pipeline_onnx_stable_diffusion_controlnet.py

# Copyright 2023 The HuggingFace Team.
# Converted for use with ONNX as part of https://github.com/Amblyopius/Stable-Diffusion-ONNX-FP16
# Special thanks to https://github.com/uchuusen for the initial conversion effort

import inspect
from typing import Callable, List, Optional, Union

import numpy as np
import torch
import PIL
from transformers import CLIPFeatureExtractor, CLIPTokenizer

from diffusers.configuration_utils import FrozenDict
from diffusers.schedulers import DDIMScheduler, LMSDiscreteScheduler, PNDMScheduler
from diffusers.utils import deprecate, logging, PIL_INTERPOLATION
from diffusers.pipelines.onnx_utils import ORT_TO_NP_TYPE, OnnxRuntimeModel
from diffusers.pipeline_utils import DiffusionPipeline
from diffusers.pipelines.stable_diffusion import StableDiffusionPipelineOutput


logger = logging.get_logger(__name__)


class OnnxStableDiffusionControlNetPipeline(DiffusionPipeline):
    vae_encoder: OnnxRuntimeModel
    vae_decoder: OnnxRuntimeModel
    text_encoder: OnnxRuntimeModel
    tokenizer: CLIPTokenizer
    unet: OnnxRuntimeModel
    controlnet: OnnxRuntimeModel
    scheduler: Union[DDIMScheduler, PNDMScheduler, LMSDiscreteScheduler]
    safety_checker: OnnxRuntimeModel
    feature_extractor: CLIPFeatureExtractor

    _optional_components = ["safety_checker", "feature_extractor"]

    def __init__(
        self,
        vae_encoder: OnnxRuntimeModel,
        vae_decoder: OnnxRuntimeModel,
        text_encoder: OnnxRuntimeModel,
        tokenizer: CLIPTokenizer,
        unet: OnnxRuntimeModel,
        controlnet: OnnxRuntimeModel,
        scheduler: Union[DDIMScheduler, PNDMScheduler, LMSDiscreteScheduler],
        safety_checker: OnnxRuntimeModel,
        feature_extractor: CLIPFeatureExtractor,
        requires_safety_checker: bool = True,
    ):
        super().__init__()

        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                f"The configuration file of this scheduler: {scheduler} is outdated.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_encoder=vae_encoder,
            vae_decoder=vae_decoder,
            text_encoder=text_encoder,
            tokenizer=tokenizer,
            unet=unet,
            controlnet=controlnet,
            scheduler=scheduler,
            safety_checker=safety_checker,
            feature_extractor=feature_extractor,
        )
        self.register_to_config(requires_safety_checker=requires_safety_checker)
        
        
    def _default_height_width(self, height, width, image):
        if isinstance(image, list):
            image = image[0]

        if height is None:
            if isinstance(image, PIL.Image.Image):
                height = image.height
            elif isinstance(image, np.ndarray):
                height = image.shape[3]

            height = (height // 8) * 8  # round down to nearest multiple of 8

        if width is None:
            if isinstance(image, PIL.Image.Image):
                width = image.width
            elif isinstance(image, np.ndarray):
                width = image.shape[2]

            width = (width // 8) * 8  # round down to nearest multiple of 8

        return height, width
        
    def prepare_image(self, image, width, height, batch_size, num_images_per_prompt, dtype):
        if not isinstance(image, np.ndarray):
            if isinstance(image, PIL.Image.Image):
                image = [image]

            if isinstance(image[0], PIL.Image.Image):
                image = [
                    np.array(i.resize((width, height), 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 = torch.from_numpy(image)
            elif isinstance(image[0], np.ndarray):
                image = np.concatenate(image, axis=0)
                image = torch.from_numpy(image)

        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)

        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, generator, latents=None):
        shape = (batch_size, num_channels_latents, height // 8, width // 8)
        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 = generator.randn(*shape).astype(dtype)


        # scale the initial noise by the standard deviation required by the scheduler
        latents = latents * self.scheduler.init_noise_sigma
        return latents
    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline.prepare_extra_step_kwargs
    def prepare_extra_step_kwargs(self, generator, eta, torch_gen):
        # 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"] = torch_gen
        return extra_step_kwargs

    def _encode_prompt(self, prompt, num_images_per_prompt, do_classifier_free_guidance, negative_prompt):
        r"""
        Encodes the prompt into text encoder hidden states.

