Upload 9 files

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+figures/main_figure.jpg filter=lfs diff=lfs merge=lfs -text
+figures/sample_bunny_2K.png filter=lfs diff=lfs merge=lfs -text
+figures/sample_icecream_4K.png filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 yhyun225
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

pipeline_diffusehigh_sdxl.py

@@ -0,0 +1,798 @@
+import numpy as np
+import PIL
+import inspect
+import os
+from tqdm import tqdm
+from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+
+import torch
+import torch.nn.functional as F
+
+from transformers import (
+    CLIPImageProcessor,
+    CLIPTextModel,
+    CLIPTextModelWithProjection,
+    CLIPTokenizer,
+    CLIPVisionModelWithProjection,
+)
+
+from diffusers.schedulers import KarrasDiffusionSchedulers
+from diffusers.image_processor import PipelineImageInput
+from diffusers import (
+    AutoencoderKL,
+    UNet2DConditionModel,
+    StableDiffusionXLPipeline,
+    DDIMScheduler,
+    EulerDiscreteScheduler,
+)
+from diffusers.utils import BaseOutput
+from diffusers.utils.torch_utils import randn_tensor
+from pytorch_wavelets import DWTForward, DWTInverse
+from torchvision.transforms import GaussianBlur
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.rescale_noise_cfg
+def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
+    """
+    Rescale `noise_cfg` according to `guidance_rescale`. Based on findings of [Common Diffusion Noise Schedules and
+    Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf). See Section 3.4
+    """
+    std_text = noise_pred_text.std(dim=list(range(1, noise_pred_text.ndim)), keepdim=True)
+    std_cfg = noise_cfg.std(dim=list(range(1, noise_cfg.ndim)), keepdim=True)
+    # rescale the results from guidance (fixes overexposure)
+    noise_pred_rescaled = noise_cfg * (std_text / std_cfg)
+    # mix with the original results from guidance by factor guidance_rescale to avoid "plain looking" images
+    noise_cfg = guidance_rescale * noise_pred_rescaled + (1 - guidance_rescale) * noise_cfg
+    return noise_cfg
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.retrieve_timesteps
+def retrieve_timesteps(
+    scheduler,
+    num_inference_steps: Optional[int] = None,
+    device: Optional[Union[str, torch.device]] = None,
+    timesteps: Optional[List[int]] = None,
+    **kwargs,
+):
+    """
+    Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles
+    custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`.
+
+    Args:
+        scheduler (`SchedulerMixin`):
+            The scheduler to get timesteps from.
+        num_inference_steps (`int`):
+            The number of diffusion steps used when generating samples with a pre-trained model. If used,
+            `timesteps` must be `None`.
+        device (`str` or `torch.device`, *optional*):
+            The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
+        timesteps (`List[int]`, *optional*):
+                Custom timesteps used to support arbitrary spacing between timesteps. If `None`, then the default
+                timestep spacing strategy of the scheduler is used. If `timesteps` is passed, `num_inference_steps`
+                must be `None`.
+
+    Returns:
+        `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
+        second element is the number of inference steps.
+    """
+    if timesteps is not None:
+        accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accepts_timesteps:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" timestep schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
+    else:
+        scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+    return timesteps, num_inference_steps
+
+
+def gaussian_blur_image_sharpening(image, kernel_size=3, sigma=(0.1, 2.0), alpha=1):
+    gaussian_blur = GaussianBlur(kernel_size=kernel_size, sigma=sigma)
+    image_blurred = gaussian_blur(image)
+    image_sharpened = (alpha + 1) * image - alpha * image_blurred
+
+    return image_sharpened
+
+
+class DiffuseHighSDXLPipelineOutput(BaseOutput):
+    """
+    Output class for Stable Diffusion pipelines.
+
+    Args:
+        images (`List[PIL.Image.Image]` or `np.ndarray`)
+            List of denoised PIL images of length `batch_size` or numpy array of shape `(batch_size, height, width,
+            num_channels)`. PIL images or numpy array present the denoised images of the diffusion pipeline.
