == "$$"
assert decoder.cleanup == False
# Modify these
decoder.prefix = "__"
assert decoder.prefix == "__"
decoder.cleanup = True
assert decoder.cleanup == True
class TestByteFallback:
def test_instantiate(self):
assert ByteFallback() is not None
assert isinstance(ByteFallback(), Decoder)
assert isinstance(ByteFallback(), ByteFallback)
assert isinstance(pickle.loads(pickle.dumps(ByteFallback())), ByteFallback)
def test_decoding(self):
decoder = ByteFallback()
assert decoder.decode(["My", " na", "me"]) == "My name"
assert decoder.decode(["<0x61>"]) == "a"
assert decoder.decode(["<0xE5>"]) == "�"
assert decoder.decode(["<0xE5>", "<0x8f>"]) == "��"
assert decoder.decode(["<0xE5>", "<0x8f>", "<0xab>"]) == "叫"
assert decoder.decode(["<0xE5>", "<0x8f>", "a"]) == "��a"
assert decoder.decode(["<0xE5>", "<0x8f>", "<0xab>", "a"]) == "叫a"
class TestFuse:
def test_instantiate(self):
assert Fuse() is not None
assert isinstance(Fuse(), Decoder)
assert isinstance(Fuse(), Fuse)
assert isinstance(pickle.loads(pickle.dumps(Fuse())), Fuse)
def test_decoding(self):
decoder = Fuse()
assert decoder.decode(["My", " na", "me"]) == "My name"
class TestStrip:
def test_instantiate(self):
assert Strip(left=0, right=0) is not None
assert isinstance(Strip(content="_", left=0, right=0), Decoder)
assert isinstance(Strip(content="_", left=0, right=0), Strip)
assert isinstance(pickle.loads(pickle.dumps(Strip(content="_", left=0, right=0))), Strip)
def test_decoding(self):
decoder = Strip(content="_", left=1, right=0)
assert decoder.decode(["_My", " na", "me", " _-", "__-"]) == "My name _-_-"
class TestMetaspace:
def test_instantiate(self):
assert Metaspace() is not None
assert Metaspace(replacement="-") is not None
with pytest.raises(ValueError, match="expected a string of length 1"):
Metaspace(replacement="")
assert Metaspace(add_prefix_space=True) is not None
assert isinstance(Metaspace(), Decoder)
assert isinstance(Metaspace(), Metaspace)
assert isinstance(pickle.loads(pickle.dumps(Metaspace())), Metaspace)
def test_decoding(self):
decoder = Metaspace()
assert decoder.decode(["▁My", "▁name", "▁is", "▁John"]) == "My name is John"
decoder = Metaspace(replacement="-", add_prefix_space=False)
assert decoder.decode(["-My", "-name", "-is", "-John"]) == " My name is John"
def test_can_modify(self):
decoder = Metaspace(replacement="*", add_prefix_space=False)
assert decoder.replacement == "*"
assert decoder.add_prefix_space == False
# Modify these
decoder.replacement = "&"
assert decoder.replacement == "&"
decoder.add_prefix_space = True
assert decoder.add_prefix_space == True
class TestBPEDecoder:
def test_instantiate(self):
assert BPEDecoder() is not None
assert BPEDecoder(suffix="_") is not None
assert isinstance(BPEDecoder(), Decoder)
assert isinstance(BPEDecoder(), BPEDecoder)
assert isinstance(pickle.loads(pickle.dumps(BPEDecoder())), BPEDecoder)
def test_decoding(self):
decoder = BPEDecoder()
assert decoder.decode(["My</w>", "na", "me</w>", "is</w>", "Jo", "hn</w>"]) == "My name is John"
decoder = BPEDecoder(suffix="_")
assert decoder.decode(["My_", "na", "me_", "is_", "Jo", "hn_"]) == "My name is John"
def test_can_modify(self):
decoder = BPEDecoder(suffix="123")
assert decoder.suffix == "123"
# Modify these
decoder.suffix = "</w>"
assert decoder.suffix == "</w>"
class TestCTCDecoder:
def test_instantiate(self):
assert CTC() is not None
assert CTC(pad_token="[PAD]") is not None
assert isinstance(CTC(), Decoder)
assert isinstance(CTC(), CTC)
assert isinstance(pickle.loads(pickle.dumps(CTC())), CTC)
def test_decoding(self):
decoder = CTC()
assert (
decoder.decode(["<pad>", "<pad>", "h", "e", "e", "l", "l", "<pad>", "l", "o", "o", "o", "<pad>"])
