diff --git a/iopaint/const.py b/iopaint/const.py index 5d272b8..baff017 100644 --- a/iopaint/const.py +++ b/iopaint/const.py @@ -30,6 +30,9 @@ DIFFUSION_MODELS = [ "Sanster/anything-4.0-inpainting", "diffusers/stable-diffusion-xl-1.0-inpainting-0.1", "Fantasy-Studio/Paint-by-Example", + "RunDiffusion/Juggernaut-XI-v11", + "SG161222/RealVisXL_V5.0", + "eienmojiki/Anything-XL", POWERPAINT_NAME, ANYTEXT_NAME, ] @@ -83,6 +86,10 @@ SDXL_CONTROLNET_CHOICES = [ "diffusers/controlnet-depth-sdxl-1.0-small", ] +SDXL_BRUSHNET_CHOICES = [ + "Regulus0725/random_mask_brushnet_ckpt_sdxl_regulus_v1" +] + LOCAL_FILES_ONLY_HELP = """ When loading diffusion models, using local files only, not connect to HuggingFace server. """ diff --git a/iopaint/model/brushnet/brushnet_xl_wrapper.py b/iopaint/model/brushnet/brushnet_xl_wrapper.py new file mode 100644 index 0000000..ed35a2e --- /dev/null +++ b/iopaint/model/brushnet/brushnet_xl_wrapper.py @@ -0,0 +1,181 @@ +import PIL.Image +import cv2 +import torch +from loguru import logger +import numpy as np + +from ..base import DiffusionInpaintModel +from ..helper.cpu_text_encoder import CPUTextEncoderWrapper +from ..original_sd_configs import get_config_files +from ..utils import ( + handle_from_pretrained_exceptions, + get_torch_dtype, + enable_low_mem, + is_local_files_only, +) +from .brushnet import BrushNetModel +from .brushnet_unet_forward import brushnet_unet_forward +from .unet_2d_blocks import ( + CrossAttnDownBlock2D_forward, + DownBlock2D_forward, + CrossAttnUpBlock2D_forward, + UpBlock2D_forward, +) +from ...schema import InpaintRequest, ModelType +from ...const import SDXL_BRUSHNET_CHOICES + + +class BrushNetXLWrapper(DiffusionInpaintModel): + name = "RunDiffusion/Juggernaut-XI-v11" + pad_mod = 8 + min_size = 1024 + model_id_or_path = "RunDiffusion/Juggernaut-XI-v11" + support_brushnet = True + support_lcm_lora = False + + def init_model(self, device: torch.device, **kwargs): + from .pipeline_brushnet_sd_xl import StableDiffusionXLBrushNetPipeline + + self.model_info = kwargs["model_info"] + self.brushnet_xl_method = SDXL_BRUSHNET_CHOICES[0] + # self.brushnet_xl_method = kwargs["brushnet_xl_method"] + + use_gpu, torch_dtype = get_torch_dtype(device, kwargs.get("no_half", False)) + self.torch_dtype = torch_dtype + + model_kwargs = { + **kwargs.get("pipe_components", {}), + "local_files_only": is_local_files_only(**kwargs), + } + self.local_files_only = model_kwargs["local_files_only"] + + disable_nsfw_checker = kwargs["disable_nsfw"] or kwargs.get( + "cpu_offload", False + ) + if disable_nsfw_checker: + logger.info("Disable Stable Diffusion Model NSFW checker") + model_kwargs.update( + dict( + safety_checker=None, + feature_extractor=None, + requires_safety_checker=False, + ) + ) + + logger.info(f"Loading BrushNet model from {self.brushnet_xl_method}") + brushnet = BrushNetModel.from_pretrained( + self.brushnet_xl_method, torch_dtype=torch_dtype + ) + + if self.model_info.is_single_file_diffusers: + if self.model_info.model_type == ModelType.DIFFUSERS_SD: + model_kwargs["num_in_channels"] = 4 + else: + model_kwargs["num_in_channels"] = 9 + + self.model = StableDiffusionXLBrushNetPipeline.from_single_file( + self.model_id_or_path, + torch_dtype=torch_dtype, + load_safety_checker=not disable_nsfw_checker, + original_config_file=get_config_files()["v1"], + brushnet=brushnet, + **model_kwargs, + ) + else: + self.model = handle_from_pretrained_exceptions( + StableDiffusionXLBrushNetPipeline.from_pretrained, + pretrained_model_name_or_path=self.model_id_or_path, + variant="fp16", + torch_dtype=torch_dtype, + brushnet=brushnet, + **model_kwargs, + ) + + enable_low_mem(self.model, kwargs.get("low_mem", False)) + + if kwargs.get("cpu_offload", False) and use_gpu: + logger.info("Enable sequential cpu offload") + self.model.enable_sequential_cpu_offload(gpu_id=0) + else: + self.model = self.model.to(device) + if kwargs["sd_cpu_textencoder"]: + logger.info("Run Stable Diffusion TextEncoder on CPU") + self.model.text_encoder = CPUTextEncoderWrapper( + self.model.text_encoder, torch_dtype + ) + + self.callback = kwargs.pop("callback", None) + + # Monkey patch the forward method of the UNet to use the brushnet_unet_forward method + self.model.unet.forward = brushnet_unet_forward.__get__( + self.model.unet, self.model.unet.__class__ + ) + + for down_block in self.model.brushnet.down_blocks: + down_block.forward = DownBlock2D_forward.__get__( + down_block, down_block.__class__ + ) + for up_block in self.model.brushnet.up_blocks: + up_block.forward = UpBlock2D_forward.__get__(up_block, up_block.__class__) + + # Monkey patch unet down_blocks to use CrossAttnDownBlock2D_forward + for down_block in self.model.unet.down_blocks: + if down_block.__class__.__name__ == "CrossAttnDownBlock2D": + down_block.forward = CrossAttnDownBlock2D_forward.__get__( + down_block, down_block.__class__ + ) + else: + down_block.forward = DownBlock2D_forward.__get__( + down_block, down_block.__class__ + ) + + for up_block in self.model.unet.up_blocks: + if up_block.__class__.__name__ == "CrossAttnUpBlock2D": + up_block.forward = CrossAttnUpBlock2D_forward.__get__( + up_block, up_block.__class__ + ) + else: + up_block.forward = UpBlock2D_forward.__get__( + up_block, up_block.__class__ + ) + + def switch_brushnet_method(self, new_method: str): + self.brushnet_method = new_method + brushnet_xl = BrushNetModel.from_pretrained( + new_method, + local_files_only=self.local_files_only, + torch_dtype=self.torch_dtype, + ).to(self.model.device) + self.model.brushnet = brushnet_xl + + def forward(self, image, mask, config: InpaintRequest): + """Input image and output image have same size + image: [H, W, C] RGB + mask: [H, W, 1] 255 means area to repaint + return: BGR IMAGE + """ + self.set_scheduler(config) + + img_h, img_w = image.shape[:2] + normalized_mask = mask[:, :].astype("float32") / 255.0 + image = image * (1 - normalized_mask) + image = image.astype(np.uint8) + output = self.model( + image=PIL.Image.fromarray(image), + prompt=config.prompt, + negative_prompt=config.negative_prompt, + mask=PIL.Image.fromarray(mask[:, :, -1], mode="L").convert("RGB"), + num_inference_steps=config.sd_steps, + # strength=config.sd_strength, + guidance_scale=config.sd_guidance_scale, + output_type="np", + callback_on_step_end=self.callback, + height=img_h, + width=img_w, + generator=torch.manual_seed(config.sd_seed), + brushnet_conditioning_scale=config.brushnet_conditioning_scale, + ).images[0] + + output = (output * 255).round().astype("uint8") + output = cv2.cvtColor(output, cv2.COLOR_RGB2BGR) + return output diff --git a/iopaint/model/brushnet/pipeline_brushnet_sd_xl.py b/iopaint/model/brushnet/pipeline_brushnet_sd_xl.py new file mode 100644 index 0000000..dbd82ca --- /dev/null +++ b/iopaint/model/brushnet/pipeline_brushnet_sd_xl.py @@ -0,0 +1,1535 @@ +# https://github.com/TencentARC/BrushNet +import inspect +from typing import Any, Callable, Dict, List, Optional, Tuple, Union + +import numpy as np +import PIL.Image +import torch +import torch.nn.functional as F +from transformers import ( + CLIPImageProcessor, + CLIPTextModel, + CLIPTextModelWithProjection, + CLIPTokenizer, + CLIPVisionModelWithProjection, +) + +from diffusers.utils.import_utils import is_invisible_watermark_available + +from diffusers.image_processor import PipelineImageInput, VaeImageProcessor +from diffusers.loaders import ( + FromSingleFileMixin, + IPAdapterMixin, + StableDiffusionXLLoraLoaderMixin, + TextualInversionLoaderMixin, +) +from diffusers.models import AutoencoderKL, ImageProjection, UNet2DConditionModel +from diffusers.models.attention_processor import ( + AttnProcessor2_0, + LoRAAttnProcessor2_0, + LoRAXFormersAttnProcessor, + XFormersAttnProcessor, +) +from diffusers.models.lora import adjust_lora_scale_text_encoder +from diffusers.schedulers import KarrasDiffusionSchedulers +from diffusers.utils import ( + USE_PEFT_BACKEND, + deprecate, + logging, + replace_example_docstring, + scale_lora_layers, + unscale_lora_layers, +) +from diffusers.utils.torch_utils import is_compiled_module, is_torch_version, randn_tensor +from diffusers.pipelines.pipeline_utils import DiffusionPipeline, StableDiffusionMixin +from diffusers.pipelines.stable_diffusion_xl.pipeline_output import StableDiffusionXLPipelineOutput +from .brushnet import BrushNetModel + + +if is_invisible_watermark_available(): + from diffusers.pipelines.stable_diffusion_xl.watermark import StableDiffusionXLWatermarker + +# from .multibrushnet import MultiBrushNetModel + +logger = logging.get_logger(__name__) # pylint: disable=invalid-name + + +EXAMPLE_DOC_STRING = """ + Examples: + ```py + >>> # !pip install opencv-python transformers accelerate + >>> from diffusers import StableDiffusionXLBrushNetPipeline, BrushNetModel, AutoencoderKL + >>> from diffusers.utils import load_image + >>> import numpy as np + >>> import torch + + >>> import cv2 + >>> from PIL import Image + + >>> prompt = "aerial view, a futuristic research complex in a bright foggy jungle, hard lighting" + >>> negative_prompt = "low quality, bad quality, sketches" + + >>> # download an image + >>> image = load_image( + ... "https://hf.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_brushnet/hf-logo.png" + ... ) + + >>> # initialize the models and pipeline + >>> brushnet_conditioning_scale = 0.5 # recommended for good generalization + >>> brushnet = BrushNetModel.from_pretrained( + ... "diffusers/brushnet-canny-sdxl-1.0", torch_dtype=torch.float16 + ... ) + >>> vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16) + >>> pipe = StableDiffusionXLBrushNetPipeline.from_pretrained( + ... "stabilityai/stable-diffusion-xl-base-1.0", brushnet=brushnet, vae=vae, torch_dtype=torch.float16 + ... ) + >>> pipe.enable_model_cpu_offload() + + >>> # get canny image + >>> image = np.array(image) + >>> image = cv2.Canny(image, 100, 200) + >>> image = image[:, :, None] + >>> image = np.concatenate([image, image, image], axis=2) + >>> canny_image = Image.fromarray(image) + + >>> # generate image + >>> image = pipe( + ... prompt, brushnet_conditioning_scale=brushnet_conditioning_scale, image=canny_image + ... ).images[0] + ``` +""" + + +class StableDiffusionXLBrushNetPipeline( + DiffusionPipeline, + StableDiffusionMixin, + TextualInversionLoaderMixin, + StableDiffusionXLLoraLoaderMixin, + IPAdapterMixin, + FromSingleFileMixin, +): + r""" + Pipeline for text-to-image generation using Stable Diffusion XL with BrushNet guidance. + + 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.StableDiffusionXLLoraLoaderMixin.load_lora_weights`] for loading LoRA weights + - [`~loaders.StableDiffusionXLLoraLoaderMixin.save_lora_weights`] for saving LoRA weights + - [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files + - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters + + Args: + vae ([`AutoencoderKL`]): + Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations. + text_encoder ([`~transformers.CLIPTextModel`]): + Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)). + text_encoder_2 ([`~transformers.CLIPTextModelWithProjection`]): + Second frozen text-encoder + ([laion/CLIP-ViT-bigG-14-laion2B-39B-b160k](https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k)). + tokenizer ([`~transformers.CLIPTokenizer`]): + A `CLIPTokenizer` to tokenize text. + tokenizer_2 ([`~transformers.CLIPTokenizer`]): + A `CLIPTokenizer` to tokenize text. + unet ([`UNet2DConditionModel`]): + A `UNet2DConditionModel` to denoise the encoded image latents. + brushnet ([`BrushNetModel`] or `List[BrushNetModel]`): + Provides additional conditioning to the `unet` during the denoising process. If you set multiple + BrushNets as a list, the outputs from each BrushNet are added together to create one combined + additional conditioning. + 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`]. + force_zeros_for_empty_prompt (`bool`, *optional*, defaults to `"True"`): + Whether the negative prompt embeddings should always be set to 0. Also see the config of + `stabilityai/stable-diffusion-xl-base-1-0`. + add_watermarker (`bool`, *optional*): + Whether to use the [invisible_watermark](https://github.com/ShieldMnt/invisible-watermark/) library to + watermark output images. If not defined, it defaults to `True` if the package is installed; otherwise no + watermarker is used. + """ + + # leave brushnet out on purpose because it iterates with unet + model_cpu_offload_seq = "text_encoder->text_encoder_2->image_encoder->unet->vae" + _optional_components = [ + "tokenizer", + "tokenizer_2", + "text_encoder", + "text_encoder_2", + "feature_extractor", + "image_encoder", + ] + _callback_tensor_inputs = ["latents", "prompt_embeds", "negative_prompt_embeds"] + + def __init__( + self, + vae: AutoencoderKL, + text_encoder: CLIPTextModel, + text_encoder_2: CLIPTextModelWithProjection, + tokenizer: CLIPTokenizer, + tokenizer_2: CLIPTokenizer, + unet: UNet2DConditionModel, + brushnet: Union[BrushNetModel, List[BrushNetModel], Tuple[BrushNetModel]], #MultiBrushNetModel], + scheduler: KarrasDiffusionSchedulers, + force_zeros_for_empty_prompt: bool = True, + add_watermarker: Optional[bool] = None, + feature_extractor: CLIPImageProcessor = None, + image_encoder: CLIPVisionModelWithProjection = None, + ): + super().