        Args:
            prompt (`str` or `List[str]`):
                prompt to be encoded
            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]`):
                The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored
                if `guidance_scale` is less than `1`).
        """
        batch_size = len(prompt) if isinstance(prompt, list) else 1

        # get prompt text embeddings
        text_inputs = self.tokenizer(
            prompt,
            padding="max_length",
            max_length=self.tokenizer.model_max_length,
            truncation=True,
            return_tensors="np",
        )
        text_input_ids = text_inputs.input_ids
        untruncated_ids = self.tokenizer(prompt, padding="max_length", return_tensors="np").input_ids

        if not np.array_equal(text_input_ids, untruncated_ids):
            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.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" {self.tokenizer.model_max_length} tokens: {removed_text}"
            )

        prompt_embeds = self.text_encoder(input_ids=text_input_ids.astype(np.int32))[0]
        prompt_embeds = np.repeat(prompt_embeds, num_images_per_prompt, axis=0)

        # get unconditional embeddings for classifier free guidance
        if do_classifier_free_guidance:
            uncond_tokens: List[str]
            if negative_prompt is None:
                uncond_tokens = [""] * batch_size
            elif 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] * batch_size
            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

            max_length = text_input_ids.shape[-1]
            uncond_input = self.tokenizer(
                uncond_tokens,
                padding="max_length",
                max_length=max_length,
                truncation=True,
                return_tensors="np",
            )
            negative_prompt_embeds = self.text_encoder(input_ids=uncond_input.input_ids.astype(np.int32))[0]
            negative_prompt_embeds = np.repeat(negative_prompt_embeds, num_images_per_prompt, axis=0)

            # For classifier free guidance, we need to do two forward passes.
            # Here we concatenate the unconditional and text embeddings into a single batch
            # to avoid doing two forward passes
            prompt_embeds = np.concatenate([negative_prompt_embeds, prompt_embeds])

        return prompt_embeds

    def __call__(
        self,
        prompt: Union[str, List[str]],
        image: Union[np.ndarray, PIL.Image.Image] = None,
        height: Optional[int] = None,
        width: Optional[int] = 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[np.random.RandomState] = None,
        latents: Optional[np.ndarray] = None,
        output_type: Optional[str] = "pil",
        return_dict: bool = True,
        callback: Optional[Callable[[int, int, np.ndarray], None]] = None,
        callback_steps: int = 1,
        controlnet_conditioning_scale: float = 1.0,
    ):
        if isinstance(prompt, str):
            batch_size = 1
        elif isinstance(prompt, list):
            batch_size = len(prompt)
        else:
            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
            
            
        if generator:
            torch_seed = generator.randint(2147483647)
            torch_gen = torch.Generator().manual_seed(torch_seed)
        else:
            generator = np.random
            torch_gen = None
            
        height, width = self._default_height_width(height, width, image)

        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)}."
            )

        # 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

        prompt_embeds = self._encode_prompt(
            prompt, num_images_per_prompt, do_classifier_free_guidance, negative_prompt
        )
        
        # 4. Prepare image
        image = self.prepare_image(
            image,
            width,
            height,
            batch_size * num_images_per_prompt,
            num_images_per_prompt,
            np.float32,
        ).numpy()
        
        if do_classifier_free_guidance:
            image = np.concatenate([image] * 2)

        # get the initial random noise unless the user supplied it
        latents_dtype = prompt_embeds.dtype
        latents_shape = (batch_size * num_images_per_prompt, 4, height // 8, width // 8)
        
        num_channels_latents = 4
        latents = self.prepare_latents(
            batch_size * num_images_per_prompt,
            num_channels_latents,
            height,
            width,
            latents_dtype,
            generator,
            latents,
        )
        