+    """
+
+    images: Union[List[PIL.Image.Image], np.ndarray]
+    guidance_images: Union[List[PIL.Image.Image], np.ndarray]
+
+
+class DiffuseHighSDXLPipeline(StableDiffusionXLPipeline):
+    def __init__(
+        self,
+        vae: AutoencoderKL,
+        text_encoder: CLIPTextModel,
+        text_encoder_2: CLIPTextModelWithProjection,
+        tokenizer: CLIPTokenizer,
+        tokenizer_2: CLIPTokenizer,
+        unet: UNet2DConditionModel,
+        scheduler: KarrasDiffusionSchedulers,
+        image_encoder: CLIPVisionModelWithProjection = None,
+        feature_extractor: CLIPImageProcessor = None,
+        force_zeros_for_empty_prompt: bool = True,
+        add_watermarker: Optional[bool] = None,
+    ):
+        super().__init__(
+            vae=vae,
+            text_encoder=text_encoder,
+            text_encoder_2=text_encoder_2,
+            tokenizer=tokenizer,
+            tokenizer_2=tokenizer_2,
+            unet=unet,
+            scheduler=scheduler,
+            image_encoder=image_encoder,
+            feature_extractor=feature_extractor,
+            force_zeros_for_empty_prompt=force_zeros_for_empty_prompt,
+            add_watermarker=add_watermarker
+        )
+
+    def _encode_vae_image(
+        self,
+        image: torch.Tensor,
+        normalize: bool = True,
+    ):
+        if normalize:
+            image = image * 2 - 1
+
+        needs_upcasting = self.vae.dtype == torch.float16 and self.vae.config.force_upcast
+
+        if needs_upcasting:
+            self.upcast_vae()
+
+        image = image.to(self.device)
+        latents = self.vae.encode(image).latent_dist.mode() * self.vae.config.scaling_factor
+
+        if needs_upcasting:
+            self.vae.to(dtype=torch.float16)
+
+        return latents.to(self.dtype)
+
+    def _decode_vae_latent(
+        self,
+        latents: torch.Tensor,
+        output_type: Optional[str] = 'pt',
+    ):
+        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)
+
+        latents = latents.to(self.device)
+        image = self.vae.decode(latents / self.vae.config.scaling_factor, return_dict=False)[0]
+        image = self.image_processor.postprocess(image, output_type=output_type)
+
+        if needs_upcasting:
+            self.vae.to(dtype=torch.float16)
+
+        return image
+
+    def edm_scheduler_step(
+        self,
+        model_output: torch.FloatTensor,
+        timestep: Union[float, torch.FloatTensor],
+        sample: torch.FloatTensor,
+        s_churn: float = 0.0,
+        s_tmin: float = 0.0,
+        s_tmax: float = 0.0,
+        s_noise: float = 1.0,
+        LL_guidance: Optional[torch.FloatTensor] = None,
+        generator: Optional[torch.Generator] = None,
+        return_pred_original_sample: bool = False,
+    ):
+        assert isinstance(self.scheduler, EulerDiscreteScheduler)
+        config = self.scheduler.config
+
+        if self.scheduler.step_index is None:
+            self.scheduler._init_step_index(timestep)
+
+        step_index = self.scheduler.step_index
+
+        sigma = self.scheduler.sigmas[step_index]
+
+        gamma = min(s_churn / (len(self.scheduler.sigmas) - 1), 2**0.5 - 1) if s_tmin <= sigma <= s_tmax else 0.0
+
+        noise = randn_tensor(
+            model_output.shape, dtype=model_output.dtype, device=model_output.device, generator=generator
+        )
+
+        eps = noise * s_noise
+        sigma_hat = sigma * (gamma + 1)
+
+        if gamma > 0:
+            sample = sample + eps * (sigma_hat**2 - sigma**2) ** 0.5
+
+        # 1. compute predicted original sample (x_0) from sigma-scaled predicted noise
+        if config.prediction_type == "original_sample" or config.prediction_type == "sample":
+            pred_original_sample = model_output
+        elif config.prediction_type == "epsilon":
+            pred_original_sample = sample - sigma_hat * model_output
+        elif config.prediction_type == "v_prediction":
+            # denoised = model_output * c_out + input * c_skip
+            pred_original_sample = model_output * (-sigma / (sigma**2 + 1) ** 0.5) + (sample / (sigma**2 + 1))
+        else:
+            raise ValueError(
+                f"prediction_type given as {config.prediction_type} must be one of `epsilon`, or `v_prediction`"
+            )
+
+        # 2. If gudiance LL component is given, perform structural guidance
+        if LL_guidance is not None:
+            pred_original_image = self._decode_vae_latent(pred_original_sample, output_type='pt')
+
+            _, HH = self.DWT(pred_original_image)
+            coeffs = (LL_guidance, HH)
+            pred_original_image = self.iDWT(coeffs)
+
+            pred_original_sample = self._encode_vae_image(pred_original_image)
+
+        # 3. Convert to an ODE derivative
+        derivative = (sample - pred_original_sample) / sigma_hat
+
+        dt = self.scheduler.sigmas[self.scheduler.step_index + 1] - sigma_hat
+
+        prev_sample = sample + derivative * dt
+
+        self.scheduler._step_index += 1
+
+        if return_pred_original_sample:
+            return (prev_sample, pred_original_sample)
+
+        return (prev_sample, )
+
+
+    @torch.no_grad()
+    def __call__(
+        self,
+        prompt: Union[str, List[str]] = None,
+        prompt_2: Optional[Union[str, List[str]]] = None,
+        num_inference_steps: int = 50,
+        timesteps: List[int] = None,
+        denoising_end: Optional[float] = None,
+        guidance_scale: float = 5,
+        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,
+        output_type: Optional[str] = "pil",
+        return_dict: bool = True,
+        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+        guidance_rescale: float = 0.0,