__init__() + + # if isinstance(brushnet, (list, tuple)): + # brushnet = MultiBrushNetModel(brushnet) + + self.register_modules( + vae=vae, + text_encoder=text_encoder, + text_encoder_2=text_encoder_2, + tokenizer=tokenizer, + tokenizer_2=tokenizer_2, + unet=unet, + brushnet=brushnet, + scheduler=scheduler, + feature_extractor=feature_extractor, + image_encoder=image_encoder, + ) + self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) + self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor, do_convert_rgb=True) + # self.control_image_processor = VaeImageProcessor( + # vae_scale_factor=self.vae_scale_factor, do_convert_rgb=True, do_normalize=False + # ) + 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 + + self.register_to_config(force_zeros_for_empty_prompt=force_zeros_for_empty_prompt) + + # 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: process 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.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.encode_image + def encode_image(self, image, device, num_images_per_prompt, output_hidden_states=None): + dtype = next(self.image_encoder.parameters()).dtype + + if not isinstance(image, torch.Tensor): + image = self.feature_extractor(image, return_tensors="pt").pixel_values + + image = image.to(device=device, dtype=dtype) + if output_hidden_states: + image_enc_hidden_states = self.image_encoder(image, output_hidden_states=True).hidden_states[-2] + image_enc_hidden_states = image_enc_hidden_states.repeat_interleave(num_images_per_prompt, dim=0) + uncond_image_enc_hidden_states = self.image_encoder( + torch.zeros_like(image), output_hidden_states=True + ).hidden_states[-2] + uncond_image_enc_hidden_states = uncond_image_enc_hidden_states.repeat_interleave( + num_images_per_prompt, dim=0 + ) + return image_enc_hidden_states, uncond_image_enc_hidden_states + else: + image_embeds = self.image_encoder(image).image_embeds + image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0) + uncond_image_embeds = torch.zeros_like(image_embeds) + + return image_embeds, uncond_image_embeds + + # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_ip_adapter_image_embeds + def prepare_ip_adapter_image_embeds( + self, ip_adapter_image, ip_adapter_image_embeds, device, num_images_per_prompt, do_classifier_free_guidance + ): + if ip_adapter_image_embeds is None: + if not isinstance(ip_adapter_image, list): + ip_adapter_image = [ip_adapter_image] + + if len(ip_adapter_image) != len(self.unet.encoder_hid_proj.image_projection_layers): + raise ValueError( + f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.unet.encoder_hid_proj.image_projection_layers)} IP Adapters." + ) + + image_embeds = [] + for single_ip_adapter_image, image_proj_layer in zip( + ip_adapter_image, self.unet.encoder_hid_proj.image_projection_layers + ): + output_hidden_state = not isinstance(image_proj_layer, ImageProjection) + single_image_embeds, single_negative_image_embeds = self.encode_image( + single_ip_adapter_image, device, 1, output_hidden_state + ) + single_image_embeds = torch.stack([single_image_embeds] * num_images_per_prompt, dim=0) + single_negative_image_embeds = torch.stack( + [single_negative_image_embeds] * num_images_per_prompt, dim=0 + ) + + if do_classifier_free_guidance: + single_image_embeds = torch.cat([single_negative_image_embeds, single_image_embeds]) + single_image_embeds = single_image_embeds.to(device) + + image_embeds.append(single_image_embeds) + else: + repeat_dims = [1] + image_embeds = [] + for single_image_embeds in ip_adapter_image_embeds: + if do_classifier_free_guidance: + single_negative_image_embeds, single_image_embeds = single_image_embeds.chunk(2) + single_image_embeds = single_image_embeds.repeat( + num_images_per_prompt, *(repeat_dims * len(single_image_embeds.shape[1:])) + ) + single_negative_image_embeds = single_negative_image_embeds.repeat( + num_images_per_prompt, *(repeat_dims * len(single_negative_image_embeds.shape[1:])) + ) + single_image_embeds = torch.cat([single_negative_image_embeds, single_image_embeds]) + else: + single_image_embeds = single_image_embeds.repeat( + num_images_per_prompt, *(repeat_dims * len(single_image_embeds.shape[1:])) + ) + image_embeds.append(single_image_embeds) + + return image_embeds + + # 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 + + def check_inputs( + self, + prompt, + prompt_2, + image, + mask, + callback_steps, + negative_prompt=None, + negative_prompt_2=None, + prompt_embeds=None, + negative_prompt_embeds=None, + pooled_prompt_embeds=None, + ip_adapter_image=None, + ip_adapter_image_embeds=None, + negative_pooled_prompt_embeds=None, + brushnet_conditioning_scale=1.0, + control_guidance_start=0.0, + control_guidance_end=1.0, + callback_on_step_end_tensor_inputs=None, + ): + 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`." + ) + + # `prompt` needs more sophisticated handling when there are multiple + # conditionings. + # if isinstance(self.brushnet, MultiBrushNetModel): + # if isinstance(prompt, list): + # logger.warning( + # f"You have {len(self.brushnet.nets)} BrushNets and you have passed {len(prompt)}" + # " prompts. The conditionings will be fixed across the prompts." + # ) + + # Check `image` + is_compiled = hasattr(F, "scaled_dot_product_attention") and isinstance( + self.brushnet, torch._dynamo.eval_frame.OptimizedModule + ) + if ( + isinstance(self.brushnet, BrushNetModel) + or is_compiled + and