File size: 14,108 Bytes
43b7e92 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 |
# Copyright 2024 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 Callable, List, Optional, Union
import torch
from ...models import UNet2DConditionModel, VQModel
from ...schedulers import DDPMScheduler
from ...utils import (
logging,
)
from ...utils.torch_utils import randn_tensor
from ..pipeline_utils import DiffusionPipeline, ImagePipelineOutput
logger = logging.get_logger(__name__) # pylint: disable=invalid-name
EXAMPLE_DOC_STRING = """
Examples:
```py
>>> import torch
>>> import numpy as np
>>> from diffusers import KandinskyV22PriorPipeline, KandinskyV22ControlnetPipeline
>>> from transformers import pipeline
>>> from diffusers.utils import load_image
>>> def make_hint(image, depth_estimator):
... image = depth_estimator(image)["depth"]
... image = np.array(image)
... image = image[:, :, None]
... image = np.concatenate([image, image, image], axis=2)
... detected_map = torch.from_numpy(image).float() / 255.0
... hint = detected_map.permute(2, 0, 1)
... return hint
>>> depth_estimator = pipeline("depth-estimation")
>>> pipe_prior = KandinskyV22PriorPipeline.from_pretrained(
... "kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16
... )
>>> pipe_prior = pipe_prior.to("cuda")
>>> pipe = KandinskyV22ControlnetPipeline.from_pretrained(
... "kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
... )
>>> pipe = pipe.to("cuda")
>>> img = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... ).resize((768, 768))
>>> hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
>>> prompt = "A robot, 4k photo"
>>> negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
>>> generator = torch.Generator(device="cuda").manual_seed(43)
>>> image_emb, zero_image_emb = pipe_prior(
... prompt=prompt, negative_prompt=negative_prior_prompt, generator=generator
... ).to_tuple()
>>> images = pipe(
... image_embeds=image_emb,
... negative_image_embeds=zero_image_emb,
... hint=hint,
... num_inference_steps=50,
... generator=generator,
... height=768,
... width=768,
... ).images
>>> images[0].save("robot_cat.png")
```
"""
# Copied from diffusers.pipelines.kandinsky2_2.pipeline_kandinsky2_2.downscale_height_and_width
def downscale_height_and_width(height, width, scale_factor=8):
new_height = height // scale_factor**2
if height % scale_factor**2 != 0:
new_height += 1
new_width = width // scale_factor**2
if width % scale_factor**2 != 0:
new_width += 1
return new_height * scale_factor, new_width * scale_factor
class KandinskyV22ControlnetPipeline(DiffusionPipeline):
"""
Pipeline for text-to-image generation using Kandinsky
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.)
Args:
scheduler ([`DDIMScheduler`]):
A scheduler to be used in combination with `unet` to generate image latents.
unet ([`UNet2DConditionModel`]):
Conditional U-Net architecture to denoise the image embedding.
movq ([`VQModel`]):
MoVQ Decoder to generate the image from the latents.
"""
model_cpu_offload_seq = "unet->movq"
def __init__(
self,
unet: UNet2DConditionModel,
scheduler: DDPMScheduler,
movq: VQModel,
):
super().__init__()
self.register_modules(
unet=unet,
scheduler=scheduler,
movq=movq,
)
self.movq_scale_factor = 2 ** (len(self.movq.config.block_out_channels) - 1)
# 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
@torch.no_grad()
def __call__(
self,
image_embeds: Union[torch.Tensor, List[torch.Tensor]],
negative_image_embeds: Union[torch.Tensor, List[torch.Tensor]],
hint: torch.Tensor,
height: int = 512,
width: int = 512,
num_inference_steps: int = 100,
guidance_scale: float = 4.0,
num_images_per_prompt: int = 1,
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
latents: Optional[torch.Tensor] = None,
output_type: Optional[str] = "pil",
callback: Optional[Callable[[int, int, torch.Tensor], None]] = None,
callback_steps: int = 1,
return_dict: bool = True,
):
"""
Function invoked when calling the pipeline for generation.