+        original_size: Optional[Tuple[int, int]] = None,
+        crops_coords_top_left: Tuple[int, int] = (0, 0),
+        target_size: Optional[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: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
+        callback_steps: Optional[int] = 1,
+        callback_on_step_end: Optional[Callable[[int, int, Dict], None]] = None,
+        callback_on_step_end_tensor_inputs: List[str] = ["latents"],
+        ### DiffuseHigh parameters ###
+        target_height: Union[int, List[int]] = [2048, 3072, 4096],
+        target_width: Union[int, List[int]] = [2048, 3072, 4096],
+        guidance_image: Optional[Union[torch.FloatTensor, PIL.Image.Image, np.ndarray]] = None,
+        noising_steps: int = 15,
+        diffusehigh_guidance_scale: float = 10.0,
+        # >>> DWT parameters
+        enable_dwt: bool = True,
+        dwt_level: Optional[int] = 1,
+        dwt_wave: Optional[str] = "db4",
+        dwt_mode: Optional[str] = "symmetric",
+        dwt_steps: Optional[int] = 5,
+        # >>> Sharpening parameters
+        enable_sharpening: bool = True,
+        sharpening_kernel_size: int = 3,
+        sharpening_sigma: Optional[Union[Tuple[float, float], float]] = (0.1, 2.0),
+        sharpening_alpha: float = 1.0,
+        **kwargs,
+    ):
+        r"""
+        Function invoked when calling the pipeline for generation.
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`.
+                instead.
+            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
+            height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor):
+                The height in pixels of the generated image. This is set to 1024 by default for the best results.
+                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. This is set to 1024 by default for the best results.
+                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.
+            timesteps (`List[int]`, *optional*):
+                Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument
+                in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
+                passed will be used. Must be in descending order.
+            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):
+                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
+                `guidance_scale` is defined as `w` of equation 2. of [Imagen
+                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
+                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
+                usually at the expense of lower image quality.
+            negative_prompt (`str` or `List[str]`, *optional*):
+                The prompt or prompts not to guide the image generation. If not defined, one has to pass
+                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
+                less than `1`).
+            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
+            num_images_per_prompt (`int`, *optional*, defaults to 1):
+                The number of images to generate per prompt.
+            eta (`float`, *optional*, defaults to 0.0):
+                Corresponds to parameter eta (η) in the DDIM paper: https://arxiv.org/abs/2010.02502. Only applies to
+                [`schedulers.DDIMScheduler`], will be ignored for others.
+            generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
+                One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html)
+                to make generation deterministic.
+            latents (`torch.FloatTensor`, *optional*):
+                Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image
+                generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
+                tensor will ge generated by sampling using the supplied random `generator`.
+            prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
+                provided, text embeddings will be generated from `prompt` input argument.
+            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
+                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
+                argument.
+            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.
+            ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
+            output_type (`str`, *optional*, defaults to `"pil"`):
+                The output format of the generate image. Choose between
+                [PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`.
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~pipelines.stable_diffusion_xl.StableDiffusionXLPipelineOutput`] instead
+                of a plain tuple.
+            cross_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
+                `self.processor` in
+                [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+            guidance_rescale (`float`, *optional*, defaults to 0.0):
+                Guidance rescale factor proposed by [Common Diffusion Noise Schedules and Sample Steps are
+                Flawed](https://arxiv.org/pdf/2305.08891.pdf) `guidance_scale` is defined as `φ` in equation 16. of
+                [Common Diffusion Noise Schedules and Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf).
+                Guidance rescale factor should fix overexposure when using zero terminal SNR.
+            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.
+            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 pipeline class.
+            target_height ('List[int]' or int):
+                The height of the image being generated. If list is given, the pipeline generates corresponding intermediate
+                resolution images in a progressive manner.