        # set timesteps
        self.scheduler.set_timesteps(num_inference_steps)
        timesteps = self.scheduler.timesteps


        # 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]
        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta, torch_gen)

        timestep_dtype = next(
            (input.type for input in self.unet.model.get_inputs() if input.name == "timestep"), "tensor(float)"
        )
        timestep_dtype = ORT_TO_NP_TYPE[timestep_dtype]
        
        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):
                # expand the latents if we are doing classifier free guidance
                latent_model_input = np.concatenate([latents] * 2) if do_classifier_free_guidance else latents
                latent_model_input = self.scheduler.scale_model_input(torch.from_numpy(latent_model_input), t)
                latent_model_input = latent_model_input.cpu().numpy()

                timestep = np.array([t], dtype=timestep_dtype)

                blocksamples = self.controlnet(
                    sample=latent_model_input,
                    timestep=timestep,
                    encoder_hidden_states=prompt_embeds,
                    controlnet_cond=image,
                    conditioning_scale=1.0
                )

                mid_block_res_sample=blocksamples[12]
                down_block_res_samples=blocksamples[0:12]

                down_block_res_samples = [
                    down_block_res_sample * controlnet_conditioning_scale
                    for down_block_res_sample in down_block_res_samples
                ]
                mid_block_res_sample *= controlnet_conditioning_scale

                # predict the noise residual
                
                noise_pred = self.unet(
                    sample=latent_model_input,
                    timestep=timestep,
                    encoder_hidden_states=prompt_embeds,
                    down_block_0=down_block_res_samples[0],
                    down_block_1=down_block_res_samples[1],
                    down_block_2=down_block_res_samples[2],
                    down_block_3=down_block_res_samples[3],
                    down_block_4=down_block_res_samples[4],
                    down_block_5=down_block_res_samples[5],
                    down_block_6=down_block_res_samples[6],
                    down_block_7=down_block_res_samples[7],
                    down_block_8=down_block_res_samples[8],
                    down_block_9=down_block_res_samples[9],
                    down_block_10=down_block_res_samples[10],
                    down_block_11=down_block_res_samples[11],
                    mid_block_additional_residual=mid_block_res_sample
                )
                noise_pred = noise_pred[0]

                # perform guidance
                if do_classifier_free_guidance:
                    noise_pred_uncond, noise_pred_text = np.split(noise_pred, 2)
                    noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)

                # compute the previous noisy sample x_t -> x_t-1
                scheduler_output = self.scheduler.step(
                    torch.from_numpy(noise_pred), t, torch.from_numpy(latents), **extra_step_kwargs
                )
                latents = scheduler_output.prev_sample.numpy()

                # 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)

        latents = 1 / 0.18215 * latents
        # image = self.vae_decoder(latent_sample=latents)[0]
        # it seems likes there is a strange result for using half-precision vae decoder if batchsize>1
        image = np.concatenate(
            [self.vae_decoder(latent_sample=latents[i : i + 1])[0] for i in range(latents.shape[0])]
        )

        image = np.clip(image / 2 + 0.5, 0, 1)
        image = image.transpose((0, 2, 3, 1))

        if self.safety_checker is not None:
            safety_checker_input = self.feature_extractor(
                self.numpy_to_pil(image), return_tensors="np"
            ).pixel_values.astype(image.dtype)

            images, has_nsfw_concept = [], []
            for i in range(image.shape[0]):
                image_i, has_nsfw_concept_i = self.safety_checker(
                    clip_input=safety_checker_input[i : i + 1], images=image[i : i + 1]
                )
                images.append(image_i)
                has_nsfw_concept.append(has_nsfw_concept_i[0])
            image = np.concatenate(images)
        else:
            has_nsfw_concept = None

        if output_type == "pil":
            image = self.numpy_to_pil(image)

        if not return_dict:
            return (image, has_nsfw_concept)

        return StableDiffusionPipelineOutput(images=image, nsfw_content_detected=has_nsfw_concept)