pipeline_t2v_base_pixel.py

# Copyright 2023 Show Labs, Alibaba DAMO-VILAB, and The HuggingFace Team. All rights reserved.
# Copyright 2023 The ModelScope Team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import html
import inspect
import re
import urllib.parse as ul
from dataclasses import dataclass
from typing import Any, Callable, Dict, List, Optional, Union

import numpy as np
import torch
import torch.utils.checkpoint
from transformers import CLIPImageProcessor, T5EncoderModel, T5Tokenizer

from diffusers import UNet3DConditionModel
from diffusers.loaders import LoraLoaderMixin
from diffusers.pipelines.pipeline_utils import DiffusionPipeline
from diffusers.schedulers import DDPMScheduler
from diffusers.utils import (
    BACKENDS_MAPPING,
    BaseOutput,
    is_accelerate_available,
    is_accelerate_version,
    is_bs4_available,
    is_ftfy_available,
    logging,
)
from diffusers.utils.torch_utils import randn_tensor


logger = logging.get_logger(__name__)  # pylint: disable=invalid-name

if is_bs4_available():
    from bs4 import BeautifulSoup

if is_ftfy_available():
    import ftfy


@dataclass
class TextToVideoPipelineOutput(BaseOutput):
    """
    Output class for text to video pipelines.

    Args:
        frames (`List[np.ndarray]` or `torch.FloatTensor`)
            List of denoised frames (essentially images) as NumPy arrays of shape `(height, width, num_channels)` or as
            a `torch` tensor. NumPy array present the denoised images of the diffusion pipeline. The length of the list
            denotes the video length i.e., the number of frames.
    """

    frames: Union[List[np.ndarray], torch.FloatTensor]


def tensor2vid(video: torch.Tensor, mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]) -> List[np.ndarray]:
    # This code is copied from https://github.com/modelscope/modelscope/blob/1509fdb973e5871f37148a4b5e5964cafd43e64d/modelscope/pipelines/multi_modal/text_to_video_synthesis_pipeline.py#L78
    # reshape to ncfhw
    mean = torch.tensor(mean, device=video.device).reshape(1, -1, 1, 1, 1)
    std = torch.tensor(std, device=video.device).reshape(1, -1, 1, 1, 1)
    # unnormalize back to [0,1]
    video = video.mul_(std).add_(mean)
    video.clamp_(0, 1)
    # prepare the final outputs
    i, c, f, h, w = video.shape
    images = video.permute(2, 3, 0, 4, 1).reshape(
        f, h, i * w, c
    )  # 1st (frames, h, batch_size, w, c) 2nd (frames, h, batch_size * w, c)
    images = images.unbind(dim=0)  # prepare a list of indvidual (consecutive frames)
    images = [(image.cpu().numpy() * 255).astype("uint8") for image in images]  # f h w c
    return images


class TextToVideoIFPipeline(DiffusionPipeline, LoraLoaderMixin):
    tokenizer: T5Tokenizer
    text_encoder: T5EncoderModel

    unet: UNet3DConditionModel
    scheduler: DDPMScheduler

    feature_extractor: Optional[CLIPImageProcessor]
    # safety_checker: Optional[IFSafetyChecker]

    # watermarker: Optional[IFWatermarker]

    bad_punct_regex = re.compile(
        r"[" + "#®•©™&@·º½¾¿¡§~" + "\)" + "\(" + "\]" + "\[" + "\}" + "\{" + "\|" + "\\" + "\/" + "\*" + r"]{1,}"
    )  # noqa

