tfmri.convex.ConvexFunctionTotalVariation

class ConvexFunctionTotalVariation(domain_shape, axes=None, scale=None, dtype=tf.float32, name=None)

Bases: tensorflow_mri.python.ops.convex_ops.ConvexFunctionLinearOperatorComposition

A ConvexFunction representing a total variation regularization term.

For a given input \(x\), computes \(\lambda \left\| Dx \right\|_1\), where \(\lambda\) is a scaling factor and \(D\) is the finite difference operator.

Parameters

• domain_shape – A 1D integer tf.Tensor. The shape of the domain. Defaults to None. The domain of this ConvexFunction may have multiple axes.

• axes –

  An int or a list of ints. The axes along which to compute the total variation. If None (default), the total variation is computed over all axes.

• scale – A float. A scaling factor.

• dtype – A tf.DType. The dtype of the inputs.

• name – A name for this ConvexFunction.

Initialize this ConvexFunction.