isinstance(self.brushnet._orig_mod, BrushNetModel) + ): + self.check_image(image, mask, prompt, prompt_embeds) + # elif ( + # isinstance(self.brushnet, MultiBrushNetModel) + # or is_compiled + # and isinstance(self.brushnet._orig_mod, MultiBrushNetModel) + # ): + # if not isinstance(image, list): + # raise TypeError("For multiple brushnets: `image` must be type `list`") + + # # When `image` is a nested list: + # # (e.g. [[canny_image_1, pose_image_1], [canny_image_2, pose_image_2]]) + # elif any(isinstance(i, list) for i in image): + # raise ValueError("A single batch of multiple conditionings are supported at the moment.") + # elif len(image) != len(self.brushnet.nets): + # raise ValueError( + # f"For multiple brushnets: `image` must have the same length as the number of brushnets, but got {len(image)} images and {len(self.brushnet.nets)} BrushNets." + # ) + + # for image_ in image: + # self.check_image(image_, prompt, prompt_embeds) + else: + assert False + + # Check `brushnet_conditioning_scale` + if ( + isinstance(self.brushnet, BrushNetModel) + or is_compiled + and isinstance(self.brushnet._orig_mod, BrushNetModel) + ): + if not isinstance(brushnet_conditioning_scale, float): + raise TypeError("For single brushnet: `brushnet_conditioning_scale` must be type `float`.") + # elif ( + # isinstance(self.brushnet, MultiBrushNetModel) + # or is_compiled + # and isinstance(self.brushnet._orig_mod, MultiBrushNetModel) + # ): + # if isinstance(brushnet_conditioning_scale, list): + # if any(isinstance(i, list) for i in brushnet_conditioning_scale): + # raise ValueError("A single batch of multiple conditionings are supported at the moment.") + # elif isinstance(brushnet_conditioning_scale, list) and len(brushnet_conditioning_scale) != len( + # self.brushnet.nets + # ): + # raise ValueError( + # "For multiple brushnets: When `brushnet_conditioning_scale` is specified as `list`, it must have" + # " the same length as the number of brushnets" + # ) + else: + assert False + + if not isinstance(control_guidance_start, (tuple, list)): + control_guidance_start = [control_guidance_start] + + if not isinstance(control_guidance_end, (tuple, list)): + control_guidance_end = [control_guidance_end] + + if len(control_guidance_start) != len(control_guidance_end): + raise ValueError( + f"`control_guidance_start` has {len(control_guidance_start)} elements, but `control_guidance_end` has {len(control_guidance_end)} elements. Make sure to provide the same number of elements to each list." + ) + + # if isinstance(self.brushnet, MultiBrushNetModel): + # if len(control_guidance_start) != len(self.brushnet.nets): + # raise ValueError( + # f"`control_guidance_start`: {control_guidance_start} has {len(control_guidance_start)} elements but there are {len(self.brushnet.nets)} brushnets available. Make sure to provide {len(self.brushnet.nets)}." + # ) + + for start, end in zip(control_guidance_start, control_guidance_end): + if start >= end: + raise ValueError( + f"control guidance start: {start} cannot be larger or equal to control guidance end: {end}." + ) + if start < 0.0: + raise ValueError(f"control guidance start: {start} can't be smaller than 0.") + if end > 1.0: + raise ValueError(f"control guidance end: {end} can't be larger than 1.0.") + + if ip_adapter_image is not None and ip_adapter_image_embeds is not None: + raise ValueError( + "Provide either `ip_adapter_image` or `ip_adapter_image_embeds`. Cannot leave both `ip_adapter_image` and `ip_adapter_image_embeds` defined." + ) + + if ip_adapter_image_embeds is not None: + if not isinstance(ip_adapter_image_embeds, list): + raise ValueError( + f"`ip_adapter_image_embeds` has to be of type `list` but is {type(ip_adapter_image_embeds)}" + ) + elif ip_adapter_image_embeds[0].ndim not in [3, 4]: + raise ValueError( + f"`ip_adapter_image_embeds` has to be a list of 3D or 4D tensors but is {ip_adapter_image_embeds[0].ndim}D" + ) + + # Copied from diffusers.pipelines.brushnet.pipeline_brushnet.StableDiffusionBrushNetPipeline.check_image + def check_image(self, image, mask, prompt, prompt_embeds): + image_is_pil = isinstance(image, PIL.Image.Image) + image_is_tensor = isinstance(image, torch.Tensor) + image_is_np = isinstance(image, np.ndarray) + image_is_pil_list = isinstance(image, list) and isinstance(image[0], PIL.Image.Image) + image_is_tensor_list = isinstance(image, list) and isinstance(image[0], torch.Tensor) + image_is_np_list = isinstance(image, list) and isinstance(image[0], np.ndarray) + + if ( + not image_is_pil + and not image_is_tensor + and not image_is_np + and not image_is_pil_list + and not image_is_tensor_list + and not image_is_np_list + ): + raise TypeError( + f"image must be passed and be one of PIL image, numpy array, torch tensor, list of PIL images, list of numpy arrays or list of torch tensors, but is {type(image)}" + ) + + mask_is_pil = isinstance(mask, PIL.Image.Image) + mask_is_tensor = isinstance(mask, torch.Tensor) + mask_is_np = isinstance(mask, np.ndarray) + mask_is_pil_list = isinstance(mask, list) and isinstance(mask[0], PIL.Image.Image) + mask_is_tensor_list = isinstance(mask, list) and isinstance(mask[0], torch.Tensor) + mask_is_np_list = isinstance(mask, list) and isinstance(mask[0], np.ndarray) + + if ( + not mask_is_pil + and not mask_is_tensor + and not mask_is_np + and not mask_is_pil_list + and not mask_is_tensor_list + and not mask_is_np_list + ): + raise TypeError( + f"mask must be passed and be one of PIL image, numpy array, torch tensor, list of PIL images, list of numpy arrays or list of torch tensors, but is {type(mask)}" + ) + + + if image_is_pil: + image_batch_size = 1 + else: + image_batch_size = len(image) + + if prompt is not None and isinstance(prompt, str): + prompt_batch_size = 1 + elif prompt is not None and isinstance(prompt, list): + prompt_batch_size = len(prompt) + elif prompt_embeds is not None: + prompt_batch_size = prompt_embeds.shape[0] + + if image_batch_size != 1 and image_batch_size != prompt_batch_size: + raise ValueError( + f"If image batch size is not 1, image batch size must be same as prompt batch size. image batch size: {image_batch_size}, prompt batch size: {prompt_batch_size}" + ) + + # Copied from diffusers.pipelines.brushnet.pipeline_brushnet.StableDiffusionBrushNetPipeline.prepare_image + def prepare_image( + self, + image, + width, + height, + batch_size, + num_images_per_prompt, + device, + dtype, + do_classifier_free_guidance=False, + guess_mode=False, + ): + image = self.image_processor.preprocess(image, height=height, width=width).to(dtype=torch.float32) + image_batch_size = image.shape[0] + + if image_batch_size == 1: + repeat_by = batch_size + else: + # image batch size is the same as prompt batch size + repeat_by = num_images_per_prompt + + image = image.repeat_interleave(repeat_by, dim=0) + + image = image.to(device=device, dtype=dtype) + + if do_classifier_free_guidance and not guess_mode: + image = torch.cat([image] * 2) + + return image + + # 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: + noise = randn_tensor(shape, generator=generator, device=device, dtype=dtype) + else: + noise = latents.to(device) + + # scale the initial noise by the standard deviation required by the scheduler + latents = noise * self.scheduler.init_noise_sigma + return latents, noise + + # 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_upscale.StableDiffusionUpscalePipeline.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, + ), + ) + # 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.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding + def get_guidance_scale_embedding(self, w, embedding_dim=512, dtype=torch.float32): + """ + See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298 + + Args: + timesteps (`torch.Tensor`): + generate embedding vectors at these timesteps + embedding_dim (`int`, *optional*, defaults to 512): + dimension of the embeddings to generate + dtype: + data type of the generated embeddings + + Returns: + `torch.FloatTensor`: Embedding vectors with shape `(len(timesteps), embedding_dim)` + """ + assert len(w.shape) == 1 + w = w * 1000.0 + + half_dim = embedding_dim // 2 + emb = torch.log(torch.tensor(10000.0)) / (half_dim - 1) + emb = torch.exp(torch.arange(half_dim, dtype=dtype) * -emb) + emb = w.to(dtype)[:, None] * emb[None, :] + emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1) + if embedding_dim % 2 == 1: # zero pad + emb = torch.nn.functional.pad(emb, (0, 1)) + assert emb.shape == (w.shape[0], embedding_dim) + return emb + + @property + def guidance_scale(self): + return self._guidance_scale + + @property + def clip_skip(self): + return self._clip_skip + + # 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. + @property + def do_classifier_free_guidance(self): + return self._guidance_scale > 1 and self.unet.config.time_cond_proj_dim is None + + @property + def cross_attention_kwargs(self): + return self._cross_attention_kwargs + + @property + def denoising_end(self): + return self._denoising_end + + @property + def num_timesteps(self): + return self._num_timesteps + + @torch.no_grad() + @replace_example_docstring(EXAMPLE_DOC_STRING) + def __call__( + self, + prompt: Union[str, List[str]] = None, + prompt_2: Optional[Union[str, List[str]]] = None, + image: PipelineImageInput = None, + mask: PipelineImageInput = None, + height: Optional[int] = None, + width: Optional[int] = None, + num_inference_steps: int = 50, + denoising_end: Optional[float] = None, + guidance_scale: float = 5.0, + negative_prompt: Optional[Union[str, List[str]]] = None, + negative_prompt_2: Optional[Union[str, List[str]]] = None, + num_images_per_prompt: Optional[int] = 1, + eta: float = 0.0, + 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, + pooled_prompt_embeds: Optional[torch.FloatTensor] = None, + negative_pooled_prompt_embeds: Optional[torch.FloatTensor] = None, + ip_adapter_image: Optional[PipelineImageInput] = None, + ip_adapter_image_embeds: Optional[List[torch.FloatTensor]] = None, + output_type: Optional[str] = "pil", + return_dict: bool = True, + cross_attention_kwargs: Optional[Dict[str, Any]] = None, + brushnet_conditioning_scale: Union[float, List[float]] = 1.0, + guess_mode: bool = False, + control_guidance_start: Union[float, List[float]] = 0.0, + control_guidance_end: Union[float, List[float]] = 1.0, + original_size: Tuple[int, int] = None, + crops_coords_top_left: Tuple[int, int] = (0, 0), + target_size: Tuple[int, int] = None, + negative_original_size: Optional[Tuple[int, int]] = None, + negative_crops_coords_top_left: Tuple[int, int] = (0, 0), + negative_target_size: Optional[Tuple[int, int]] = None, + clip_skip: Optional[int] = None, + callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None, + callback_on_step_end_tensor_inputs: List[str] = ["latents"], + **kwargs, + ): + r""" + The call function to the pipeline for generation. + + Args: + prompt (`str` or `List[str]`, *optional*): + The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`. + prompt_2 (`str` or `List[str]`, *optional*): + The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is + used in both text-encoders. + image (`torch.FloatTensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.FloatTensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: + `List[List[torch.FloatTensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): + The BrushNet input condition to provide guidance to the `unet` for generation. If the type is + specified as `torch.FloatTensor`, it is passed to BrushNet as is. `PIL.Image.Image` can also be + accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height + and/or width are passed, `image` is resized accordingly. If multiple BrushNets are specified in + `init`, images must be passed as a list such that each element of the list can be correctly batched for + input to a single BrushNet. + height (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`): + The height in pixels of the generated image. Anything below 512 pixels won't work well for + [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) + and checkpoints that are not specifically fine-tuned on low resolutions. + width (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`): + The width in pixels of the generated image. Anything below 512 pixels won't work well for + [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) + and checkpoints that are not specifically fine-tuned on low resolutions. + 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 5.0): + 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`. + 