Spaces:
Runtime error
Runtime error
File size: 7,457 Bytes
8c1c4dc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 |
# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
# Copyright 2022 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 inspect
from typing import List, Optional, Tuple, Union
import numpy as np
import paddle
import PIL
from ...models import UNet2DModel, VQModel
from ...pipeline_utils import DiffusionPipeline, ImagePipelineOutput
from ...schedulers import (
DDIMScheduler,
DPMSolverMultistepScheduler,
EulerAncestralDiscreteScheduler,
EulerDiscreteScheduler,
LMSDiscreteScheduler,
PNDMScheduler,
)
from ...utils import PIL_INTERPOLATION
def preprocess(image):
w, h = image.size
w, h = map(lambda x: x - x % 32, (w, h)) # resize to integer multiple of 32
image = image.resize((w, h), resample=PIL_INTERPOLATION["lanczos"])
image = np.array(image).astype(np.float32) / 255.0
image = image[None].transpose(0, 3, 1, 2)
image = paddle.to_tensor(image)
return 2.0 * image - 1.0
class LDMSuperResolutionPipeline(DiffusionPipeline):
r"""
A pipeline for image super-resolution using Latent
This class 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 xxxx, etc.)
Parameters:
vqvae ([`VQModel`]):
Vector-quantized (VQ) VAE Model to encode and decode images to and from latent representations.
unet ([`UNet2DModel`]): U-Net architecture to denoise the encoded image.
scheduler ([`SchedulerMixin`]):
A scheduler to be used in combination with `unet` to denoise the encoded image latens. Can be one of
[`DDIMScheduler`], [`LMSDiscreteScheduler`], [`EulerDiscreteScheduler`],
[`EulerAncestralDiscreteScheduler`], [`DPMSolverMultistepScheduler`], or [`PNDMScheduler`].
"""
def __init__(
self,
vqvae: VQModel,
unet: UNet2DModel,
scheduler: Union[
DDIMScheduler,
PNDMScheduler,
LMSDiscreteScheduler,
EulerDiscreteScheduler,
EulerAncestralDiscreteScheduler,
DPMSolverMultistepScheduler,
],
):
super().__init__()
self.register_modules(vqvae=vqvae, unet=unet, scheduler=scheduler)
@paddle.no_grad()
def __call__(
self,
image: Union[paddle.Tensor, PIL.Image.Image],
batch_size: Optional[int] = 1,
num_inference_steps: Optional[int] = 100,
eta: Optional[float] = 0.0,
generator: Optional[Union[paddle.Generator, List[paddle.Generator]]] = None,
output_type: Optional[str] = "pil",
return_dict: bool = True,
**kwargs,
) -> Union[Tuple, ImagePipelineOutput]:
r"""
Args:
image (`paddle.Tensor` or `PIL.Image.Image`):
`Image`, or tensor representing an image batch, that will be used as the starting point for the
process.
batch_size (`int`, *optional*, defaults to 1):
Number of images to generate.
num_inference_steps (`int`, *optional*, defaults to 100):
The number of denoising steps. More denoising steps usually lead to a higher quality image at the
expense of slower inference.
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 (`paddle.Generator`, *optional*):
One or a list of paddle generator(s) to make generation deterministic.
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*):
Whether or not to return a [`~pipeline_utils.ImagePipelineOutput`] instead of a plain tuple.
Returns:
[`~pipeline_utils.ImagePipelineOutput`] or `tuple`: [`~pipelines.utils.ImagePipelineOutput`] if
`return_dict` is True, otherwise a `tuple. When returning a tuple, the first element is a list with the
generated images.
"""
if isinstance(image, PIL.Image.Image):
batch_size = 1
elif isinstance(image, paddle.Tensor):
batch_size = image.shape[0]
else:
raise ValueError(f"`image` has to be of type `PIL.Image.Image` or `paddle.Tensor` but is {type(image)}")
if isinstance(image, PIL.Image.Image):
image = preprocess(image)
height, width = image.shape[-2:]
# in_channels should be 6: 3 for latents, 3 for low resolution image
latents_shape = (batch_size, self.unet.in_channels // 2, height, width)
latents_dtype = next(self.unet.named_parameters())[1].dtype
latents = paddle.randn(latents_shape, generator=generator, dtype=latents_dtype)
image = image.cast(latents_dtype)
self.scheduler.set_timesteps(num_inference_steps)
timesteps_tensor = self.scheduler.timesteps
# scale the initial noise by the standard deviation required by the scheduler
latents = latents * self.scheduler.init_noise_sigma
# 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_kwargs = {}
if accepts_eta:
extra_kwargs["eta"] = eta
for t in self.progress_bar(timesteps_tensor):
# concat latents and low resolution image in the channel dimension.
latents_input = paddle.concat([latents, image], axis=1)
latents_input = self.scheduler.scale_model_input(latents_input, t)
# predict the noise residual
noise_pred = self.unet(latents_input, t).sample
# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(noise_pred, t, latents, **extra_kwargs).prev_sample
# decode the image latents with the VQVAE
image = self.vqvae.decode(latents).sample
image = paddle.clip(image, -1.0, 1.0)
image = image / 2 + 0.5
image = image.transpose([0, 2, 3, 1]).cast("float32").numpy()
if output_type == "pil":
image = self.numpy_to_pil(image)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=image)
|