dival.measure module

Provides the abstract Measure base class and some popular measures.

Measure instances are identified by a unique short_name, which is used by the evaluation module.

dival.measure.gen_unique_name(name_orig)[source]

class dival.measure.Measure(short_name=None)[source]

Bases: ABC

Abstract base class for measures used for evaluation.

In subclasses, either __init__() should be inherited or it should call super().__init__() in order to register the short_name and to ensure it is unique.

measure_type

The measure type. Measures with type 'distance' should attain small values if the reconstruction is good. Measures with type 'quality' should attain large values if the reconstruction is good.

Type:: {'distance', 'quality'}

short_name

Short name of the measure, used as identifier (key in measure_dict).

Type:: str

name

Name of the measure.

Type:: str

description

Description of the measure.

Type:: str

measure_type = None: Class attribute, default value for measure_type.

name = '': Class attribute, default value for name.

description = '': Class attribute, default value for description.

measure_dict = {'l2': <dival.measure.L2Measure object>, 'mse': <dival.measure.MSEMeasure object>, 'psnr': <dival.measure.PSNRMeasure object>, 'ssim': <dival.measure.SSIMMeasure object>}: Class attribute, registry of all measures with their short_name as key.

__init__(short_name=None)[source]

Parameters:: short_name (str, optional) – The short name of this measure, used as identifier in measure_dict. If None is passed and short_name was not set by the subclass, the __name__ of the subclass is used. If short_name is already taken by another instance, a unique short name is generated by appending a suffix of format '_%d'.

short_name = '': Class attribute, default value for short_name.

abstract apply(reconstruction, ground_truth)[source]

Calculate the value of this measure.

Parameters:

reconstruction (odl element) – The reconstruction.
ground_truth (odl element) – The ground truth to compare with.

Returns:

value – The value of this measure for the given reconstruction and ground_truth.

Return type:

float

classmethod get_by_short_name(short_name)[source]

Return Measure instance with given short name by registry lookup.

Parameters:: short_name (str) – Short name, identifier in measure_dict.
Returns:: measure – The instance.
Return type:: Measure

as_operator_for_fixed_ground_truth(ground_truth)[source]

Return an odl operator that can be applied to different reconstructions for fixed ground truth.

Returns:: op – odl operator.
Return type:: odl operator

class dival.measure.L2Measure(short_name=None)[source]

Bases: Measure

The euclidean (l2) distance measure.

measure_type = 'distance': Class attribute, default value for measure_type.

short_name = 'l2': Class attribute, default value for short_name.

name = 'euclidean distance': Class attribute, default value for name.

description = 'distance given by sqrt(sum((reconstruction-ground_truth)**2))': Class attribute, default value for description.

apply(reconstruction, ground_truth)[source]

Calculate the value of this measure.

Parameters:

reconstruction (odl element) – The reconstruction.
ground_truth (odl element) – The ground truth to compare with.

Returns:

value – The value of this measure for the given reconstruction and ground_truth.

Return type:

float

class dival.measure.MSEMeasure(short_name=None)[source]

Bases: Measure

The mean squared error distance measure.

measure_type = 'distance': Class attribute, default value for measure_type.

short_name = 'mse': Class attribute, default value for short_name.

name = 'mean squared error': Class attribute, default value for name.

description = 'distance given by 1/n * sum((reconstruction-ground_truth)**2)': Class attribute, default value for description.

apply(reconstruction, ground_truth)[source]

Calculate the value of this measure.

Parameters:

reconstruction (odl element) – The reconstruction.
ground_truth (odl element) – The ground truth to compare with.

Returns:

value – The value of this measure for the given reconstruction and ground_truth.

Return type:

float

dival.measure.MSE = <dival.measure.MSEMeasure object>: Instance of MSEMeasure, with short_name='mse'.

class dival.measure.PSNRMeasure(data_range=None, short_name=None)[source]

Bases: Measure

The peak signal-to-noise ratio (PSNR) measure.

The data range is automatically determined from the ground truth if not given to the constructor.

data_range

The data range (max-min possible value). If data_range is None, np.max(ground_truth) - np.min(ground_truth) is used in apply().

Type:: float or None

measure_type = 'quality': Class attribute, default value for measure_type.

short_name = 'psnr': Class attribute, default value for short_name.

name = 'peak signal-to-noise ratio': Class attribute, default value for name.

description = 'quality given by 10*log10(MAX**2/MSE)': Class attribute, default value for description.

__init__(data_range=None, short_name=None)[source]

Parameters:

data_range (float, optional) – The data range (max-min possible value). If data_range is None, np.max(ground_truth) - np.min(ground_truth) is used in apply().
short_name (str, optional) – Short name.

apply(reconstruction, ground_truth)[source]

Calculate the value of this measure.

Parameters:

reconstruction (odl element) – The reconstruction.
ground_truth (odl element) – The ground truth to compare with.

Returns:

value – The value of this measure for the given reconstruction and ground_truth.

Return type:

float

dival.measure.PSNR = <dival.measure.PSNRMeasure object>: Instance of PSNRMeasure, with data_range=None and short_name='psnr'.

class dival.measure.SSIMMeasure(short_name=None, **kwargs)[source]

Bases: Measure

The structural similarity index measure.

measure_type = 'quality': Class attribute, default value for measure_type.

short_name = 'ssim': Class attribute, default value for short_name.

name = 'structural similarity index': Class attribute, default value for name.

description = 'The (M)SSIM like described in `Wang et al. 2014 <https://doi.org/10.1109/TIP.2003.819861>`_.': Class attribute, default value for description.

__init__(short_name=None, **kwargs)[source]

This is a wrapper for skimage.metrics.structural_similarity(). The data range is automatically determined from the ground truth if not given to the constructor.

Parameters:

short_name (str, optional) – Short name.
kwargs (dict, optional) – Keyword arguments that will be passed to structural_similarity() in apply(). If data_range is not specified, np.max(ground_truth) - np.min(ground_truth) is used.

apply(reconstruction, ground_truth)[source]

Calculate the value of this measure.

Parameters:

reconstruction (odl element) – The reconstruction.
ground_truth (odl element) – The ground truth to compare with.

Returns:

value – The value of this measure for the given reconstruction and ground_truth.

Return type:

float

dival.measure.SSIM = <dival.measure.SSIMMeasure object>: Instance of SSIMMeasure, with data_range=None and short_name='ssim'.