Initialize the step_index counter for the scheduler.

Initialize the step_index counter for the scheduler.

Autoencoder3d adapted from Stable Diffusion 1.x, 2.x and XL.

Performs an all-gather operation on the input tensor along the specified dimension. Args: input_ (torch.Tensor): Input tensor of shape [B

Convert audio bytes to a 16kHz mono torch tensor in [-1, 1]. Args: audio_bytes (bytes): Raw audio bytes target_sr (int):

Convert the model output to the corresponding type the DPMSolver/DPMSolver++ algorithm needs. DPM-Solver is designed to discretize an

r""" Convert the model output to the corresponding type the UniPC algorithm needs. Args: model_output (`torch.Tensor`):

One step for the first-order DPMSolver (equivalent to DDIM). Args: model_output (`torch.Tensor`): The dir

videos: A list of videos each with shape [C, T, H, W].

r""" Initialize model parameters using Xavier initialization.

Initialize the sequence parallel group.

Load configuration from yaml file

keys in each kwarg: e seq_lens grid_sizes freqs context context_lens

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

Implementation of sinc, i.e. sin(pi * x) / (pi * x) __Warning__: Different to julius.sinc, the input is multiplied by `pi`!

Implementation of sinc, i.e. sin(pi * x) / (pi * x) __Warning__: Different to julius.sinc, the input is multiplied by `pi`!

Scale (h, w) to match a target area if provided, then snap both dimensions to the nearest multiple of 32 (min 32). Args: h (

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