Same as to in torch module Don't really underestand why this isn't a module in the first place

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

Resize an image to the target size, adding padding to the shorter side to maintain aspect ratio. :param img: img_path (str) or PIL image

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

img_h: int, an image height img_w: int, an image width marg: int, a margin for a box starting coordinate p_irr: float, 0 <= p_irr <=

Create a linear module.

Imagenet Superresolution Dataloader Performs following ops in order: 1. crops a crop of size s from image either as random o

replace background pixel with random color in rendering

Concatenate x and its positional encodings, following NeRF.

Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without affecting the

Save the current parameters for restoring later. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to be

Zero out the parameters of a module and return it.