|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
import inspect |
|
from typing import Any, Callable, Dict, List, Optional, Union |
|
|
|
import PIL |
|
import numpy as np |
|
import torch |
|
import torchvision.transforms as T |
|
from diffusers.configuration_utils import FrozenDict |
|
from diffusers.image_processor import VaeImageProcessor |
|
from diffusers.loaders import FromSingleFileMixin, LoraLoaderMixin, TextualInversionLoaderMixin |
|
from diffusers.models import AutoencoderKL, UNet2DConditionModel |
|
from diffusers.models.lora import adjust_lora_scale_text_encoder |
|
from diffusers.pipelines.pipeline_utils import DiffusionPipeline |
|
from diffusers.pipelines.stable_diffusion import StableDiffusionPipelineOutput |
|
from diffusers.pipelines.stable_diffusion.safety_checker import StableDiffusionSafetyChecker |
|
from diffusers.schedulers import KarrasDiffusionSchedulers |
|
from diffusers.utils import ( |
|
PIL_INTERPOLATION, |
|
deprecate, |
|
logging, |
|
replace_example_docstring, |
|
) |
|
from diffusers.utils.torch_utils import randn_tensor |
|
from transformers import BertModel, BertTokenizer |
|
from transformers import CLIPImageProcessor, CLIPTextModel, CLIPTokenizer |
|
|
|
from ..modules.models import HunYuanDiT |
|
|
|
logger = logging.get_logger(__name__) |
|
|
|
EXAMPLE_DOC_STRING = """ |
|
Examples: |
|
```py |
|
>>> import requests |
|
>>> import torch |
|
>>> from PIL import Image |
|
>>> from io import BytesIO |
|
|
|
>>> from diffusers import StableDiffusionImg2ImgPipeline |
|
|
|
>>> device = "cuda" |
|
>>> model_id_or_path = "runwayml/stable-diffusion-v1-5" |
|
>>> pipe = StableDiffusionImg2ImgPipeline.from_pretrained(model_id_or_path, torch_dtype=torch.float16) |
|
>>> pipe = pipe.to(device) |
|
|
|
>>> url = "https://raw.githubusercontent.com/CompVis/stable-diffusion/main/assets/stable-samples/img2img/sketch-mountains-input.jpg" |
|
|
|
>>> response = requests.get(url) |
|
>>> init_image = Image.open(BytesIO(response.content)).convert("RGB") |
|
>>> init_image = init_image.resize((768, 512)) |
|
|
|
>>> prompt = "A fantasy landscape, trending on artstation" |
|
|
|
>>> images = pipe(prompt=prompt, image=init_image, strength=0.75, guidance_scale=7.5).images |
|
>>> images[0].save("fantasy_landscape.png") |
|
``` |
|
""" |
|
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) |
|
|
|
noise_pred_rescaled = noise_cfg * (std_text / std_cfg) |
|
|
|
noise_cfg = guidance_rescale * noise_pred_rescaled + (1 - guidance_rescale) * noise_cfg |
|
return noise_cfg |
|
|
|
def preprocess(image): |
|
deprecation_message = "The preprocess method is deprecated and will be removed in diffusers 1.0.0. Please use VaeImageProcessor.preprocess(...) instead" |
|
deprecate("preprocess", "1.0.0", deprecation_message, standard_warn=False) |
|
if isinstance(image, torch.Tensor): |
|
return image |
|
elif isinstance(image, PIL.Image.Image): |
|
image = [image] |
|
|
|
if isinstance(image[0], PIL.Image.Image): |
|
w, h = image[0].size |
|
w, h = (x - x % 8 for x in (w, h)) |
|
|
|
image = [np.array(i.resize((w, h), resample=PIL_INTERPOLATION["lanczos"]))[None, :] for i in image] |
|
image = np.concatenate(image, axis=0) |
|
image = np.array(image).astype(np.float32) / 255.0 |
|
image = image.transpose(0, 3, 1, 2) |
|
image = 2.0 * image - 1.0 |
|
image = torch.from_numpy(image) |
|
elif isinstance(image[0], torch.Tensor): |
|
image = torch.cat(image, dim=0) |
|
return image |
|
|
|
|
|
class StableDiffusionPipeline( |
|
DiffusionPipeline, TextualInversionLoaderMixin, LoraLoaderMixin, FromSingleFileMixin |
|
): |
|
r""" |
|
Pipeline for text-guided image-to-image generation using Stable Diffusion. |
|
|
|
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.). |
|
|
|