== "hello"
)
decoder = CTC(pad_token="[PAD]")
assert (
decoder.decode(["[PAD]", "[PAD]", "h", "e", "e", "l", "l", "[PAD]", "l", "o", "o", "o", "[PAD]"])
== "hello"
)
def test_can_modify(self):
decoder = CTC(pad_token="[PAD]")
assert decoder.pad_token == "[PAD]"
assert decoder.word_delimiter_token == "|"
assert decoder.cleanup == True
# Modify these
decoder.pad_token = "{pad}"
assert decoder.pad_token == "{pad}"
decoder.word_delimiter_token = "_"
assert decoder.word_delimiter_token == "_"
decoder.cleanup = False
assert decoder.cleanup == False
class TestSequenceDecoder:
def test_instantiate(self):
assert Sequence([]) is not None
assert Sequence([CTC()]) is not None
assert isinstance(Sequence([]), Decoder)
assert isinstance(Sequence([]), Sequence)
serialized = pickle.dumps(Sequence([]))
assert isinstance(pickle.loads(serialized), Sequence)
def test_decoding(self):
decoder = Sequence([CTC(), Metaspace()])
initial = ["▁", "▁", "H", "H", "i", "i", "▁", "y", "o", "u"]
expected = "Hi you"
assert decoder.decode(initial) == expected
| tokenizers/bindings/python/tests/bindings/test_decoders.py/0 | {
"file_path": "tokenizers/bindings/python/tests/bindings/test_decoders.py",
"repo_id": "tokenizers",
"token_count": 3521
} | 212 |
import pytest
from tokenizers import CharBPETokenizer
from ..utils import data_dir, multiprocessing_with_parallelism, openai_files
class TestCharBPETokenizer:
def test_basic_encode(self, openai_files):
tokenizer = CharBPETokenizer.from_file(openai_files["vocab"], openai_files["merges"])
output = tokenizer.encode("My name is John", "pair")
assert output.ids == [0, 253, 1362, 544, 0, 7, 12662, 2688]
assert output.tokens == [
"<unk>",
"y</w>",
"name</w>",
"is</w>",
"<unk>",
"o",
"hn</w>",
"pair</w>",
]
assert output.offsets == [
(0, 1),
(1, 2),
(3, 7),
(8, 10),
(11, 12),
(12, 13),
(13, 15),
(0, 4),
]
assert output.type_ids == [0, 0, 0, 0, 0, 0, 0, 1]
def test_lowercase(self, openai_files):
tokenizer = CharBPETokenizer.from_file(openai_files["vocab"], openai_files["merges"], lowercase=True)
output = tokenizer.encode("My name is John", "pair", add_special_tokens=False)
assert output.ids == [547, 1362, 544, 2476, 2688]
assert output.tokens == ["my</w>", "name</w>", "is</w>", "john</w>", "pair</w>"]
assert output.offsets == [(0, 2), (3, 7), (8, 10), (11, 15), (0, 4)]
assert output.type_ids == [0, 0, 0, 0, 1]
def test_decoding(self, openai_files):
tokenizer = CharBPETokenizer.from_file(openai_files["vocab"], openai_files["merges"], lowercase=True)
decoded = tokenizer.decode(tokenizer.encode("my name is john").ids)
assert decoded == "my name is john"
def test_multiprocessing_with_parallelism(self, openai_files):
tokenizer = CharBPETokenizer.from_file(openai_files["vocab"], openai_files["merges"])
multiprocessing_with_parallelism(tokenizer, False)
multiprocessing_with_parallelism(tokenizer, True)
def test_train_from_iterator(self):
text = ["A first sentence", "Another sentence", "And a last one"]
tokenizer = CharBPETokenizer()
tokenizer.train_from_iterator(text, show_progress=False)
output = tokenizer.encode("A sentence")
assert output.tokens == ["A</w>", "sentence</w>"]
| tokenizers/bindings/python/tests/implementations/test_char_bpe.py/0 | {
"file_path": "tokenizers/bindings/python/tests/implementations/test_char_bpe.py",
"repo_id": "tokenizers",
"token_count": 1099
} | 213 |
# Tokenizer
<tokenizerslangcontent>
<python>
## Tokenizer
[[autodoc]] tokenizers.Tokenizer
- all
- decoder
- model
- normalizer
- padding
- post_processor
- pre_tokenizer
- truncation
</python>
<rust>
The Rust API Reference is available directly on the [Docs.rs](https://docs.rs/tokenizers/latest/tokenizers/) website.