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`). + negative_prompt_2 (`str` or `List[str]`, *optional*): + The prompt or prompts to guide what to not include in image generation. This is sent to `tokenizer_2` + and `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders. + 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. + 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 is 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 (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. + pooled_prompt_embeds (`torch.FloatTensor`, *optional*): + Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If + not provided, pooled text embeddings are 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 (prompt + weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input + argument. + ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. + ip_adapter_image_embeds (`List[torch.FloatTensor]`, *optional*): + Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. + Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding + if `do_classifier_free_guidance` is set to `True`. + If not provided, embeddings are computed from the `ip_adapter_image` 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. + 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). + brushnet_conditioning_scale (`float` or `List[float]`, *optional*, defaults to 1.0): + The outputs of the BrushNet are multiplied by `brushnet_conditioning_scale` before they are added + to the residual in the original `unet`. If multiple BrushNets are specified in `init`, you can set + the corresponding scale as a list. + guess_mode (`bool`, *optional*, defaults to `False`): + The BrushNet encoder tries to recognize the content of the input image even if you remove all + prompts. A `guidance_scale` value between 3.0 and 5.0 is recommended. + control_guidance_start (`float` or `List[float]`, *optional*, defaults to 0.0): + The percentage of total steps at which the BrushNet starts applying. + control_guidance_end (`float` or `List[float]`, *optional*, defaults to 1.0): + The percentage of total steps at which the BrushNet stops applying. + 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 `(height, width)` 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 `(height, width)`. 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). + negative_original_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): + To negatively condition the generation process based on a specific image resolution. 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). For more + information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. + negative_crops_coords_top_left (`Tuple[int]`, *optional*, defaults to (0, 0)): + To negatively condition the generation process based on a specific crop coordinates. 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). For more + information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. + negative_target_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): + To negatively condition the generation process based on a target image resolution. It should be as same + as the `target_size` for most cases. 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). For more + information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. + 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. + 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 pipeine class. + + Examples: + + Returns: + [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`: + If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] is returned, + otherwise a `tuple` is returned containing the output images. + """ + + 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 using `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 using `callback_on_step_end`", + ) + + brushnet = self.brushnet._orig_mod if is_compiled_module(self.brushnet) else self.brushnet + + # align format for control guidance + if not isinstance(control_guidance_start, list) and isinstance(control_guidance_end, list): + control_guidance_start = len(control_guidance_end) * [control_guidance_start] + elif not isinstance(control_guidance_end, list) and isinstance(control_guidance_start, list): + control_guidance_end = len(control_guidance_start) * [control_guidance_end] + elif not isinstance(control_guidance_start, list) and not isinstance(control_guidance_end, list): + # mult = len(brushnet.nets) if isinstance(brushnet, MultiBrushNetModel) else 1 + mult = 1 + control_guidance_start, control_guidance_end = ( + mult * [control_guidance_start], + mult * [control_guidance_end], + ) + + # 1. Check inputs. Raise error if not correct + self.check_inputs( + prompt, + prompt_2, + image, + mask, + callback_steps, + negative_prompt, + negative_prompt_2, + prompt_embeds, + negative_prompt_embeds, + pooled_prompt_embeds, + ip_adapter_image, + ip_adapter_image_embeds, + negative_pooled_prompt_embeds, + brushnet_conditioning_scale, + control_guidance_start, + control_guidance_end, + callback_on_step_end_tensor_inputs, + ) + + self._guidance_scale = guidance_scale + self._clip_skip = clip_skip + self._cross_attention_kwargs = cross_attention_kwargs + self._denoising_end = denoising_end + + # 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 + + # if isinstance(brushnet, MultiBrushNetModel) and isinstance(brushnet_conditioning_scale, float): + # brushnet_conditioning_scale = [brushnet_conditioning_scale] * len(brushnet.nets) + + global_pool_conditions = ( + brushnet.config.global_pool_conditions + if isinstance(brushnet, BrushNetModel) + else brushnet.nets[0].config.global_pool_conditions + ) + guess_mode = guess_mode or global_pool_conditions + + # 3.1 Encode input prompt + text_encoder_lora_scale = ( + self.cross_attention_kwargs.get("scale", None) if self.