Args:
prompt (`str` or `List[str]`):
The prompt or prompts to guide the image generation.
hint (`torch.Tensor`):
The controlnet condition.
image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for text prompt, that will be used to condition the image generation.
negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`):
The clip image embeddings for negative text prompt, will be used to condition the image generation.
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 `guidance_scale` is less than `1`).
height (`int`, *optional*, defaults to 512):
The height in pixels of the generated image.
width (`int`, *optional*, defaults to 512):
The width in pixels of the generated image.
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.
guidance_scale (`float`, *optional*, defaults to 4.0):
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.
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.Tensor`, *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`).
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.Tensor)`.
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.
return_dict (`bool`, *optional*, defaults to `True`):
Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple.
Examples:
Returns:
[`~pipelines.ImagePipelineOutput`] or `tuple`
"""
device = self._execution_device
do_classifier_free_guidance = guidance_scale > 1.0
if isinstance(image_embeds, list):
image_embeds = torch.cat(image_embeds, dim=0)
if isinstance(negative_image_embeds, list):
negative_image_embeds = torch.cat(negative_image_embeds, dim=0)
if isinstance(hint, list):
hint = torch.cat(hint, dim=0)
batch_size = image_embeds.shape[0] * num_images_per_prompt
if do_classifier_free_guidance:
image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
negative_image_embeds = negative_image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
hint = hint.repeat_interleave(num_images_per_prompt, dim=0)
image_embeds = torch.cat([negative_image_embeds, image_embeds], dim=0).to(
dtype=self.unet.dtype, device=device
)
hint = torch.cat([hint, hint], dim=0).to(dtype=self.unet.dtype, device=device)
self.scheduler.set_timesteps(num_inference_steps, device=device)
timesteps_tensor = self.scheduler.timesteps
num_channels_latents = self.movq.config.latent_channels
height, width = downscale_height_and_width(height, width, self.movq_scale_factor)
# create initial latent
latents = self.prepare_latents(
(batch_size, num_channels_latents, height, width),
image_embeds.dtype,
device,
generator,
latents,
self.scheduler,
)
for i, t in enumerate(self.progress_bar(timesteps_tensor)):
# expand the latents if we are doing classifier free guidance
latent_model_input = torch.cat([latents] * 2) if do_classifier_free_guidance else latents
added_cond_kwargs = {"image_embeds": image_embeds, "hint": hint}
noise_pred = self.unet(
sample=latent_model_input,
timestep=t,
encoder_hidden_states=None,
added_cond_kwargs=added_cond_kwargs,
return_dict=False,
)[0]
if do_classifier_free_guidance:
noise_pred, variance_pred = noise_pred.split(latents.shape[1], dim=1)
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
_, variance_pred_text = variance_pred.chunk(2)
noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
noise_pred = torch.cat([noise_pred, variance_pred_text], dim=1)
if not (
hasattr(self.scheduler.config, "variance_type")
and self.scheduler.config.variance_type in ["learned", "learned_range"]
):
noise_pred, _ = noise_pred.split(latents.shape[1], dim=1)
# compute the previous noisy sample x_t -> x_t-1
latents = self.scheduler.step(
noise_pred,
t,
latents,
generator=generator,
)[0]
if callback is not None and i % callback_steps == 0:
step_idx = i // getattr(self.scheduler, "order", 1)
callback(step_idx, t, latents)
# post-processing
image = self.movq.decode(latents, force_not_quantize=True)["sample"]
# Offload all models
self.maybe_free_model_hooks()
if output_type not in ["pt", "np", "pil"]:
raise ValueError(f"Only the output types `pt`, `pil` and `np` are supported not output_type={output_type}")
if output_type in ["np", "pil"]:
image = image * 0.5 + 0.5
image = image.clamp(0, 1)
image = image.cpu().permute(0, 2, 3, 1).float().numpy()
if output_type == "pil":
image = self.numpy_to_pil(image)
if not return_dict:
return (image,)
return ImagePipelineOutput(images=image)
|