</rust>
<node>
The node API has not been documented yet.
</node>
</tokenizerslangcontent> | tokenizers/docs/source-doc-builder/api/tokenizer.mdx/0 | {
"file_path": "tokenizers/docs/source-doc-builder/api/tokenizer.mdx",
"repo_id": "tokenizers",
"token_count": 156
} | 214 |
.highlight .c1, .highlight .sd{
color: #999
}
.highlight .nn, .highlight .k, .highlight .s1, .highlight .nb, .highlight .bp, .highlight .kc, .highlight .kt {
color: #FB8D68;
}
.highlight .kn, .highlight .nv, .highlight .s2, .highlight .ow, .highlight .kd, .highlight .kr, .highlight .s {
color: #6670FF;
}
.highlight .gp {
color: #FB8D68;
}
| tokenizers/docs/source/_static/css/code-snippets.css/0 | {
"file_path": "tokenizers/docs/source/_static/css/code-snippets.css",
"repo_id": "tokenizers",
"token_count": 166
} | 215 |
Quicktour
====================================================================================================
Let's have a quick look at the 🤗 Tokenizers library features. The library provides an
implementation of today's most used tokenizers that is both easy to use and blazing fast.
.. only:: python
It can be used to instantiate a :ref:`pretrained tokenizer <pretrained>` but we will start our
quicktour by building one from scratch and see how we can train it.
Build a tokenizer from scratch
----------------------------------------------------------------------------------------------------
To illustrate how fast the 🤗 Tokenizers library is, let's train a new tokenizer on `wikitext-103
<https://blog.einstein.ai/the-wikitext-long-term-dependency-language-modeling-dataset/>`__ (516M of
text) in just a few seconds. First things first, you will need to download this dataset and unzip it
with:
.. code-block:: bash
wget https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-raw-v1.zip
unzip wikitext-103-raw-v1.zip
Training the tokenizer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. entities:: python
BpeTrainer
:class:`~tokenizers.trainers.BpeTrainer`
vocab_size
:obj:`vocab_size`
min_frequency
:obj:`min_frequency`
special_tokens
:obj:`special_tokens`
unk_token
:obj:`unk_token`
pad_token
:obj:`pad_token`
.. entities:: rust
BpeTrainer
:rust_struct:`~tokenizers::models::bpe::BpeTrainer`
vocab_size
:obj:`vocab_size`
min_frequency
:obj:`min_frequency`
special_tokens
:obj:`special_tokens`
unk_token
:obj:`unk_token`
pad_token
:obj:`pad_token`
.. entities:: node
BpeTrainer
BpeTrainer
vocab_size
:obj:`vocabSize`
min_frequency
:obj:`minFrequency`
special_tokens
:obj:`specialTokens`
unk_token
:obj:`unkToken`
pad_token
:obj:`padToken`
In this tour, we will build and train a Byte-Pair Encoding (BPE) tokenizer. For more information
about the different type of tokenizers, check out this `guide
<https://huggingface.co/docs/transformers/main/en/tokenizer_summary#summary-of-the-tokenizers>`__ in the 🤗 Transformers
documentation. Here, training the tokenizer means it will learn merge rules by:
- Start with all the characters present in the training corpus as tokens.
- Identify the most common pair of tokens and merge it into one token.
- Repeat until the vocabulary (e.g., the number of tokens) has reached the size we want.
The main API of the library is the :entity:`class` :entity:`Tokenizer`, here is how we instantiate
one with a BPE model:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START init_tokenizer
:end-before: END init_tokenizer
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_init_tokenizer
:end-before: END quicktour_init_tokenizer
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START init_tokenizer
:end-before: END init_tokenizer
:dedent: 4
To train our tokenizer on the wikitext files, we will need to instantiate a `trainer`, in this case
a :entity:`BpeTrainer`
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START init_trainer
:end-before: END init_trainer
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_init_trainer
:end-before: END quicktour_init_trainer
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START init_trainer
:end-before: END init_trainer
:dedent: 4
We can set the training arguments like :entity:`vocab_size` or :entity:`min_frequency` (here left at
their default values of 30,000 and 0) but the most important part is to give the
:entity:`special_tokens` we plan to use later on (they are not used at all during training) so that
they get inserted in the vocabulary.