The pipeline also inherits the following loading methods: |
|
- [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings |
|
- [`~loaders.LoraLoaderMixin.load_lora_weights`] for loading LoRA weights |
|
- [`~loaders.LoraLoaderMixin.save_lora_weights`] for saving LoRA weights |
|
- [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files |
|
|
|
Args: |
|
vae ([`AutoencoderKL`]): |
|
Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations. |
|
text_encoder (Optional[`~transformers.BertModel`, `~transformers.CLIPTextModel`]): |
|
Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)). |
|
tokenizer (Optional[`~transformers.BertTokenizer`, `~transformers.CLIPTokenizer`]): |
|
A `BertTokenizer` or `CLIPTokenizer` to tokenize text. |
|
unet (Optional[`HunYuanDiT`, `UNet2DConditionModel`]): |
|
A `UNet2DConditionModel` 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`]. |
|
safety_checker ([`StableDiffusionSafetyChecker`]): |
|
Classification module that estimates whether generated images could be considered offensive or harmful. |
|
Please refer to the [model card](https://huggingface.co/runwayml/stable-diffusion-v1-5) for more details |
|
about a model's potential harms. |
|
feature_extractor ([`~transformers.CLIPImageProcessor`]): |
|
A `CLIPImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`. |
|
""" |
|
model_cpu_offload_seq = "text_encoder->unet->vae" |
|
_optional_components = ["safety_checker", "feature_extractor"] |
|
_exclude_from_cpu_offload = ["safety_checker"] |
|
|
|
def __init__( |
|
self, |
|
vae: AutoencoderKL, |
|
text_encoder: Union[BertModel, CLIPTextModel], |
|
tokenizer: Union[BertTokenizer, CLIPTokenizer], |
|
unet: Union[HunYuanDiT, UNet2DConditionModel], |
|
scheduler: KarrasDiffusionSchedulers, |
|
safety_checker: StableDiffusionSafetyChecker, |
|
feature_extractor: CLIPImageProcessor, |
|
requires_safety_checker: bool = True, |
|
progress_bar_config: Dict[str, Any] = None, |
|
embedder_t5=None, |
|
infer_mode='torch', |
|
): |
|
super().__init__() |
|
|
|
|
|
self.embedder_t5 = embedder_t5 |
|
self.infer_mode = infer_mode |
|
|
|
|
|
if progress_bar_config is None: |
|
progress_bar_config = {} |
|
if not hasattr(self, '_progress_bar_config'): |
|
self._progress_bar_config = {} |
|
self._progress_bar_config.update(progress_bar_config) |
|
|
|
|
|
if hasattr(scheduler.config, "steps_offset") and scheduler.config.steps_offset != 1: |
|
deprecation_message = ( |
|
f"The configuration file of this scheduler: {scheduler} is outdated. `steps_offset`" |
|
f" should be set to 1 instead of {scheduler.config.steps_offset}. Please make sure " |
|
"to update the config accordingly as leaving `steps_offset` might led to incorrect results" |
|
" in future versions. If you have downloaded this checkpoint from the Hugging Face Hub," |
|
" it would be very nice if you could open a Pull request for the `scheduler/scheduler_config.json`" |
|
" file" |
|
) |
|
deprecate("steps_offset!=1", "1.0.0", deprecation_message, standard_warn=False) |
|
new_config = dict(scheduler.config) |
|
new_config["steps_offset"] = 1 |
|
scheduler._internal_dict = FrozenDict(new_config) |
|
|
|
if hasattr(scheduler.config, "clip_sample") and scheduler.config.clip_sample is True: |
|
deprecation_message = ( |
|
f"The configuration file of this scheduler: {scheduler} has not set the configuration `clip_sample`." |
|
" `clip_sample` should be set to False in the configuration file. Please make sure to update the" |
|
" config accordingly as not setting `clip_sample` in the config might lead to incorrect results in" |
|
" future versions. If you have downloaded this checkpoint from the Hugging Face Hub, it would be very" |
|
" nice if you could open a Pull request for the `scheduler/scheduler_config.json` file" |
|
) |
|
