Run TAEHV roundtrip reconstruction on the given video paths.

Aggregate all ode pairs inside a folder into a lmdb dataset. Each pt file should contain a (key, value) pair representing a video's ODE t

Aggregate all ode pairs inside a folder into a lmdb dataset. Each pt file should contain a (key, value) pair representing a video's ODE t

One step for the second-order multistep DPMSolver. Args: model_output_list (`List[torch.Tensor]`): The di

One step for the third-order multistep DPMSolver. Args: model_output_list (`List[torch.Tensor]`): The dir

One step for the UniC (B(h) version). Args: this_model_output (`torch.Tensor`): The model outputs at `x_

One step for the UniP (B(h) version). Alternatively, `self.solver_p` is used if is specified. Args: model_output (`torch

Patch TGrow layers to use a smaller kernel if needed. Args: sd: state dict to patch

A few custom initialization steps that should be called after the object is created. Currently, the only one we have is to bind a few

直接从帧张量准备batch（跳过视频文件加载） Args: video_tensors: List[torch.Tensor], 每个张量形状为 [T, C, H, W] prompts: List[str], 提示文

Args: video_tensors: List[torch.Tensor], shape [T, C, H, W] prompts: List[str]

Inputs: video_paths: List[str], B paths of the videos. prompts: List[str], B prompts for the videos. eval

Save all configurations to a JSON file.

Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with the multistep UniPC.

Input: - timestep: the timestep with shape [B*T] Output: the corresponding weighting [B*T]

Preprocess inputs to token sequences and return a batch

Base class for denoising loss. Input: - x: the clean data with shape [B, F, C, H, W] - x_pred: the predicted

Outputs: - prompts: List of Strings - latents: Tensor of shape (num_denoising_steps, num_frames, num_channels, height

Outputs: - prompts: List of Strings - latents: Tensor of shape (num_denoising_steps, num_frames, num_chan

Returns: dict: A dictionary containing: - image: PIL Image - caption: str - targe

Args: data_dir (str): Path to the directory containing: - target_crop_info_*.json (metadata file)

r""" Initializes the image-to-video generation model components. Args: config (EasyDict): Object containi

r""" Initializes the Wan text-to-video generation model components. Args: config (EasyDict): Object conta

r""" Initialize the diffusion model backbone. Args: model_type (`str`, *optional*, defaults to 't2v'): Mo