.. note::
The order in which you write the special tokens list matters: here :obj:`"[UNK]"` will get the
ID 0, :obj:`"[CLS]"` will get the ID 1 and so forth.
We could train our tokenizer right now, but it wouldn't be optimal. Without a pre-tokenizer that
will split our inputs into words, we might get tokens that overlap several words: for instance we
could get an :obj:`"it is"` token since those two words often appear next to each other. Using a
pre-tokenizer will ensure no token is bigger than a word returned by the pre-tokenizer. Here we want
to train a subword BPE tokenizer, and we will use the easiest pre-tokenizer possible by splitting
on whitespace.
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START init_pretok
:end-before: END init_pretok
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_init_pretok
:end-before: END quicktour_init_pretok
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START init_pretok
:end-before: END init_pretok
:dedent: 4
Now, we can just call the :entity:`Tokenizer.train` method with any list of files we want
to use:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START train
:end-before: END train
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_train
:end-before: END quicktour_train
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START train
:end-before: END train
:dedent: 4
This should only take a few seconds to train our tokenizer on the full wikitext dataset!
To save the tokenizer in one file that contains all its configuration and vocabulary, just use the
:entity:`Tokenizer.save` method:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START save
:end-before: END save
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_save
:end-before: END quicktour_save
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START save
:end-before: END save
:dedent: 4
and you can reload your tokenizer from that file with the :entity:`Tokenizer.from_file`
:entity:`classmethod`:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START reload_tokenizer
:end-before: END reload_tokenizer
:dedent: 12
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_reload_tokenizer
:end-before: END quicktour_reload_tokenizer
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START reload_tokenizer
:end-before: END reload_tokenizer
:dedent: 4
Using the tokenizer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Now that we have trained a tokenizer, we can use it on any text we want with the
:entity:`Tokenizer.encode` method:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START encode
:end-before: END encode
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_encode
:end-before: END quicktour_encode
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START encode
:end-before: END encode
:dedent: 4
This applied the full pipeline of the tokenizer on the text, returning an
:entity:`Encoding` object. To learn more about this pipeline, and how to apply (or
customize) parts of it, check out :doc:`this page <pipeline>`.
This :entity:`Encoding` object then has all the attributes you need for your deep
learning model (or other). The :obj:`tokens` attribute contains the segmentation of your text in
tokens:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_tokens
:end-before: END print_tokens
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_tokens
:end-before: END quicktour_print_tokens
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_tokens
:end-before: END print_tokens
:dedent: 4
Similarly, the :obj:`ids` attribute will contain the index of each of those tokens in the
tokenizer's vocabulary:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_ids
:end-before: END print_ids
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_ids
:end-before: END quicktour_print_ids
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_ids
:end-before: END print_ids
:dedent: 4
An important feature of the 🤗 Tokenizers library is that it comes with full alignment tracking,
meaning you can always get the part of your original sentence that corresponds to a given token.
Those are stored in the :obj:`offsets` attribute of our :entity:`Encoding` object. For
instance, let's assume we would want to find back what caused the :obj:`"[UNK]"` token to appear,
which is the token at index 9 in the list, we can just ask for the offset at the index:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_offsets
:end-before: END print_offsets
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_offsets
:end-before: END quicktour_print_offsets
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_offsets
:end-before: END print_offsets
:dedent: 4
and those are the indices that correspond to the emoji in the original sentence:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START use_offsets
:end-before: END use_offsets
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_use_offsets
:end-before: END quicktour_use_offsets
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START use_offsets
:end-before: END use_offsets
:dedent: 4
Post-processing
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
We might want our tokenizer to automatically add special tokens, like :obj:`"[CLS]"` or
:obj:`"[SEP]"`. To do this, we use a post-processor. :entity:`TemplateProcessing` is the
most commonly used, you just have to specify a template for the processing of single sentences and
pairs of sentences, along with the special tokens and their IDs.