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_2, + device, + num_images_per_prompt, + self.do_classifier_free_guidance, + negative_prompt, + 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, + clip_skip=self.clip_skip, + ) + + # 3.2 Encode ip_adapter_image + if ip_adapter_image is not None or ip_adapter_image_embeds is not None: + image_embeds = self.prepare_ip_adapter_image_embeds( + ip_adapter_image, + ip_adapter_image_embeds, + device, + batch_size * num_images_per_prompt, + self.do_classifier_free_guidance, + ) + + # 4. Prepare image + if isinstance(brushnet, BrushNetModel): + image = self.prepare_image( + image=image, + width=width, + height=height, + batch_size=batch_size * num_images_per_prompt, + num_images_per_prompt=num_images_per_prompt, + device=device, + dtype=brushnet.dtype, + do_classifier_free_guidance=self.do_classifier_free_guidance, + guess_mode=guess_mode, + ) + original_mask = self.prepare_image( + image=mask, + width=width, + height=height, + batch_size=batch_size * num_images_per_prompt, + num_images_per_prompt=num_images_per_prompt, + device=device, + dtype=brushnet.dtype, + do_classifier_free_guidance=self.do_classifier_free_guidance, + guess_mode=guess_mode, + ) + original_mask=(original_mask.sum(1)[:,None,:,:] < 0).to(image.dtype) + height, width = image.shape[-2:] + # elif isinstance(brushnet, MultiBrushNetModel): + # images = [] + + # for image_ in image: + # image_ = self.prepare_image( + # image=image_, + # width=width, + # height=height, + # batch_size=batch_size * num_images_per_prompt, + # num_images_per_prompt=num_images_per_prompt, + # device=device, + # dtype=brushnet.dtype, + # do_classifier_free_guidance=self.do_classifier_free_guidance, + # guess_mode=guess_mode, + # ) + + # images.append(image_) + + # image = images + # height, width = image[0].shape[-2:] + else: + assert False + + # 5. Prepare timesteps + self.scheduler.set_timesteps(num_inference_steps, device=device) + timesteps = self.scheduler.timesteps + self._num_timesteps = len(timesteps) + + # 6. Prepare latent variables + num_channels_latents = self.unet.config.in_channels + latents, noise = self.prepare_latents( + batch_size * num_images_per_prompt, + num_channels_latents, + height, + width, + prompt_embeds.dtype, + device, + generator, + latents, + ) + + # 6.1 prepare condition latents + conditioning_latents=self.vae.encode(image).latent_dist.sample() * self.vae.config.scaling_factor + mask = torch.nn.functional.interpolate( + original_mask, + size=( + conditioning_latents.shape[-2], + conditioning_latents.shape[-1] + ) + ) + conditioning_latents = torch.concat([conditioning_latents,mask],1) + + + # 6.5 Optionally get Guidance Scale Embedding + timestep_cond = None + if self.unet.config.time_cond_proj_dim is not None: + guidance_scale_tensor = torch.tensor(self.guidance_scale - 1).repeat(batch_size * num_images_per_prompt) + timestep_cond = self.get_guidance_scale_embedding( + guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim + ).to(device=device, dtype=latents.dtype) + + # 7. 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) + + # 7.1 Create tensor stating which brushnets to keep + brushnet_keep = [] + for i in range(len(timesteps)): + keeps = [ + 1.0 - float(i / len(timesteps) < s or (i + 1) / len(timesteps) > e) + for s, e in zip(control_guidance_start, control_guidance_end) + ] + brushnet_keep.append(keeps[0] if isinstance(brushnet, BrushNetModel) else keeps) + + # 7.2 Prepare added time ids & embeddings + if isinstance(image, list): + original_size = original_size or image[0].shape[-2:] + else: + original_size = original_size or image.shape[-2:] + target_size = target_size or (height, width) + + 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 negative_original_size is not None and negative_target_size is not None: + negative_add_time_ids = self._get_add_time_ids( + negative_original_size, + negative_crops_coords_top_left, + negative_target_size, + dtype=prompt_embeds.dtype, + text_encoder_projection_dim=text_encoder_projection_dim, + ) + else: + negative_add_time_ids = add_time_ids + + if self.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([negative_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_images_per_prompt, 1) + + # 8. Denoising loop + num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order + + # 8.1 Apply denoising_end + if ( + self.denoising_end is not None + and isinstance(self.denoising_end, float) + and self.denoising_end > 0 + and self.denoising_end < 1 + ): + discrete_timestep_cutoff = int( + round( + self.scheduler.config.num_train_timesteps + - (self.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] + + is_unet_compiled = is_compiled_module(self.unet) + is_brushnet_compiled = is_compiled_module(self.brushnet) + is_torch_higher_equal_2_1 = is_torch_version(">=", "2.1") + with self.progress_bar(total=num_inference_steps) as progress_bar: + for i, t in enumerate(timesteps): + # Relevant thread: + # https://dev-discuss.pytorch.org/t/cudagraphs-in-pytorch-2-0/1428 + if (is_unet_compiled and is_brushnet_compiled) and is_torch_higher_equal_2_1: + torch._inductor.cudagraph_mark_step_begin() + # expand the latents if we are doing classifier free guidance + latent_model_input = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents + latent_model_input = self.scheduler.scale_model_input(latent_model_input, t) + + added_cond_kwargs = {"text_embeds": add_text_embeds, "time_ids": add_time_ids} + + # brushnet(s) inference + if guess_mode and self.do_classifier_free_guidance: + # Infer BrushNet only for the conditional batch. + control_model_input = latents + control_model_input = self.scheduler.scale_model_input(control_model_input, t) + brushnet_prompt_embeds = prompt_embeds.chunk(2)[1] + brushnet_added_cond_kwargs = { + "text_embeds": add_text_embeds.chunk(2)[1], + "time_ids": add_time_ids.chunk(2)[1], + } + else: + control_model_input = latent_model_input + brushnet_prompt_embeds = prompt_embeds + brushnet_added_cond_kwargs = added_cond_kwargs + + if isinstance(brushnet_keep[i], list): + cond_scale = [c * s for c, s in zip(brushnet_conditioning_scale, brushnet_keep[i])] + else: + brushnet_cond_scale = brushnet_conditioning_scale + if isinstance(brushnet_cond_scale, list): + brushnet_cond_scale = brushnet_cond_scale[0] + cond_scale = brushnet_cond_scale * brushnet_keep[i] + + down_block_res_samples, mid_block_res_sample, up_block_res_samples = self.brushnet( + control_model_input, + t, + encoder_hidden_states=brushnet_prompt_embeds, + brushnet_cond=conditioning_latents, + conditioning_scale=cond_scale, + guess_mode=guess_mode, + added_cond_kwargs=brushnet_added_cond_kwargs, + return_dict=False, + ) + + if