Package parsimony :: Package functions :: Package nesterov :: Module l1tv

Module l1tv

The :mod:`parsimony.functions.nesterov.L1TV` module contains the loss function for the L1 + TV penalty, smoothed together using Nesterov's technique.

Created on Mon Feb 3 17:04:14 2014

Author: Tommy Löfstedt, Vincent Guillemot, Edouard Duchesnay and Fouad Hadj-Selem

License: BSD 3-clause.

Classes

[hide private]

L1TV
The proximal operator of the smoothed sum of the TV and L1 functions

Functions

[hide private]

A_from_mask(*args, **kwargs)

source code

linear_operator_from_mask(mask, num_variables, penalty_start=0)
Generates the linear operator for the total variation Nesterov function from a mask for a 3D image.

source code

A_from_shape(*args, **kwargs)

source code

linear_operator_from_shape(shape, num_variables, penalty_start=0)
Generates the linear operator for the total variation Nesterov function from the shape of a 3D image.

source code

Variables

[hide private]

__package__ = 'parsimony.functions.nesterov'

Function Details

[hide private]

A_from_mask(*args, **kwargs)

source code

Decorators:

@utils.deprecated("linear_operator_from_mask")

linear_operator_from_mask(mask, num_variables, penalty_start=0)

source code

Generates the linear operator for the total variation Nesterov function
from a mask for a 3D image.

Parameters
----------
mask : Numpy array. The mask. The mask does not involve any intercept
        variables.

num_variables : Positive integer. The total number of variables, including
        the intercept variable(s).

penalty_start : Non-negative integer. The number of variables to exempt
        from penalisation. Equivalently, the first index to be penalised.
        Default is 0, all variables are included.

A_from_shape(*args, **kwargs)

source code

Decorators:

@utils.deprecated("linear_operator_from_shape")

linear_operator_from_shape(shape, num_variables, penalty_start=0)

source code

Generates the linear operator for the total variation Nesterov function
from the shape of a 3D image.

Parameters
----------
shape : List or tuple with 1, 2 or 3 elements. The shape of the 1D, 2D or
        3D image. shape has the form (Z, Y, X), where Z is the number of
        "layers", Y is the number of rows and X is the number of columns.
        The shape does not involve any intercept variables.

num_variables : Positive integer. The total number of variables, including
        the intercept variable(s).

penalty_start : Non-negative integer. The number of variables to exempt
        from penalisation. Equivalently, the first index to be penalised.
        Default is 0, all variables are included.