    _optional_components = [
        "tokenizer",
        "text_encoder",
        "safety_checker",
        "feature_extractor",
        "watermarker",
    ]

    def __init__(
        self,
        tokenizer: T5Tokenizer,
        text_encoder: T5EncoderModel,
        unet: UNet3DConditionModel,
        scheduler: DDPMScheduler,
        feature_extractor: Optional[CLIPImageProcessor],
    ):
        super().__init__()

        self.register_modules(
            tokenizer=tokenizer,
            text_encoder=text_encoder,
            unet=unet,
            scheduler=scheduler,
            feature_extractor=feature_extractor,
        )
        self.safety_checker = None

    def enable_sequential_cpu_offload(self, gpu_id=0):
        r"""
        Offloads all models to CPU using accelerate, significantly reducing memory usage. When called, the pipeline's
        models have their state dicts saved to CPU and then are moved to a `torch.device('meta') and loaded to GPU only
        when their specific submodule has its `forward` method called.
        """
        if is_accelerate_available():
            from accelerate import cpu_offload
        else:
            raise ImportError("Please install accelerate via `pip install accelerate`")

        device = torch.device(f"cuda:{gpu_id}")

        models = [
            self.text_encoder,
            self.unet,
        ]
        for cpu_offloaded_model in models:
            if cpu_offloaded_model is not None:
                cpu_offload(cpu_offloaded_model, device)

        if self.safety_checker is not None:
            cpu_offload(self.safety_checker, execution_device=device, offload_buffers=True)

    def enable_model_cpu_offload(self, gpu_id=0):
        r"""
        Offloads all models to CPU using accelerate, reducing memory usage with a low impact on performance. Compared
        to `enable_sequential_cpu_offload`, this method moves one whole model at a time to the GPU when its `forward`
        method is called, and the model remains in GPU until the next model runs. Memory savings are lower than with
        `enable_sequential_cpu_offload`, but performance is much better due to the iterative execution of the `unet`.
        """
        if is_accelerate_available() and is_accelerate_version(">=", "0.17.0.dev0"):
            from accelerate import cpu_offload_with_hook
        else:
            raise ImportError("`enable_model_cpu_offload` requires `accelerate v0.17.0` or higher.")

        device = torch.device(f"cuda:{gpu_id}")

        self.unet.train()

        if self.device.type != "cpu":
            self.to("cpu", silence_dtype_warnings=True)
            torch.cuda.empty_cache()  # otherwise we don't see the memory savings (but they probably exist)

        hook = None

        if self.text_encoder is not None:
            _, hook = cpu_offload_with_hook(self.text_encoder, device, prev_module_hook=hook)

            # Accelerate will move the next model to the device _before_ calling the offload hook of the
            # previous model. This will cause both models to be present on the device at the same time.
            # IF uses T5 for its text encoder which is really large. We can manually call the offload
            # hook for the text encoder to ensure it's moved to the cpu before the unet is moved to
            # the GPU.
            self.text_encoder_offload_hook = hook

        _, hook = cpu_offload_with_hook(self.unet, device, prev_module_hook=hook)

        # if the safety checker isn't called, `unet_offload_hook` will have to be called to manually offload the unet
        self.unet_offload_hook = hook

        if self.safety_checker is not None:
            _, hook = cpu_offload_with_hook(self.safety_checker, device, prev_module_hook=hook)

        # We'll offload the last model manually.
        self.final_offload_hook = hook

    def remove_all_hooks(self):
        if is_accelerate_available():
            from accelerate.hooks import remove_hook_from_module
        else:
            raise ImportError("Please install accelerate via `pip install accelerate`")

        for model in [self.text_encoder, self.unet, self.safety_checker]:
            if model is not None:
                remove_hook_from_module(model, recurse=True)

        self.unet_offload_hook = None
        self.text_encoder_offload_hook = None
        self.final_offload_hook = None

    @property
    # Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.StableDiffusionPipeline._execution_device
    def _execution_device(self):
        r"""
        Returns the device on which the pipeline's models will be executed. After calling
        `pipeline.enable_sequential_cpu_offload()` the execution device can only be inferred from Accelerate's module
        hooks.
        """
        if not hasattr(self.unet, "_hf_hook"):
            return self.device
        for module in self.unet.modules():
            if (
                hasattr(module, "_hf_hook")
                and hasattr(module._hf_hook, "execution_device")
                and module._hf_hook.execution_device is not None
            ):
                return torch.device(module._hf_hook.execution_device)
        return self.device

    @torch.no_grad()
    def encode_prompt(
        self,
        prompt,
        do_classifier_free_guidance=True,
        num_images_per_prompt=1,
        device=None,
        negative_prompt=None,
        prompt_embeds: Optional[torch.FloatTensor] = None,
        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
        clean_caption: bool = False,
    ):
        r"""
        Encodes the prompt into text encoder hidden states.