deprecate("clip_sample not set", "1.0.0", deprecation_message, standard_warn=False) |
|
new_config = dict(scheduler.config) |
|
new_config["clip_sample"] = False |
|
scheduler._internal_dict = FrozenDict(new_config) |
|
|
|
if safety_checker is None and requires_safety_checker: |
|
logger.warning( |
|
f"You have disabled the safety checker for {self.__class__} by passing `safety_checker=None`. Ensure" |
|
" that you abide to the conditions of the Stable Diffusion license and do not expose unfiltered" |
|
" results in services or applications open to the public. Both the diffusers team and Hugging Face" |
|
" strongly recommend to keep the safety filter enabled in all public facing circumstances, disabling" |
|
" it only for use-cases that involve analyzing network behavior or auditing its results. For more" |
|
" information, please have a look at https://github.com/huggingface/diffusers/pull/254 ." |
|
) |
|
|
|
if safety_checker is not None and feature_extractor is None: |
|
raise ValueError( |
|
"Make sure to define a feature extractor when loading {self.__class__} if you want to use the safety" |
|
" checker. If you do not want to use the safety checker, you can pass `'safety_checker=None'` instead." |
|
) |
|
|
|
self.register_modules( |
|
vae=vae, |
|
text_encoder=text_encoder, |
|
tokenizer=tokenizer, |
|
unet=unet, |
|
scheduler=scheduler, |
|
safety_checker=safety_checker, |
|
feature_extractor=feature_extractor, |
|
) |
|
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.register_to_config(requires_safety_checker=requires_safety_checker) |
|
|
|
|
|
def _encode_prompt( |
|
self, |
|
prompt, |
|
device, |
|
num_images_per_prompt, |
|
do_classifier_free_guidance, |
|
negative_prompt=None, |
|
prompt_embeds: Optional[torch.FloatTensor] = None, |
|
negative_prompt_embeds: Optional[torch.FloatTensor] = None, |
|
lora_scale: Optional[float] = None, |
|
): |
|
deprecation_message = "`_encode_prompt()` is deprecated and it will be removed in a future version. Use `encode_prompt()` instead. Also, be aware that the output format changed from a concatenated tensor to a tuple." |
|
deprecate("_encode_prompt()", "1.0.0", deprecation_message, standard_warn=False) |
|
|
|
prompt_embeds_tuple = self.encode_prompt( |
|
prompt=prompt, |
|
device=device, |
|
num_images_per_prompt=num_images_per_prompt, |
|
do_classifier_free_guidance=do_classifier_free_guidance, |
|
negative_prompt=negative_prompt, |
|
prompt_embeds=prompt_embeds, |
|
negative_prompt_embeds=negative_prompt_embeds, |
|
lora_scale=lora_scale, |
|
) |
|
|
|
|
|
prompt_embeds = torch.cat([prompt_embeds_tuple[1], prompt_embeds_tuple[0]]) |
|
|
|
return prompt_embeds |
|
|
|
|
|
def encode_prompt( |
|
self, |
|
prompt, |
|
device, |
|
num_images_per_prompt, |
|
do_classifier_free_guidance, |
|
negative_prompt=None, |
|
prompt_embeds: Optional[torch.FloatTensor] = None, |
|
negative_prompt_embeds: Optional[torch.FloatTensor] = None, |
|
lora_scale: Optional[float] = None, |
|
embedder=None, |
|
): |
|
r""" |
|
Encodes the prompt into text encoder hidden states. |
|
|
|
Args: |
|
prompt (`str` or `List[str]`, *optional*): |
|
prompt to be encoded |
|
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`). |
|
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. |
|
lora_scale (`float`, *optional*): |
|
A lora scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded. |
|
embedder: |
|
T5 embedder (including text encoder and tokenizer) |
|
""" |
|
if embedder is None: |
|
text_encoder = self.text_encoder |
|
tokenizer = self.tokenizer |
|
max_length = self.tokenizer.model_max_length |
|
else: |
|
text_encoder = embedder.model |
|
tokenizer = embedder.tokenizer |
|
max_length = embedder.max_length |
|
|
|
|
|
|
|
if lora_scale is not None and isinstance(self, LoraLoaderMixin): |