+            target_width ('List[int]' or int):
+                The width of the image being generated. If list is given, the pipeline generates corresponding intermediate
+                resolution images in a progressive manner.
+
+
+
+        Examples:
+
+        Returns:
+            [`DiffuseHighSDXLPipelineOutput`] or `tuple`:
+            [`DiffuseHighSDXLPipelineOutput`] if `return_dict` is True, otherwise a
+            `tuple`. When returning a tuple, the first element is a list with the generated images.
+        """
+        # 0. Default height and width to unet
+        height = self.default_sample_size * self.vae_scale_factor
+        width = self.default_sample_size * self.vae_scale_factor
+
+        original_size = original_size or (height, width)
+        target_size = target_size or (height, width)
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(
+            prompt,
+            prompt_2,
+            height,
+            width,
+            callback_steps,
+            negative_prompt,
+            negative_prompt_2,
+            prompt_embeds,
+            negative_prompt_embeds,
+            pooled_prompt_embeds,
+            negative_pooled_prompt_embeds,
+            callback_on_step_end_tensor_inputs,
+        )
+
+        self._guidance_scale = guidance_scale
+        self._guidance_rescale = guidance_rescale
+        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
+
+        # 3. Encode input prompt
+        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,
+            prompt_2=prompt_2,
+            device=device,
+            num_images_per_prompt=num_images_per_prompt,
+            do_classifier_free_guidance=self.do_classifier_free_guidance,
+            negative_prompt=negative_prompt,
+            negative_prompt_2=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=lora_scale,
+            clip_skip=self.clip_skip,
+        )
+
+        # 4. Prepare timesteps
+        timesteps, num_inference_steps = retrieve_timesteps(self.scheduler, num_inference_steps, device, timesteps)
+
+        # 5. Prepare latent variables
+        num_channels_latents = self.unet.config.in_channels
+        latents = self.prepare_latents(
+            batch_size * num_images_per_prompt,
+            num_channels_latents,
+            height,
+            width,
+            prompt_embeds.dtype,
+            device,
+            generator,
+            latents,
+        )
+
+        # 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+        extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+
+        # 7. Prepare added time ids & embeddings
+        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)
+
+        if ip_adapter_image is not None:
+            image_embeds, negative_image_embeds = self.encode_image(ip_adapter_image, device, num_images_per_prompt)
+            if self.do_classifier_free_guidance:
+                image_embeds = torch.cat([negative_image_embeds, image_embeds])
+                image_embeds = image_embeds.to(device)
+
+        # 8. Denoising loop
+        num_warmup_steps = max(len(timesteps) - num_inference_steps * self.scheduler.order, 0)
+
+        # 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]
+
+        # 9. 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)
+
+        # 10. Obtain clean image for structral guidance (can be given by user or generated)
+        if guidance_image is None:
+            self._num_timesteps = len(timesteps)
+            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 = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents
+
+                    latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+                    # predict the noise residual
+                    added_cond_kwargs = {"text_embeds": add_text_embeds, "time_ids": add_time_ids}
+                    if ip_adapter_image is not None:
+                        added_cond_kwargs["image_embeds"] = image_embeds
+                    noise_pred = self.unet(
+                        latent_model_input,
+                        t,
+                        encoder_hidden_states=prompt_embeds,
+                        timestep_cond=timestep_cond,
+                        cross_attention_kwargs=self.cross_attention_kwargs,
+                        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 + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                    if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
+                        # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+                        noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale)
+
+                    # 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)
+                        add_text_embeds = callback_outputs.pop("add_text_embeds", add_text_embeds)
+                        negative_pooled_prompt_embeds = callback_outputs.pop(
+                            "negative_pooled_prompt_embeds", negative_pooled_prompt_embeds
+                        )
+                        add_time_ids = callback_outputs.pop("add_time_ids", add_time_ids)
+                        negative_add_time_ids = callback_outputs.pop("negative_add_time_ids", negative_add_time_ids)
+
+                    # 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)
