SupLoss

class deepinv.loss.SupLoss(metric: Metric | Module = MSELoss())[source]

Bases: Loss

Standard supervised loss

The supervised loss is defined as

\[\|x-\inverse{y}\|^2\]

where \(\inverse{y}\) is the reconstructed signal and \(x\) is the ground truth target.

By default, the error is computed using the MSE metric, however any other metric (e.g., \(\ell_1\)) can be used as well. If called with arguments x_net, x, this is simply a wrapper for the metric metric.

Parameters:

metric (Metric, torch.nn.Module) – metric used for computing data consistency, which is set as the mean squared error by default.

forward(x_net, x, **kwargs)[source]

Computes the loss.

Parameters:
Returns:

(torch.Tensor) loss.

Examples using SupLoss:

Imaging inverse problems with adversarial networks

Imaging inverse problems with adversarial networks

Training a reconstruction network.

Training a reconstruction network.

Learned Iterative Soft-Thresholding Algorithm (LISTA) for compressed sensing

Learned Iterative Soft-Thresholding Algorithm (LISTA) for compressed sensing

Vanilla Unfolded algorithm for super-resolution

Vanilla Unfolded algorithm for super-resolution

Learned iterative custom prior

Learned iterative custom prior

Deep Equilibrium (DEQ) algorithms for image deblurring

Deep Equilibrium (DEQ) algorithms for image deblurring

Learned Primal-Dual algorithm for CT scan.

Learned Primal-Dual algorithm for CT scan.

Unfolded Chambolle-Pock for constrained image inpainting

Unfolded Chambolle-Pock for constrained image inpainting