|
self._lora_scale = lora_scale |
|
|
|
|
|
adjust_lora_scale_text_encoder(self.text_encoder, lora_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] |
|
|
|
if prompt_embeds is None: |
|
|
|
if isinstance(self, TextualInversionLoaderMixin): |
|
prompt = self.maybe_convert_prompt(prompt, tokenizer) |
|
|
|
text_inputs = tokenizer( |
|
prompt, |
|
padding="max_length", |
|
max_length=max_length, |
|
truncation=True, |
|
return_attention_mask=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}" |
|
) |
|
|
|
attention_mask = text_inputs.attention_mask.to(device) |
|
prompt_embeds = text_encoder( |
|
text_input_ids.to(device), |
|
attention_mask=attention_mask, |
|
) |
|
prompt_embeds = prompt_embeds[0] |
|
attention_mask = attention_mask.repeat(num_images_per_prompt, 1) |
|
else: |
|
attention_mask = None |
|
|
|
if text_encoder is not None: |
|
prompt_embeds_dtype = text_encoder.dtype |
|
elif self.unet is not None: |
|
prompt_embeds_dtype = self.unet.dtype |
|
else: |
|
prompt_embeds_dtype = prompt_embeds.dtype |
|
|
|
prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype, device=device) |
|
|
|
bs_embed, seq_len, _ = prompt_embeds.shape |
|
|
|
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 and negative_prompt_embeds is None: |
|
uncond_tokens: List[str] |
|
if negative_prompt is None: |
|
uncond_tokens = [""] * batch_size |
|
elif 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 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 |
|
|
|
|
|
if isinstance(self, TextualInversionLoaderMixin): |
|
uncond_tokens = self.maybe_convert_prompt(uncond_tokens, tokenizer) |
|
|
|
max_length = prompt_embeds.shape[1] |
|
uncond_input = tokenizer( |
|
uncond_tokens, |
|
padding="max_length", |
|
max_length=max_length, |
|
truncation=True, |
|
return_tensors="pt", |
|
) |
|
|
|
uncond_attention_mask = uncond_input.attention_mask.to(device) |
|
negative_prompt_embeds = text_encoder( |
|
uncond_input.input_ids.to(device), |
|
attention_mask=uncond_attention_mask, |
|
) |
|
negative_prompt_embeds = negative_prompt_embeds[0] |
|
uncond_attention_mask = uncond_attention_mask.repeat(num_images_per_prompt, 1) |
|
else: |
|
uncond_attention_mask = None |
|
|
|
if do_classifier_free_guidance: |
|
|
|
seq_len = negative_prompt_embeds.shape[1] |
|
|
|
negative_prompt_embeds = negative_prompt_embeds.to(dtype=prompt_embeds_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) |
|
|
|
return prompt_embeds, negative_prompt_embeds, attention_mask, uncond_attention_mask |
|
|
|
def _convert_to_rgb(self, image): |
|
return image.convert('RGB') |
|
|
|
def image_transform(self, image_size=224): |
|
transform = T.Compose([ |
|
T.Resize((image_size, image_size), interpolation=T.InterpolationMode.BICUBIC), |
|
self._convert_to_rgb, |
|
T.ToTensor(), |
|
T.Normalize((0.48145466, 0.4578275, 0.40821073), (0.26862954, 0.26130258, 0.27577711)), |
|
]) |
|
return transform |
|
|
|
def encode_img(self, img, device, do_classifier_free_guidance): |
|
|
|
|
|
img = img[0] |
|
image_preprocess = self.image_transform(224) |
|
img_for_clip = image_preprocess(img) |
|
|
|
img_for_clip = img_for_clip.unsqueeze(0) |
|
img_clip_embedding = self.img_encoder(img_for_clip.to(device)).to(dtype=torch.float16) |
|
|
|
if do_classifier_free_guidance: |
|
negative_img_clip_embedding = torch.zeros_like(img_clip_embedding) |
|
return img_clip_embedding, negative_img_clip_embedding |
|
|
|
|
|
|
|
def run_safety_checker(self, image, device, dtype): |
|
if self.safety_checker is None: |
|
has_nsfw_concept = None |
|
else: |
|
if torch.is_tensor(image): |
|
feature_extractor_input = self.image_processor.postprocess(image, output_type="pil") |
|
else: |
|
feature_extractor_input = self.image_processor.numpy_to_pil(image) |