When we built our tokenizer, we set :obj:`"[CLS]"` and :obj:`"[SEP]"` in positions 1 and 2 of our
list of special tokens, so this should be their IDs. To double-check, we can use the
:entity:`Tokenizer.token_to_id` method:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START check_sep
:end-before: END check_sep
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_check_sep
:end-before: END quicktour_check_sep
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START check_sep
:end-before: END check_sep
:dedent: 4
Here is how we can set the post-processing to give us the traditional BERT inputs:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START init_template_processing
:end-before: END init_template_processing
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_init_template_processing
:end-before: END quicktour_init_template_processing
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START init_template_processing
:end-before: END init_template_processing
:dedent: 4
Let's go over this snippet of code in more details. First we specify the template for single
sentences: those should have the form :obj:`"[CLS] $A [SEP]"` where :obj:`$A` represents our
sentence.
Then, we specify the template for sentence pairs, which should have the form
:obj:`"[CLS] $A [SEP] $B [SEP]"` where :obj:`$A` represents the first sentence and :obj:`$B` the
second one. The :obj:`:1` added in the template represent the `type IDs` we want for each part of
our input: it defaults to 0 for everything (which is why we don't have :obj:`$A:0`) and here we set
it to 1 for the tokens of the second sentence and the last :obj:`"[SEP]"` token.
Lastly, we specify the special tokens we used and their IDs in our tokenizer's vocabulary.
To check out this worked properly, let's try to encode the same sentence as before:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_special_tokens
:end-before: END print_special_tokens
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_special_tokens
:end-before: END quicktour_print_special_tokens
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_special_tokens
:end-before: END print_special_tokens
:dedent: 4
To check the results on a pair of sentences, we just pass the two sentences to
:entity:`Tokenizer.encode`:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_special_tokens_pair
:end-before: END print_special_tokens_pair
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_special_tokens_pair
:end-before: END quicktour_print_special_tokens_pair
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_special_tokens_pair
:end-before: END print_special_tokens_pair
:dedent: 4
You can then check the type IDs attributed to each token is correct with
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_type_ids
:end-before: END print_type_ids
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_type_ids
:end-before: END quicktour_print_type_ids
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_type_ids
:end-before: END print_type_ids
:dedent: 4
If you save your tokenizer with :entity:`Tokenizer.save`, the post-processor will be saved along.
Encoding multiple sentences in a batch
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
To get the full speed of the 🤗 Tokenizers library, it's best to process your texts by batches by
using the :entity:`Tokenizer.encode_batch` method:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START encode_batch
:end-before: END encode_batch
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_encode_batch
:end-before: END quicktour_encode_batch
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START encode_batch
:end-before: END encode_batch
:dedent: 4
The output is then a list of :entity:`Encoding` objects like the ones we saw before. You
can process together as many texts as you like, as long as it fits in memory.
To process a batch of sentences pairs, pass two lists to the
:entity:`Tokenizer.encode_batch` method: the list of sentences A and the list of sentences
B:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START encode_batch_pair
:end-before: END encode_batch_pair
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_encode_batch_pair
:end-before: END quicktour_encode_batch_pair
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START encode_batch_pair
:end-before: END encode_batch_pair
:dedent: 4
When encoding multiple sentences, you can automatically pad the outputs to the longest sentence
present by using :entity:`Tokenizer.enable_padding`, with the :entity:`pad_token` and its ID
(which we can double-check the id for the padding token with
:entity:`Tokenizer.token_to_id` like before):
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START enable_padding
:end-before: END enable_padding
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_enable_padding
:end-before: END quicktour_enable_padding
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START enable_padding
:end-before: END enable_padding
:dedent: 4
We can set the :obj:`direction` of the padding (defaults to the right) or a given :obj:`length` if
we want to pad every sample to that specific number (here we leave it unset to pad to the size of
the longest text).
