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

Initialize the GANLoss class. Parameters: gan_mode (str) - - the type of GAN objective. It currently supports vanilla, lsgan, an

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

Create a linear module.

Similar to np.fft.fftshift but applies to PyTorch Tensors Args: x: A PyTorch tensor. dim: Which dimension to fftshift. Re

Similar to np.fft.ifftshift but applies to PyTorch Tensors Args: x: A PyTorch tensor. dim: Which dimension to ifftshift.

Normalize img in arbitrary range to [0, 1]

Zero out the parameters of a module and return it.

Normalize img in arbitrary range to [0, 1]

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

A counter for the `thop` package to count the operations in an attention operation. Meant to be used like: macs, params = thop.pr

Copy model parameters into a (differently-shaped) list of full-precision parameters.

Generate a random tlhw with configuration. Args: config: Config should have configuration including img Returns: tuple: (top

Similar to np.roll but applies to PyTorch Tensors. Args: x: A PyTorch tensor. shift: Amount to roll. dim: Which dimen

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

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

Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of (1-beta) over time from t = [

Convert the torso of the model to float16.

Return a Gaussian

FFT for multi-coil

Apply centered 2 dimensional Fast Fourier Transform. Args: data: Complex valued input data containing at least 3 dimensions:

Get a pre-defined beta schedule for the given name. The beta schedule library consists of beta schedules which remain similar in the lim

Create label tensors with the same size as the input. Parameters: prediction (tensor) - - tpyically the prediction from a discrim

Apply centered 2-dimensional Inverse Fast Fourier Transform. Args: data: Complex valued input data containing at least 3 dimensions:

Copy the master parameter data back into the model parameters.

Copy the gradients from the model parameters into the master parameters from make_master_params().

Nonlinear deblur requires external codes (bkse).

Compute the mean of the diffusion posteriro: q(x_{t-1} | x_t, x_0)

Compute the mean of the diffusion posteriro: q(x_{t-1} | x_t, x_0)

Diffuse the data for a given number of diffusion steps. In other words, sample from q(x_t | x_0). :param x_start: the initi

Generate a random sqaure mask for inpainting