|
safety_checker_input = self.feature_extractor(feature_extractor_input, return_tensors="pt").to(device) |
|
image, has_nsfw_concept = self.safety_checker( |
|
images=image, clip_input=safety_checker_input.pixel_values.to(dtype) |
|
) |
|
return image, has_nsfw_concept |
|
|
|
|
|
def decode_latents(self, latents): |
|
deprecation_message = "The decode_latents method is deprecated and will be removed in 1.0.0. Please use VaeImageProcessor.postprocess(...) instead" |
|
deprecate("decode_latents", "1.0.0", deprecation_message, standard_warn=False) |
|
|
|
latents = 1 / self.vae.config.scaling_factor * latents |
|
image = self.vae.decode(latents, return_dict=False)[0] |
|
image = (image / 2 + 0.5).clamp(0, 1) |
|
|
|
image = image.cpu().permute(0, 2, 3, 1).float().numpy() |
|
return image |
|
|
|
|
|
def prepare_extra_step_kwargs(self, generator, eta): |
|
|
|
|
|
|
|
|
|
|
|
accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys()) |
|
extra_step_kwargs = {} |
|
if accepts_eta: |
|
extra_step_kwargs["eta"] = eta |
|
|
|
|
|
accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys()) |
|
if accepts_generator: |
|
extra_step_kwargs["generator"] = generator |
|
return extra_step_kwargs |
|
|
|
def check_inputs( |
|
self, |
|
prompt, |
|
height, |
|
width, |
|
callback_steps, |
|
negative_prompt=None, |
|
prompt_embeds=None, |
|
negative_prompt_embeds=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 None) or ( |
|
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 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}." |
|
) |
|
|
|
def get_timesteps(self, num_inference_steps, strength, device): |
|
|
|
init_timestep = min(int(num_inference_steps * strength), num_inference_steps) |
|
|
|
t_start = max(num_inference_steps - init_timestep, 0) |
|
timesteps = self.scheduler.timesteps[t_start * self.scheduler.order :] |
|
|
|
return timesteps, num_inference_steps - t_start |
|
|
|
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) |
|
|
|
|
|
latents = latents * self.scheduler.init_noise_sigma |
|
return latents |
|
|
|
@torch.no_grad() |
|
@replace_example_docstring(EXAMPLE_DOC_STRING) |
|
def __call__( |
|
self, |
|
height: int, |
|
width: int, |
|
prompt: Union[str, List[str]] = None, |
|
num_inference_steps: Optional[int] = 50, |
|
guidance_scale: Optional[float] = 7.5, |
|
negative_prompt: Optional[Union[str, List[str]]] = None, |
|
num_images_per_prompt: Optional[int] = 1, |
|
eta: Optional[float] = 0.0, |
|
generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, |
|
latents: Optional[torch.FloatTensor] = None, |
|
prompt_embeds: Optional[torch.FloatTensor] = None, |
|
prompt_embeds_t5: Optional[torch.FloatTensor] = None, |
|
negative_prompt_embeds: Optional[torch.FloatTensor] = None, |
|
negative_prompt_embeds_t5: Optional[torch.FloatTensor] = None, |
|
output_type: Optional[str] = "pil", |
|
return_dict: bool = True, |
|
callback: Optional[Callable[[int, int, torch.FloatTensor, torch.FloatTensor], None]] = None, |
|
callback_steps: int = 1, |
|
cross_attention_kwargs: Optional[Dict[str, Any]] = None, |
|
guidance_rescale: float = 0.0, |
|
image_meta_size: Optional[torch.LongTensor] = None, |
|
style: Optional[torch.LongTensor] = None, |
|
progress: bool = True, |
|
use_fp16: bool = False, |
|
freqs_cis_img: Optional[tuple] = None, |
|
learn_sigma: bool = True, |
|
): |
|
r""" |
|
The call function to the pipeline for generation. |
|
|
|
Args: |
|
height (`int`): |
|
The height in pixels of the generated image. |
|
width (`int`): |
|
The width in pixels of the generated image. |
|
prompt (`str` or `List[str]`, *optional*): |
|
The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`. |
|
image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`): |
|