        Args:
             prompt (`str` or `List[str]`, *optional*):
                prompt to be encoded
            device: (`torch.device`, *optional*):
                torch device to place the resulting embeddings on
            num_images_per_prompt (`int`, *optional*, defaults to 1):
                number of images that should be generated per prompt
            do_classifier_free_guidance (`bool`, *optional*, defaults to `True`):
                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. If not defined, one has to pass `negative_prompt_embeds`. instead.
                Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`).
            prompt_embeds (`torch.FloatTensor`, *optional*):
                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                provided, text embeddings will be generated from `prompt` input argument.
            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                argument.
        """
        if prompt is not None and negative_prompt is not None:
            if 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)}."
                )

        if device is None:
            device = self._execution_device

        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]

        # while T5 can handle much longer input sequences than 77, the text encoder was trained with a max length of 77 for IF
        max_length = 77

        if prompt_embeds is None:
            prompt = self._text_preprocessing(prompt, clean_caption=clean_caption)
            text_inputs = self.tokenizer(
                prompt,
                padding="max_length",
                max_length=max_length,
                truncation=True,
                add_special_tokens=True,
                return_tensors="pt",
            )
            text_input_ids = text_inputs.input_ids
            untruncated_ids = self.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 = self.tokenizer.batch_decode(untruncated_ids[:, max_length - 1 : -1])
                logger.warning(
                    "The following part of your input was truncated because CLIP can only handle sequences up to"
                    f" {max_length} tokens: {removed_text}"
                )

            attention_mask = text_inputs.attention_mask.to(device)

            prompt_embeds = self.text_encoder(
                text_input_ids.to(device),
                attention_mask=attention_mask,
            )
            prompt_embeds = prompt_embeds[0]

        if self.text_encoder is not None:
            dtype = self.text_encoder.dtype
        elif self.unet is not None:
            dtype = self.unet.dtype
        else:
            dtype = None

        prompt_embeds = prompt_embeds.to(dtype=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)

        # get unconditional embeddings for classifier free guidance
        if do_classifier_free_guidance and negative_prompt_embeds is None:
            uncond_tokens: List[str]
            if negative_prompt is None:
                uncond_tokens = [""] * batch_size
            elif isinstance(negative_prompt, str):
                uncond_tokens = [negative_prompt]
            elif batch_size != len(negative_prompt):
                raise ValueError(
                    f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
                    f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
                    " the batch size of `prompt`."
                )
            else:
                uncond_tokens = negative_prompt

            uncond_tokens = self._text_preprocessing(uncond_tokens, clean_caption=clean_caption)
            max_length = prompt_embeds.shape[1]
            uncond_input = self.tokenizer(
                uncond_tokens,
                padding="max_length",
                max_length=max_length,
                truncation=True,
                return_attention_mask=True,
                add_special_tokens=True,
                return_tensors="pt",
            )
            attention_mask = uncond_input.attention_mask.to(device)

            negative_prompt_embeds = self.text_encoder(
                uncond_input.input_ids.to(device),
                attention_mask=attention_mask,
            )
            negative_prompt_embeds = negative_prompt_embeds[0]

        if do_classifier_free_guidance:
            # duplicate unconditional embeddings for each generation per prompt, using mps friendly method
            seq_len = negative_prompt_embeds.shape[1]

            negative_prompt_embeds = negative_prompt_embeds.to(dtype=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)

