Reconstructor

class deepinv.models.Reconstructor(device='cpu')

Bases: Module

Base class for reconstruction models.

Provides a template for defining reconstruction models.

Reconstructors provide a signal estimate x_hat as x_hat = model(y, physics) where y are the measurements and physics is the forward model \(A\) (possibly including information about the noise distribution too).

The base class inherits from torch.nn.Module.

forward(y, physics, **kwargs)

Applies reconstruction model \(\inversef{y}{A}\).

Parameters:

• y (torch.Tensor) – measurements.

• physics (deepinv.physics.Physics) – forward model \(A\).

Returns:

(torch.Tensor) reconstructed tensor.

Examples using Reconstructor:

Reconstructing an image using the deep image prior.

Reconstructing an image using the deep image prior.

Training a reconstruction network.

Training a reconstruction network.

Image reconstruction with a diffusion model

Image reconstruction with a diffusion model

Self-supervised learning with measurement splitting

Self-supervised learning with measurement splitting

Self-supervised denoising with the UNSURE loss.

Self-supervised denoising with the UNSURE loss.

Self-supervised denoising with the SURE loss.

Self-supervised denoising with the SURE loss.

Self-supervised denoising with the Neighbor2Neighbor loss.

Self-supervised denoising with the Neighbor2Neighbor loss.

Self-supervised learning from incomplete measurements of multiple operators.

Self-supervised learning from incomplete measurements of multiple operators.

Imaging inverse problems with adversarial networks

Imaging inverse problems with adversarial networks