from typing import Any, Callable, Dict, List, Optional, Union import torch from diffusers.utils import replace_example_docstring from transformers import CLIPTokenizer from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion import StableDiffusionPipeline, EXAMPLE_DOC_STRING from diffusers.pipelines.stable_diffusion import StableDiffusionPipelineOutput from NestedScheduler import NestedScheduler class NestedStableDiffusionPipeline(StableDiffusionPipeline): r""" Pipeline for text-to-image generation using Nested Stable Diffusion. 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.) In addition the pipeline inherits the following loading methods: - *Textual-Inversion*: [`loaders.TextualInversionLoaderMixin.load_textual_inversion`] - *LoRA*: [`loaders.LoraLoaderMixin.load_lora_weights`] - *Ckpt*: [`loaders.FromCkptMixin.from_ckpt`] as well as the following saving methods: - *LoRA*: [`loaders.LoraLoaderMixin.save_lora_weights`] 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 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. tokenizer (`CLIPTokenizer`): 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`]. 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 details. feature_extractor ([`CLIPImageProcessor`]): Model that extracts features from generated images to be used as inputs for the `safety_checker`. """ @torch.no_grad() @replace_example_docstring(EXAMPLE_DOC_STRING) def __call__( self, prompt: Union[str, List[str]] = None, height: Optional[int] = None, width: Optional[int] = None, num_inference_steps: int = 5, num_inner_steps: int = 20, guidance_scale: float = 7.5, negative_prompt: Optional[Union[str, List[str]]] = None, num_images_per_prompt: Optional[int] = 1, eta: float = 0.0, inner_eta: float = 0.85, generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None, latents: Optional[torch.FloatTensor] = None, prompt_embeds: Optional[torch.FloatTensor] = None, negative_prompt_embeds: Optional[torch.FloatTensor] = None, output_type: Optional[str] = "pil", 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, coroutine_mode=True): r""" 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. 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 5): The number of outer denoising steps. num_inner_steps (`int`, *optional*, defaults to 20): The number of inner denoising steps. 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`). 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) in the outer diffusion process inner_eta (`float`, *optional*, defaults to 0.85): Corresponds to parameter eta (η) in the inner diffusion process 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`. 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. 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.StableDiffusionPipelineOutput`] 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). Examples: Returns: [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`: [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] if `return_dict` is True, otherwise a `tuple. When returning a tuple, the first element is a list with the generated images, and the second element is a list of `bool`s denoting whether the corresponding generated image likely represents "not-safe-for-work" (nsfw) content, according to the `safety_checker`. """ # 0. Default height and width to unet height = height or self.unet.config.sample_size * self.vae_scale_factor width = width or self.unet.config.sample_size * self.vae_scale_factor # 1. Check inputs. Raise error if not correct self.check_inputs( prompt, height, width, callback_steps, negative_prompt, prompt_embeds, negative_prompt_embeds ) # 2. Define call parameters 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 # 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 prompt_embeds = 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, ) # 4. Prepare timesteps self.scheduler.set_timesteps(num_inference_steps + 1, device=device) timesteps = self.scheduler.timesteps[:-1] # 5. Prepare latent variables 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, ) # 6. Prepare extra step kwargs. extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta) inner_extra_step_kwargs = self.prepare_extra_step_kwargs(generator, inner_eta) # 7. Denoising loop outer_latents = latents.clone() num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order # running the outer diffusion procees anytime_latent = outer_latents.clone() with self.progress_bar(total=num_inference_steps) as progress_bar: for i, t in enumerate(timesteps): # creating the inner diffusion process self.inner_scheduler = NestedScheduler(beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear", clip_sample=False, set_alpha_to_one=False, thresholding=False) self.inner_scheduler.set_timesteps(num_inner_steps, max_timestep=t.item(), device=device) inner_timesteps = self.inner_scheduler.timesteps latents = outer_latents.clone() # running the inner diffusion procees for j, t_tag in enumerate(inner_timesteps): yield (i, j, self.decode_latents(anytime_latent)) # 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.inner_scheduler.scale_model_input(latent_model_input, t_tag) # predict the noise residual noise_pred = self.unet(latent_model_input, t_tag, encoder_hidden_states=prompt_embeds).sample # 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) latents = self.inner_scheduler.step(noise_pred, t_tag, latents, **inner_extra_step_kwargs).prev_sample anytime_latent = latents.clone() # compute the previous noisy sample x_t -> x_t-1 outer_latents = self.scheduler.step(latents, t, outer_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) yield (i+1, j+1, self.decode_latents(outer_latents))