            # 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
        else:
            negative_prompt_embeds = None

        return prompt_embeds, negative_prompt_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,
        callback_steps,
        negative_prompt=None,
        prompt_embeds=None,
        negative_prompt_embeds=None,
    ):
        if (callback_steps is None) or (
            callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0)
        ):
            raise ValueError(
                f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
                f" {type(callback_steps)}."
            )

        if prompt is not None and prompt_embeds is not None:
            raise ValueError(
                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
                " only forward one of the two."
            )
        elif prompt is None and prompt_embeds is None:
            raise ValueError(
                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
            )
        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")

        if negative_prompt is not None and negative_prompt_embeds is not None:
            raise ValueError(
                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
            )

        if prompt_embeds is not None and negative_prompt_embeds is not None:
            if prompt_embeds.shape != negative_prompt_embeds.shape:
                raise ValueError(
                    "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
                    f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
                    f" {negative_prompt_embeds.shape}."
                )

    def prepare_intermediate_images(
        self,
        batch_size,
        num_channels,
        num_frames,
        height,
        width,
        dtype,
        device,
        generator,
    ):
        shape = (batch_size, num_channels, num_frames, height, width)
        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."
            )

        intermediate_images = randn_tensor(shape, generator=generator, device=device, dtype=dtype)

        # scale the initial noise by the standard deviation required by the scheduler
        intermediate_images = intermediate_images * self.scheduler.init_noise_sigma
        return intermediate_images

    def _text_preprocessing(self, text, clean_caption=False):
        if clean_caption and not is_bs4_available():
            logger.warn(BACKENDS_MAPPING["bs4"][-1].format("Setting `clean_caption=True`"))
            logger.warn("Setting `clean_caption` to False...")
            clean_caption = False

        if clean_caption and not is_ftfy_available():
            logger.warn(BACKENDS_MAPPING["ftfy"][-1].format("Setting `clean_caption=True`"))
            logger.warn("Setting `clean_caption` to False...")
            clean_caption = False

        if not isinstance(text, (tuple, list)):
            text = [text]

        def process(text: str):
            if clean_caption:
                text = self._clean_caption(text)
                text = self._clean_caption(text)
            else:
                text = text.lower().strip()
            return text

        return [process(t) for t in text]

    def _clean_caption(self, caption):
        caption = str(caption)
        caption = ul.unquote_plus(caption)
        caption = caption.strip().lower()
        caption = re.sub("<person>", "person", caption)
        # urls:
        caption = re.sub(
            r"\b((?:https?:(?:\/{1,3}|[a-zA-Z0-9%])|[a-zA-Z0-9.\-]+[.](?:com|co|ru|net|org|edu|gov|it)[\w/-]*\b\/?(?!@)))",  # noqa
            "",
            caption,
        )  # regex for urls
        caption = re.sub(
            r"\b((?:www:(?:\/{1,3}|[a-zA-Z0-9%])|[a-zA-Z0-9.\-]+[.](?:com|co|ru|net|org|edu|gov|it)[\w/-]*\b\/?(?!@)))",  # noqa
            "",
            caption,
        )  # regex for urls
        # html:
        caption = BeautifulSoup(caption, features="html.parser").text

        # @<nickname>
        caption = re.sub(r"@[\w\d]+\b", "", caption)

        # 31C0—31EF CJK Strokes
        # 31F0—31FF Katakana Phonetic Extensions
        # 3200—32FF Enclosed CJK Letters and Months
        # 3300—33FF CJK Compatibility
        # 3400—4DBF CJK Unified Ideographs Extension A
        # 4DC0—4DFF Yijing Hexagram Symbols
        # 4E00—9FFF CJK Unified Ideographs
        caption = re.sub(r"[\u31c0-\u31ef]+", "", caption)
        caption = re.sub(r"[\u31f0-\u31ff]+", "", caption)
        caption = re.sub(r"[\u3200-\u32ff]+", "", caption)
        caption = re.sub(r"[\u3300-\u33ff]+", "", caption)
        caption = re.sub(r"[\u3400-\u4dbf]+", "", caption)
        caption = re.sub(r"[\u4dc0-\u4dff]+", "", caption)
        caption = re.sub(r"[\u4e00-\u9fff]+", "", caption)
        #######################################################

