Args: parameters: Iterable of `torch.nn.Parameter`; usually the result of `model.parameters()`. decay: The exponential decay.

r""" Concatenates two noise :math:`N = N_1 \circ N_2` via the mul operation The resulting operator will add the noise from both noise

Create the score model.

1x1 convolution with DDPM initialization.

r"""Downsample a batch of 2D images with the given filter. Accepts a batch of 2D images of the shape `[N, C, H, W]` or `[N, H, W, C]` and

The probability flow ODE sampler with simple Euler discretization. Args: model: A velocity model. z: If present, generate

Evaluate trained models. Args: config: Configuration to use. workdir: Working directory for checkpoints. eval_folder: The subfolder for

Evaluate trained models. Args: config: Configuration to use. workdir: Working directory for checkpoints. eval_folder: The subfo

Calculation of FID. Params: -- images1/2 : Images of shape (N, H, W, C) -- batch_size : The images numpy array is split into batches wi

Runs the rematching finetune pipeline. Args: config: Configuration to use. workdir: Working directory for checkpoints and TF summaries. If

t is not used

Get Inception feature maps Parameters ---------- inp : torch.autograd.Variable Input tensor of shape Bx3xHxW. Val