`Image`, numpy array or tensor representing an image batch to be used as the starting point. For both |
|
numpy array and pytorch tensor, the expected value range is between `[0, 1]` If it's a tensor or a list |
|
or tensors, the expected shape should be `(B, C, H, W)` or `(C, H, W)`. If it is a numpy array or a |
|
list of arrays, the expected shape should be `(B, H, W, C)` or `(H, W, C)` It can also accept image |
|
latents as `image`, but if passing latents directly it is not encoded again. |
|
strength (`float`, *optional*, defaults to 1.0): |
|
Indicates extent to transform the reference `image`. Must be between 0 and 1. `image` is used as a |
|
starting point and more noise is added the higher the `strength`. The number of denoising steps depends |
|
on the amount of noise initially added. When `strength` is 1, added noise is maximum and the denoising |
|
process runs for the full number of iterations specified in `num_inference_steps`. A value of 1 |
|
essentially ignores `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. This parameter is modulated by `strength`. |
|
guidance_scale (`float`, *optional*, defaults to 7.5): |
|
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`. |
|
negative_prompt (`str` or `List[str]`, *optional*): |
|
The prompt or prompts to guide what to not include in image generation. If not defined, you need to |
|
pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`). |
|
num_images_per_prompt (`int`, *optional*, defaults to 1): |
|
The number of images to generate per prompt. |
|
eta (`float`, *optional*, defaults to 0.0): |
|
Corresponds to parameter eta (η) from the [DDIM](https://arxiv.org/abs/2010.02502) paper. Only applies |
|
to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers. |
|
generator (`torch.Generator` or `List[torch.Generator]`, *optional*): |
|
A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make |
|
generation deterministic. |
|
prompt_embeds (`torch.FloatTensor`, *optional*): |
|
Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not |
|
provided, text embeddings are generated from the `prompt` input argument. |
|
negative_prompt_embeds (`torch.FloatTensor`, *optional*): |
|
Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If |
|
not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. |
|
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 [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a |
|
plain tuple. |
|
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, |
|
pred_x0: 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. |
|
cross_attention_kwargs (`dict`, *optional*): |
|
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). |
|
|
|
Examples: |
|
|
|
Returns: |
|
[`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`: |
|
If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] is returned, |
|
otherwise a `tuple` is returned where the first element is a list with the generated images and the |
|
second element is a list of `bool`s indicating whether the corresponding generated image contains |
|
"not-safe-for-work" (nsfw) content. |
|
""" |
|
|
|
self.check_inputs( |
|
prompt, height, width, callback_steps, negative_prompt, prompt_embeds, negative_prompt_embeds |
|
) |
|
|
|
|
|
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] |
|
|
|
device = self._execution_device |
|
|
|
|
|
|
|
do_classifier_free_guidance = guidance_scale > 1.0 |
|
|
|
|
|
text_encoder_lora_scale = ( |
|
cross_attention_kwargs.get("scale", None) if cross_attention_kwargs is not None else None |
|
) |
|
|
|
prompt_embeds, negative_prompt_embeds, attention_mask, uncond_attention_mask = \ |
|
self.encode_prompt(prompt, |
|
device, |
|
num_images_per_prompt, |
|
do_classifier_free_guidance, |
|
negative_prompt, |
|
prompt_embeds=prompt_embeds, |