        # все виды тире / all types of dash --> "-"
        caption = re.sub(
            r"[\u002D\u058A\u05BE\u1400\u1806\u2010-\u2015\u2E17\u2E1A\u2E3A\u2E3B\u2E40\u301C\u3030\u30A0\uFE31\uFE32\uFE58\uFE63\uFF0D]+",  # noqa
            "-",
            caption,
        )

        # кавычки к одному стандарту
        caption = re.sub(r"[`´«»“”¨]", '"', caption)
        caption = re.sub(r"[‘’]", "'", caption)

        # &quot;
        caption = re.sub(r"&quot;?", "", caption)
        # &amp
        caption = re.sub(r"&amp", "", caption)

        # ip adresses:
        caption = re.sub(r"\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}", " ", caption)

        # article ids:
        caption = re.sub(r"\d:\d\d\s+$", "", caption)

        # \n
        caption = re.sub(r"\\n", " ", caption)

        # "#123"
        caption = re.sub(r"#\d{1,3}\b", "", caption)
        # "#12345.."
        caption = re.sub(r"#\d{5,}\b", "", caption)
        # "123456.."
        caption = re.sub(r"\b\d{6,}\b", "", caption)
        # filenames:
        caption = re.sub(r"[\S]+\.(?:png|jpg|jpeg|bmp|webp|eps|pdf|apk|mp4)", "", caption)

        #
        caption = re.sub(r"[\"\']{2,}", r'"', caption)  # """AUSVERKAUFT"""
        caption = re.sub(r"[\.]{2,}", r" ", caption)  # """AUSVERKAUFT"""

        caption = re.sub(self.bad_punct_regex, r" ", caption)  # ***AUSVERKAUFT***, #AUSVERKAUFT
        caption = re.sub(r"\s+\.\s+", r" ", caption)  # " . "

        # this-is-my-cute-cat / this_is_my_cute_cat
        regex2 = re.compile(r"(?:\-|\_)")
        if len(re.findall(regex2, caption)) > 3:
            caption = re.sub(regex2, " ", caption)

        caption = ftfy.fix_text(caption)
        caption = html.unescape(html.unescape(caption))

        caption = re.sub(r"\b[a-zA-Z]{1,3}\d{3,15}\b", "", caption)  # jc6640
        caption = re.sub(r"\b[a-zA-Z]+\d+[a-zA-Z]+\b", "", caption)  # jc6640vc
        caption = re.sub(r"\b\d+[a-zA-Z]+\d+\b", "", caption)  # 6640vc231

        caption = re.sub(r"(worldwide\s+)?(free\s+)?shipping", "", caption)
        caption = re.sub(r"(free\s)?download(\sfree)?", "", caption)
        caption = re.sub(r"\bclick\b\s(?:for|on)\s\w+", "", caption)
        caption = re.sub(r"\b(?:png|jpg|jpeg|bmp|webp|eps|pdf|apk|mp4)(\simage[s]?)?", "", caption)
        caption = re.sub(r"\bpage\s+\d+\b", "", caption)

        caption = re.sub(r"\b\d*[a-zA-Z]+\d+[a-zA-Z]+\d+[a-zA-Z\d]*\b", r" ", caption)  # j2d1a2a...

        caption = re.sub(r"\b\d+\.?\d*[xх×]\d+\.?\d*\b", "", caption)

        caption = re.sub(r"\b\s+\:\s+", r": ", caption)
        caption = re.sub(r"(\D[,\./])\b", r"\1 ", caption)
        caption = re.sub(r"\s+", " ", caption)

        caption.strip()

        caption = re.sub(r"^[\"\']([\w\W]+)[\"\']$", r"\1", caption)
        caption = re.sub(r"^[\'\_,\-\:;]", r"", caption)
        caption = re.sub(r"[\'\_,\-\:\-\+]$", r"", caption)
        caption = re.sub(r"^\.\S+$", "", caption)

        return caption.strip()