guess_mode and self.do_classifier_free_guidance: + # Infered BrushNet only for the conditional batch. + # To apply the output of BrushNet to both the unconditional and conditional batches, + # add 0 to the unconditional batch to keep it unchanged. + down_block_res_samples = [torch.cat([torch.zeros_like(d), d]) for d in down_block_res_samples] + mid_block_res_sample = torch.cat([torch.zeros_like(mid_block_res_sample), mid_block_res_sample]) + up_block_res_samples = [torch.cat([torch.zeros_like(d), d]) for d in up_block_res_samples] + + if ip_adapter_image is not None or ip_adapter_image_embeds is not None: + added_cond_kwargs["image_embeds"] = image_embeds + + # predict the noise residual + noise_pred = self.unet( + latent_model_input, + t, + encoder_hidden_states=prompt_embeds, + timestep_cond=timestep_cond, + cross_attention_kwargs=self.cross_attention_kwargs, + down_block_add_samples=down_block_res_samples, + mid_block_add_sample=mid_block_res_sample, + up_block_add_samples=up_block_res_samples, + added_cond_kwargs=added_cond_kwargs, + return_dict=False, + )[0] + + # perform guidance + if self.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) + + # compute the previous noisy sample x_t -> x_t-1 + latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0] + + 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) + prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds) + negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds) + + # 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: + step_idx = i // getattr(self.scheduler, "order", 1) + callback(step_idx, t, latents) + + 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) + + # unscale/denormalize the latents + # denormalize with the mean and std if available and not None + has_latents_mean = hasattr(self.vae.config, "latents_mean") and self.vae.config.latents_mean is not None + has_latents_std = hasattr(self.vae.config, "latents_std") and self.vae.config.latents_std is not None + if has_latents_mean and has_latents_std: + latents_mean = ( + torch.tensor(self.vae.config.latents_mean).view(1, 4, 1, 1).to(latents.device, latents.dtype) + ) + latents_std = ( + torch.tensor(self.vae.config.latents_std).view(1, 4, 1, 1).to(latents.device, latents.dtype) + ) + latents = latents * latents_std / self.vae.config.scaling_factor + latents_mean + else: + latents = latents / self.vae.config.scaling_factor + + image = self.vae.decode(latents, return_dict=False)[0] + + # cast back to fp16 if needed + if needs_upcasting: + self.vae.to(dtype=torch.float16) + else: + image = latents + + if not output_type == "latent": + # 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 all models + self.maybe_free_model_hooks() + + if not return_dict: + return (image,) + + return StableDiffusionXLPipelineOutput(images=image) diff --git a/iopaint/model/sd.py b/iopaint/model/sd.py index 2f6698c..c690755 100644 --- a/iopaint/model/sd.py +++ b/iopaint/model/sd.py @@ -29,7 +29,7 @@ class SD(DiffusionInpaintModel): **kwargs.get("pipe_components", {}), "local_files_only": is_local_files_only(**kwargs), } - disable_nsfw_checker = kwargs["disable_nsfw"] or kwargs.get( + disable_nsfw_checker = kwargs.get("disable_nsfw", False) or kwargs.get( "cpu_offload", False ) if disable_nsfw_checker: @@ -71,7 +71,7 @@ class SD(DiffusionInpaintModel): self.model.enable_sequential_cpu_offload(gpu_id=0) else: self.model = self.model.to(device) - if kwargs["sd_cpu_textencoder"]: + if kwargs.get("sd_cpu_textencoder", False): logger.info("Run Stable Diffusion TextEncoder on CPU") self.model.text_encoder = CPUTextEncoderWrapper( self.model.text_encoder, torch_dtype diff --git a/iopaint/model_manager.py b/iopaint/model_manager.py index 40c66c2..90d9c83 100644 --- a/iopaint/model_manager.py +++ b/iopaint/model_manager.py @@ -8,6 +8,7 @@ from iopaint.download import scan_models from iopaint.helper import switch_mps_device from iopaint.model import models, ControlNet, SD, SDXL from iopaint.model.brushnet.brushnet_wrapper import BrushNetWrapper +from iopaint.model.brushnet.brushnet_xl_wrapper import BrushNetXLWrapper from iopaint.model.power_paint.power_paint_v2 import PowerPaintV2 from iopaint.model.utils import torch_gc, is_local_files_only from iopaint.schema import InpaintRequest, ModelInfo, ModelType @@ -63,7 +64,10 @@ class ModelManager: return ControlNet(device, **kwargs) if model_info.support_brushnet and self.enable_brushnet: - return BrushNetWrapper(device, **kwargs) + if model_info.model_type == ModelType.DIFFUSERS_SD: + return BrushNetWrapper(device, **kwargs) + elif model_info.model_type == ModelType.DIFFUSERS_SDXL: + return BrushNetXLWrapper(device, **kwargs) if model_info.support_powerpaint_v2 and self.enable_powerpaint_v2: return PowerPaintV2(device, **kwargs) diff --git a/iopaint/schema.py b/iopaint/schema.py index 40d4551..108a2d6 100644 --- a/iopaint/schema.py +++ b/iopaint/schema.py @@ -14,6 +14,7 @@ from iopaint.const import ( SD2_CONTROLNET_CHOICES, SD_CONTROLNET_CHOICES, SD_BRUSHNET_CHOICES, + SDXL_BRUSHNET_CHOICES ) from pydantic import BaseModel, Field, computed_field, model_validator @@ -70,6 +71,8 @@ class ModelInfo(BaseModel): def brushnets(self) -> List[str]: if self.model_type in [ModelType.DIFFUSERS_SD]: return SD_BRUSHNET_CHOICES + if self.model_type in [ModelType.DIFFUSERS_SDXL]: + return SDXL_BRUSHNET_CHOICES return [] @computed_field @@ -117,6 +120,7 @@ class ModelInfo(BaseModel): def support_brushnet(self) -> bool: return self.model_type in [ ModelType.DIFFUSERS_SD, + ModelType.DIFFUSERS_SDXL, ] @computed_field diff --git a/web_app/src/components/SidePanel/DiffusionOptions.tsx b/web_app/src/components/SidePanel/DiffusionOptions.tsx index 5489584..6e969bf 100644 --- a/web_app/src/components/SidePanel/DiffusionOptions.tsx +++ b/web_app/src/components/SidePanel/DiffusionOptions.tsx @@ -142,14 +142,14 @@ const DiffusionOptions = () => { }} /> - {/* + updateSettings({ brushnetConditioningScale: vals[0] / 100 }) @@ -159,12 +159,12 @@ const DiffusionOptions = () => { id="brushnet-weight" className="w-[50px] rounded-full" numberValue={settings.brushnetConditioningScale} - allowFloat={false} + allowFloat onNumberValueChange={(val) => { updateSettings({ brushnetConditioningScale: val }) }} /> - */} +