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_batch_tokens
:end-before: END print_batch_tokens
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_batch_tokens
:end-before: END quicktour_print_batch_tokens
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_batch_tokens
:end-before: END print_batch_tokens
:dedent: 4
In this case, the `attention mask` generated by the tokenizer takes the padding into account:
.. only:: python
.. literalinclude:: ../../bindings/python/tests/documentation/test_quicktour.py
:language: python
:start-after: START print_attention_mask
:end-before: END print_attention_mask
:dedent: 8
.. only:: rust
.. literalinclude:: ../../tokenizers/tests/documentation.rs
:language: rust
:start-after: START quicktour_print_attention_mask
:end-before: END quicktour_print_attention_mask
:dedent: 4
.. only:: node
.. literalinclude:: ../../bindings/node/examples/documentation/quicktour.test.ts
:language: javascript
:start-after: START print_attention_mask
:end-before: END print_attention_mask
:dedent: 4
.. _pretrained:
.. only:: python
Using a pretrained tokenizer
------------------------------------------------------------------------------------------------
You can load any tokenizer from the Hugging Face Hub as long as a `tokenizer.json` file is
available in the repository.
.. code-block:: python
from tokenizers import Tokenizer
tokenizer = Tokenizer.from_pretrained("bert-base-uncased")
Importing a pretrained tokenizer from legacy vocabulary files
------------------------------------------------------------------------------------------------
You can also import a pretrained tokenizer directly in, as long as you have its vocabulary file.
For instance, here is how to import the classic pretrained BERT tokenizer:
.. code-block:: python
from tokenizers import BertWordPieceTokenizer
tokenizer = BertWordPieceTokenizer("bert-base-uncased-vocab.txt", lowercase=True)
as long as you have downloaded the file `bert-base-uncased-vocab.txt` with
.. code-block:: bash
wget https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt
| tokenizers/docs/source/quicktour.rst/0 | {
"file_path": "tokenizers/docs/source/quicktour.rst",
"repo_id": "tokenizers",
"token_count": 8904
} | 216 |
<div align="center">
<h1><code>wasm-pack-template</code></h1>
<strong>A template for kick starting a Rust and WebAssembly project using <a href="https://github.com/rustwasm/wasm-pack">wasm-pack</a>.</strong>
<p>
<a href="https://travis-ci.org/rustwasm/wasm-pack-template"><img src="https://img.shields.io/travis/rustwasm/wasm-pack-template.svg?style=flat-square" alt="Build Status" /></a>
</p>
<h3>
<a href="https://rustwasm.github.io/docs/wasm-pack/tutorials/npm-browser-packages/index.html">Tutorial</a>
<span> | </span>
<a href="https://discordapp.com/channels/442252698964721669/443151097398296587">Chat</a>
</h3>
<sub>Built with 🦀🕸 by <a href="https://rustwasm.github.io/">The Rust and WebAssembly Working Group</a></sub>
</div>
## About
This is an example project showing off a very basic use case for `wasm` tokenizers
usage.
[**📚 Read this template tutorial! 📚**][template-docs]
This template is designed for compiling Rust libraries into WebAssembly and
publishing the resulting package to NPM.
Be sure to check out [other `wasm-pack` tutorials online][tutorials] for other
templates and usages of `wasm-pack`.
[tutorials]: https://rustwasm.github.io/docs/wasm-pack/tutorials/index.html
[template-docs]: https://rustwasm.github.io/docs/wasm-pack/tutorials/npm-browser-packages/index.html
## 🚴 Usage
### 🐑 Use `cargo generate` to Clone this Template
[Learn more about `cargo generate` here.](https://github.com/ashleygwilliams/cargo-generate)
```
cargo generate --git https://github.com/rustwasm/wasm-pack-template.git --name my-project
cd my-project
```
### 🛠️ Build with `wasm-pack build`
```
wasm-pack build
```
### 🔬 Test in Headless Browsers with `wasm-pack test`
```
wasm-pack test --headless --firefox
```
### 🎁 Publish to NPM with `wasm-pack publish`
```
wasm-pack publish
```
## 🔋 Batteries Included
* [`wasm-bindgen`](https://github.com/rustwasm/wasm-bindgen) for communicating
between WebAssembly and JavaScript.
* [`console_error_panic_hook`](https://github.com/rustwasm/console_error_panic_hook)
for logging panic messages to the developer console.
* [`wee_alloc`](https://github.com/rustwasm/wee_alloc), an allocator optimized
for small code size.
| tokenizers/tokenizers/examples/unstable_wasm/README.md/0 | {
"file_path": "tokenizers/tokenizers/examples/unstable_wasm/README.md",
"repo_id": "tokenizers",
"token_count": 811
} | 217 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.