    @torch.no_grad()
    def __call__(
        self,
        prompt: Union[str, List[str]] = None,
        num_inference_steps: int = 100,
        timesteps: List[int] = None,
        guidance_scale: float = 7.0,
        negative_prompt: Optional[Union[str, List[str]]] = None,
        num_images_per_prompt: Optional[int] = 1,
        height: Optional[int] = None,
        width: Optional[int] = None,
        num_frames: int = 16,
        eta: float = 0.0,
        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
        prompt_embeds: Optional[torch.FloatTensor] = None,
        negative_prompt_embeds: Optional[torch.FloatTensor] = None,
        output_type: Optional[str] = "np",
        return_dict: bool = True,
        callback: Optional[Callable[[int, int, torch.FloatTensor], None]] = None,
        callback_steps: int = 1,
        clean_caption: bool = True,
        cross_attention_kwargs: Optional[Dict[str, Any]] = None,
    ):
        """
        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.
            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. If not defined, equal spaced `num_inference_steps`
                timesteps are used. Must be in descending order.
            guidance_scale (`float`, *optional*, defaults to 7.5):
                Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://arxiv.org/abs/2207.12598).
                `guidance_scale` is defined as `w` of equation 2. of [Imagen
                Paper](https://arxiv.org/pdf/2205.11487.pdf). Guidance scale is enabled by setting `guidance_scale >
                1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`,
                usually at the expense of lower image quality.
            negative_prompt (`str` or `List[str]`, *optional*):
                The prompt or prompts not to guide the image generation. If not defined, one has to pass
                `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
                less than `1`).
            num_images_per_prompt (`int`, *optional*, defaults to 1):
                The number of images to generate per prompt.
            height (`int`, *optional*, defaults to self.unet.config.sample_size):
                The height in pixels of the generated image.
            width (`int`, *optional*, defaults to self.unet.config.sample_size):
                The width in pixels of the generated image.
            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.
            prompt_embeds (`torch.FloatTensor`, *optional*):
                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
                provided, text embeddings will be generated from `prompt` input argument.
            negative_prompt_embeds (`torch.FloatTensor`, *optional*):
                Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
                weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
                argument.
            output_type (`str`, *optional*, defaults to `"pil"`):
                The output format of the generate image. Choose between
                [PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`.
            return_dict (`bool`, *optional*, defaults to `True`):
                Whether or not to return a [`~pipelines.stable_diffusion.IFPipelineOutput`] instead of a plain tuple.
            callback (`Callable`, *optional*):
                A function that will be called every `callback_steps` steps during inference. The function will be
                called with the following arguments: `callback(step: int, timestep: int, latents: torch.FloatTensor)`.
            callback_steps (`int`, *optional*, defaults to 1):
                The frequency at which the `callback` function will be called. If not specified, the callback will be
                called at every step.
            clean_caption (`bool`, *optional*, defaults to `True`):
                Whether or not to clean the caption before creating embeddings. Requires `beautifulsoup4` and `ftfy` to
                be installed. If the dependencies are not installed, the embeddings will be created from the raw
                prompt.
            cross_attention_kwargs (`dict`, *optional*):
                A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
                `self.processor` in
                [diffusers.cross_attention](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/cross_attention.py).

        Examples:

        Returns:
            [`~pipelines.stable_diffusion.IFPipelineOutput`] or `tuple`:
            [`~pipelines.stable_diffusion.IFPipelineOutput`] if `return_dict` is True, otherwise a `tuple. When
            returning a tuple, the first element is a list with the generated images, and the second element is a list
            of `bool`s denoting whether the corresponding generated image likely represents "not-safe-for-work" (nsfw)
            or watermarked content, according to the `safety_checker`.
        """
        # 1. Check inputs. Raise error if not correct
        self.check_inputs(
            prompt,
            callback_steps,
            negative_prompt,
            prompt_embeds,
            negative_prompt_embeds,
        )

        # 2. Define call parameters
        height = height or self.unet.config.sample_size
        width = width or self.unet.config.sample_size

        if prompt is not None and isinstance(prompt, str):
            batch_size = 1
        elif prompt is not None and isinstance(prompt, list):
            batch_size = len(prompt)
        else:
            batch_size = prompt_embeds.shape[0]

        device = self._execution_device

        # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
        # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
        # corresponds to doing no classifier free guidance.
        do_classifier_free_guidance = guidance_scale > 1.0

        # 3. Encode input prompt
        prompt_embeds, negative_prompt_embeds = self.encode_prompt(
            prompt,
            do_classifier_free_guidance,
            num_images_per_prompt=num_images_per_prompt,
            device=device,
            negative_prompt=negative_prompt,
            prompt_embeds=prompt_embeds,
            negative_prompt_embeds=negative_prompt_embeds,
            clean_caption=clean_caption,
        )

        if do_classifier_free_guidance:
            prompt_embeds = torch.cat([negative_prompt_embeds, prompt_embeds])

        # 4. Prepare timesteps
        if timesteps is not None:
            self.scheduler.set_timesteps(timesteps=timesteps, device=device)
            timesteps = self.scheduler.timesteps
            num_inference_steps = len(timesteps)
        else:
            self.scheduler.set_timesteps(num_inference_steps, device=device)
            timesteps = self.scheduler.timesteps

        # 5. Prepare intermediate images
        intermediate_images = self.prepare_intermediate_images(
            batch_size * num_images_per_prompt,
            self.unet.config.in_channels,
            num_frames,
            height,
            width,
            prompt_embeds.dtype,
            device,
            generator,
        )

        # 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)

        # HACK: see comment in `enable_model_cpu_offload`
        if hasattr(self, "text_encoder_offload_hook") and self.text_encoder_offload_hook is not None:
            self.text_encoder_offload_hook.offload()

        # 7. Denoising loop
        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):
                model_input = (
                    torch.cat([intermediate_images] * 2) if do_classifier_free_guidance else intermediate_images
                )
                model_input = self.scheduler.scale_model_input(model_input, t)

                # predict the noise residual
                noise_pred = self.unet(
                    model_input,
                    t,
                    encoder_hidden_states=prompt_embeds,
                    cross_attention_kwargs=cross_attention_kwargs,
                ).sample

                # perform guidance
                if do_classifier_free_guidance:
                    noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
                    noise_pred_uncond, _ = noise_pred_uncond.split(model_input.shape[1], dim=1)
                    noise_pred_text, predicted_variance = noise_pred_text.split(model_input.shape[1], dim=1)
                    noise_pred = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond)
                    noise_pred = torch.cat([noise_pred, predicted_variance], dim=1)

                if self.scheduler.config.variance_type not in [
                    "learned",
                    "learned_range",
                ]:
                    noise_pred, _ = noise_pred.split(model_input.shape[1], dim=1)

                # reshape latents
                bsz, channel, frames, height, width = intermediate_images.shape
                intermediate_images = intermediate_images.permute(0, 2, 1, 3, 4).reshape(
                    bsz * frames, channel, height, width
                )
                noise_pred = noise_pred.permute(0, 2, 1, 3, 4).reshape(bsz * frames, -1, height, width)

                # compute the previous noisy sample x_t -> x_t-1
                intermediate_images = self.scheduler.step(
                    noise_pred, t, intermediate_images, **extra_step_kwargs
                ).prev_sample

                # reshape latents back
                intermediate_images = (
                    intermediate_images[None, :].reshape(bsz, frames, channel, height, width).permute(0, 2, 1, 3, 4)
                )

                # 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, intermediate_images)

        video_tensor = intermediate_images

        if output_type == "pt":
            video = video_tensor
        else:
            video = tensor2vid(video_tensor)

        # Offload last model to CPU
        if hasattr(self, "final_offload_hook") and self.final_offload_hook is not None:
            self.final_offload_hook.offload()

        if not return_dict:
            return (video,)

        return TextToVideoPipelineOutput(frames=video)