+
+            image = self._decode_vae_latent(latents, output_type='pt')
+        else:
+            image = self.image_processor.preprocess(guidance_image, height, width)
+            if self.image_processor.config.do_normalize:
+               image = (image + 1.) * 0.5
+
+            image = image.to(self.device)
+
+        original_guidance_image = image
+
+        # |-------------------------------- DiffuseHigh process --------------------------------|
+        # DWT & inverse DWT works on torch.float32
+        if enable_dwt:
+            self.DWT = DWTForward(J=dwt_level, wave=dwt_wave, mode=dwt_mode).to(self.device)
+            self.iDWT = DWTInverse(wave=dwt_wave, mode=dwt_mode).to(self.device)
+
+        # 11. Prepare progressive DiffuseHigh pipeline
+        self.scheduler.set_timesteps(num_inference_steps)
+        diffusehigh_timesteps = self.scheduler.timesteps[-noising_steps:]
+        self.enable_vae_tiling() # Vae tiling mode in order to prevent OOM issues
+
+        if isinstance(target_width, int):
+            target_width = [target_width]
+        if isinstance(target_height, int):
+            target_height = [target_height]
+
+        assert len(target_width) == len(target_height)
+
+        #12. Progressive DiffuseHigh Pipeline
+        for h, w in zip(target_height, target_width):
+            # interpolate the image to the desired resolution
+            guidance_image = F.interpolate(image, (h, w), mode="bicubic", align_corners=False)
+
+            # apply sharpening operation to the image
+            if enable_sharpening:
+                guidance_image = gaussian_blur_image_sharpening(
+                    guidance_image,
+                    kernel_size=sharpening_kernel_size,
+                    sigma=sharpening_sigma,
+                    alpha=sharpening_alpha,
+                )
+
+            # extract low-frequency component (structural guidance) from the guidance image
+            if enable_dwt:
+                LL, _ = self.DWT(guidance_image)
+
+            # obtain latent of the interpolated image and noise it
+            latents = self._encode_vae_image(guidance_image)
+            noise = randn_tensor(latents.shape, generator, device=latents.device, dtype=latents.dtype)
+            latents = self.scheduler.add_noise(latents, noise, diffusehigh_timesteps[None, 0])
+
+            for i, t in tqdm(enumerate(diffusehigh_timesteps), total=diffusehigh_timesteps.shape[0]):
+                # 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)
+
+                # predict the noise residual
+                added_cond_kwargs = {"text_embeds": add_text_embeds, "time_ids": add_time_ids}
+
+                noise_pred = self.unet(
+                    latent_model_input,
+                    t,
+                    encoder_hidden_states=prompt_embeds,
+                    timestep_cond=timestep_cond,
+                    cross_attention_kwargs=self.cross_attention_kwargs, # None
+                    added_cond_kwargs=added_cond_kwargs, # None
+                    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 + diffusehigh_guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+                # EDM sampler step
+                latents = self.edm_scheduler_step(
+                    noise_pred,
+                    t,
+                    latents,
+                    **extra_step_kwargs,
+                    LL_guidance=LL if (enable_dwt and i < dwt_steps) else None,
+                )[0]
+
+            image = self._decode_vae_latent(latents)
+
+            if isinstance(self.scheduler, EulerDiscreteScheduler):
+                self.scheduler._step_index = None
+
+        # Offload all models
+        self.maybe_free_model_hooks()
+
+        if output_type != 'pt':
+            image = self.image_processor.postprocess(image * 2 - 1, output_type=output_type)
+            guidance_image = self.image_processor.postprocess(original_guidance_image * 2 -1 , output_type=output_type)
+
+        if not return_dict:
+            return (image, guidance_image)
+
+        return DiffuseHighSDXLPipelineOutput(images=image, guidance_image=guidance_image)
+
+
+def set_seeds(seed):
+    os.environ["PYTHONHASHSEED"] = str(seed)
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = True
+
+# DEBUGGING
+if __name__ == "__main__":
+    set_seeds(23)
+
+    model = DiffuseHighSDXLPipeline.from_pretrained(
+        "stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16, #scheduler=scheduler
+    ).to("cuda")
+
+    prompt = "Cinematic photo of delicious chocolate icecream."
+
+    negative_prompt = "blurry, ugly, duplicate, poorly drawn, deformed, mosaic"
+
+    image = model(
+        prompt,
+        negative_prompt=negative_prompt,
+        target_height=[2048, 3072, 4096],
+        target_width=[2048, 3072, 4096],
+        enable_dwt=True,
+        dwt_steps=5,
+        enable_sharpening=True,
+        sharpness_factor=1.0,
+    ).images[0]
+
+    image.save("sample.png")

requirements.txt

@@ -0,0 +1,5 @@
+diffusers==0.24.0
+accelerate
+transformers
+pywavelets
+pytorch-wavelets

utils/utils.py

@@ -0,0 +1,11 @@
+import os
+import torch
+import numpy as np
+
+def set_seeds(seed):
+    os.environ["PYTHONHASHSEED"] = str(seed)
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = True