Loads the class's state. Args: state_dict (dict): scaler state. Should be an object returned from a call to :meth:

Download a file from `url` into `model_dir`, using the file present if possible. Returns the path to the downloaded file.

Expand the tensor `v` to the dim `dims`. Args: `v`: a PyTorch tensor with shape [N]. `dim`: a `int`. Returns: a

Converts a denoiser output to a Karras ODE derivative.

Returns the state of the class as a :class:`dict`.

r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distributi

Conv layer with padding, normalization, activation mode: CNA --> Conv -> Norm -> Act NAC --> Norm -> Act --> Conv (Identity Mappings

Compute lambda_t = log(alpha_t) - log(sigma_t) of a given continuous-time label t in [0, T].

Performs all_gather operation on the provided tensors. *** Warning ***: torch.distributed.all_gather has no gradient.

Calculates the noise level (sigma_down) to step down to and the amount of noise to add (sigma_up) when doing an ancestral sampling step.

Convert the model to the noise prediction model or the data prediction model.