|
negative_prompt_embeds=negative_prompt_embeds, |
|
lora_scale=text_encoder_lora_scale, |
|
) |
|
prompt_embeds_t5, negative_prompt_embeds_t5, attention_mask_t5, uncond_attention_mask_t5 = \ |
|
self.encode_prompt(prompt, |
|
device, |
|
num_images_per_prompt, |
|
do_classifier_free_guidance, |
|
negative_prompt, |
|
prompt_embeds=prompt_embeds_t5, |
|
negative_prompt_embeds=negative_prompt_embeds_t5, |
|
lora_scale=text_encoder_lora_scale, |
|
embedder=self.embedder_t5, |
|
) |
|
|
|
|
|
|
|
|
|
if do_classifier_free_guidance: |
|
prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds]) |
|
attention_mask = torch.cat([uncond_attention_mask, attention_mask]) |
|
prompt_embeds_t5 = torch.cat([negative_prompt_embeds_t5, prompt_embeds_t5]) |
|
attention_mask_t5 = torch.cat([uncond_attention_mask_t5, attention_mask_t5]) |
|
|
|
|
|
self.scheduler.set_timesteps(num_inference_steps, device=device) |
|
timesteps = self.scheduler.timesteps |
|
|
|
|
|
num_channels_latents = self.unet.config.in_channels |
|
latents = self.prepare_latents(batch_size * num_images_per_prompt, |
|
num_channels_latents, |
|
height, |
|
width, |
|
prompt_embeds.dtype, |
|
device, |
|
generator, |
|
latents, |
|
) |
|
|
|
|
|
extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta) |
|
|
|
|
|
num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order |
|
with self.progress_bar(total=num_inference_steps) as progress_bar: |
|
for i, t in enumerate(timesteps): |
|
|
|
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) |
|
|
|
t_expand = torch.tensor([t] * latent_model_input.shape[0], device=latent_model_input.device) |
|
|
|
if use_fp16: |
|
latent_model_input = latent_model_input.half() |
|
t_expand = t_expand.half() |
|
prompt_embeds = prompt_embeds.half() |
|
ims = image_meta_size.half() if image_meta_size is not None else None |
|
else: |
|
ims = image_meta_size if image_meta_size is not None else None |
|
|
|
|
|
if self.infer_mode in ["fa", "torch"]: |
|
noise_pred = self.unet( |
|
latent_model_input, |
|
t_expand, |
|
encoder_hidden_states=prompt_embeds, |
|
text_embedding_mask=attention_mask, |
|
encoder_hidden_states_t5=prompt_embeds_t5, |
|
text_embedding_mask_t5=attention_mask_t5, |
|
image_meta_size=ims, |
|
style=style, |
|
cos_cis_img=freqs_cis_img[0], |
|
sin_cis_img=freqs_cis_img[1], |
|
return_dict=False, |
|
) |
|
elif self.infer_mode == "trt": |
|
raise NotImplementedError("TensorRT model is not supported yet.") |
|
else: |
|
raise ValueError("[ERROR] invalid inference mode! please check your config file") |
|
if learn_sigma: |
|
noise_pred, _ = noise_pred.chunk(2, dim=1) |
|
|
|
|
|
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: |
|
|
|
noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=guidance_rescale) |
|
|
|
|
|
results = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=True) |
|
latents = results.prev_sample |
|
pred_x0 = results.pred_original_sample if hasattr(results, 'pred_original_sample') else None |
|
|
|
|
|
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, pred_x0) |
|
|
|
if not output_type == "latent": |
|
image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0] |
|
image, has_nsfw_concept = self.run_safety_checker(image, device, prompt_embeds.dtype) |
|
else: |
|
image = latents |
|
has_nsfw_concept = None |
|
|
|
if has_nsfw_concept is None: |
|
do_denormalize = [True] * image.shape[0] |
|
else: |
|
do_denormalize = [not has_nsfw for has_nsfw in has_nsfw_concept] |
|
|
|
image = self.image_processor.postprocess(image, output_type=output_type, do_denormalize=do_denormalize) |
|
|
|
|
|
self.maybe_free_model_hooks() |
|
|
|
if not return_dict: |
|
return (image, has_nsfw_concept) |
|
|
|
return StableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept) |
|
|