Zero out the parameters of a module and return it.

Create a 1D, 2D, or 3D convolution module.

Evaluate a function without caching intermediate activations, allowing for reduced memory at the expense of extra compute in the backward pas

Create a linear module.

Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.

Create sinusoidal timestep embeddings. :param timesteps: a 1-D Tensor of N indices, one per batch element. These may be

Create a 1D, 2D, or 3D average pooling module.

read and parse .h5 file, and obtain the absolute actions and the action differences

Convertion from the original agibotworld data format to the expected inference data format