Diffusion model with a Transformer backbone.

Diffusion model with a Transformer backbone.

Load the FaceForensics video files Args: target_video_len (int): the number of video frames will be load. align_transform (ca

r""" A feed-forward layer. Parameters: dim (`int`): The number of channels in the input. dim_out (`int`, *optional*): The num

Load the UCF101 video files Args: target_video_len (int): the number of video frames will be load. align_transform (calla

Load the UCF101 video files Args: target_video_len (int): the number of video frames will be load. align_transform (callable)

r""" Norm layer adaptive layer norm single (adaLN-single). As proposed in PixArt-Alpha (see: https://arxiv.org/abs/2310.00426; Section 2.3).

r""" A basic Transformer block. Parameters: dim (`int`): The number of channels in the input and output. num_attention_heads

Using Decord(https://github.com/dmlc/decord) to initialize the video_reader.

Using Decord(https://github.com/dmlc/decord) to initialize the video_reader.

Using Decord(https://github.com/dmlc/decord) to initialize the video_reader.

Using Decord(https://github.com/dmlc/decord) to initialize the video_reader.

Load the FaceForensics video files Args: target_video_len (int): the number of video frames will be load. align_transform

The final layer of Latte.

The final layer of Latte.

Uses the CLIP transformer encoder for text (from Hugging Face)

r""" A gated self-attention dense layer that combines visual features and object features. Parameters: query_dim (`int`): The number

Utilities for training and sampling diffusion models. Original ported from this codebase: https://github.com/hojonathanho/diffusion/bl

Embeds class labels into vector representations. Also handles label dropout for classifier-free guidance.

Embeds class labels into vector representations. Also handles label dropout for classifier-free guidance.

r""" Pipeline for text-to-image generation using PixArt-Alpha. This model inherits from [`DiffusionPipeline`]. Check the superclass documenta

Flip the video clip along the horizontal direction with a given probability Args: p (float): probability of the clip being flipped

A diffusion process which can skip steps in a base diffusion process. :param use_timesteps: a collection (sequence or set) of timesteps fro

Temporally crop the given frame indices at a random location. Args: size (int): Desired length of frames will be seen in the model.

Embeds text prompt into vector representations. Also handles text dropout for classifier-free guidance.

Embeds scalar timesteps into vector representations.

Embeds scalar timesteps into vector representations.

Convert tensor data type from uint8 to float, divide value by 255.0 and permute the dimensions of clip tensor

The output of [`Transformer2DModel`]. Args: sample (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)` or `(ba

A Latte tansformer block with adaptive layer norm zero (adaLN-Zero) conditioning.

A Latte block with adaptive layer norm zero (adaLN-Zero) conditioning.

First scale to the specified size in equal proportion to the short edge, then center cropping

First use the short side for cropping length, center crop video, then resize to the specified size

r"""Collects the scalars broadcasted by `report()` and `report0()` and computes their long-term averages (mean and standard deviation) over us

Convenience class that behaves like a dict but allows access with the attribute syntax.

Slide along the long edge, with the short edge as crop size. And resie to the desired size.

Redirect stderr to stdout, optionally print stdout to a file, and optionally force flushing on both stdout and the file.

Which type of output the model predicts.

What is used as the model's output variance. The LEARNED_RANGE option has been added to allow the model to predict values between FIXE

Normalize the video clip by mean subtraction and division by standard deviation Args: mean (3-tuple): pixel RGB mean std

A distribution over timesteps in the diffusion process, intended to reduce variance of the objective. By default, samplers perform unb