Bregman_ICNN

class deepinv.optim.Bregman_ICNN(forw_model, conj_model=None)

Bases: Bregman

Module for the using a deep ICNN as Bregman potential.

conjugate(x)

Computes the convex conjugate potential.

Parameters:

x (torch.Tensor) – Variable \(x\) at which the conjugate is computed.

Returns:

(torch.tensor) conjugate potential \(\phi^*(y)\).

fn(x)

Computes the Bregman potential.

Parameters:

x (torch.Tensor) – Variable \(x\) at which the potential is computed.

Returns:

(torch.tensor) potential \(\phi(x)\).