This content refers to the previous stable release of PyMVPA. Please visit www.pymvpa.org for the most recent version of PyMVPA and its documentation.

measures.base

Module: measures.base

Inheritance diagram for mvpa.measures.base:

Base class for data measures: algorithms that quantify properties of datasets.

Besides the DatasetMeasure base class this module also provides the (abstract) FeaturewiseDatasetMeasure class. The difference between a general measure and the output of the FeaturewiseDatasetMeasure is that the latter returns a 1d map (one value per feature in the dataset). In contrast there are no restrictions on the returned value of DatasetMeasure except for that it has to be in some iterable container.

Classes

BoostedClassifierSensitivityAnalyzer

class mvpa.measures.base.BoostedClassifierSensitivityAnalyzer(*args_, **kwargs_)

Bases: mvpa.measures.base.Sensitivity

Set sensitivity analyzers to be merged into a single output

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

Sensitivity

Initialize instance of BoostedClassifierSensitivityAnalyzer

Parameters:

• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled

combined_analyzer

untrain()

Untrain BoostedClassifierSensitivityAnalyzer

CombinedFeaturewiseDatasetMeasure

class mvpa.measures.base.CombinedFeaturewiseDatasetMeasure(analyzers=None, combiner=None, **kwargs)

Bases: mvpa.measures.base.FeaturewiseDatasetMeasure

Set sensitivity analyzers to be merged into a single output

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm
• sensitivities: Sensitivities produced by each analyzer

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

FeaturewiseDatasetMeasure

Initialize CombinedFeaturewiseDatasetMeasure

Parameters:

• analyzers (list or None) – List of analyzers to be used. There is no logic to populate such a list in __call__, so it must be either provided to the constructor or assigned to .analyzers prior calling
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled
• combiner (Functor) – The combiner is only applied if the computed featurewise dataset measure is more than one-dimensional. This is different from a transformer, which is always applied. By default, the sum of absolute values along the second axis is computed.
• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.

analyzers

Used analyzers

untrain()

Untrain CombinedFDM

DatasetMeasure

class mvpa.measures.base.DatasetMeasure(transformer=None, null_dist=None, **kwargs)

Bases: mvpa.misc.state.ClassWithCollections

A measure computed from a Dataset

All dataset measures support arbitrary transformation of the measure after it has been computed. Transformation are done by processing the measure with a functor that is specified via the transformer keyword argument of the constructor. Upon request, the raw measure (before transformations are applied) is stored in the raw_results state variable.

Additionally all dataset measures support the estimation of the probabilit(y,ies) of a measure under some distribution. Typically this will be the NULL distribution (no signal), that can be estimated with permutation tests. If a distribution estimator instance is passed to the null_dist keyword argument of the constructor the respective probabilities are automatically computed and stored in the null_prob state variable.

Note

For developers: All subclasses shall get all necessary parameters via their constructor, so it is possible to get the same type of measure for multiple datasets by passing them to the __call__() method successively.

See also

Please refer to the documentation of the base class for more information:

ClassWithCollections

Note

Available state variables:

• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

ClassWithCollections

Does nothing special.

Parameters:

• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled

null_dist

Return Null Distribution estimator

null_prob = None

Stores the probability of a measure under the NULL hypothesis

null_t = None

Stores the t-score corresponding to null_prob under assumption of Normal distribution

transformer

Return transformer

untrain()

‘Untraining’ Measure

Some derived classes might used classifiers, so we need to untrain those

FeatureSelectionClassifierSensitivityAnalyzer

class mvpa.measures.base.FeatureSelectionClassifierSensitivityAnalyzer(*args_, **kwargs_)

Bases: mvpa.measures.base.ProxyClassifierSensitivityAnalyzer

Set sensitivity analyzer output be reverse mapped using mapper of the slave classifier

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• clf_sensitivities: Stores sensitivities of the proxied classifier
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

ProxyClassifierSensitivityAnalyzer

Initialize instance of ProxyClassifierSensitivityAnalyzer

Parameters:

• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled

FeaturewiseDatasetMeasure

class mvpa.measures.base.FeaturewiseDatasetMeasure(combiner=<function SecondAxisSumOfAbs at 0x4892f50>, **kwargs)

Bases: mvpa.measures.base.DatasetMeasure

A per-feature-measure computed from a Dataset (base class).

Should behave like a DatasetMeasure.

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

DatasetMeasure

Initialize

Parameters:

• combiner (Functor) – The combiner is only applied if the computed featurewise dataset measure is more than one-dimensional. This is different from a transformer, which is always applied. By default, the sum of absolute values along the second axis is computed.
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled
• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.

combiner

Return combiner

MappedClassifierSensitivityAnalyzer

class mvpa.measures.base.MappedClassifierSensitivityAnalyzer(*args_, **kwargs_)

Bases: mvpa.measures.base.ProxyClassifierSensitivityAnalyzer

Set sensitivity analyzer output be reverse mapped using mapper of the slave classifier

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• clf_sensitivities: Stores sensitivities of the proxied classifier
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

ProxyClassifierSensitivityAnalyzer

Initialize instance of ProxyClassifierSensitivityAnalyzer

Parameters:

• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled

ProxyClassifierSensitivityAnalyzer

class mvpa.measures.base.ProxyClassifierSensitivityAnalyzer(*args_, **kwargs_)

Bases: mvpa.measures.base.Sensitivity

Set sensitivity analyzer output just to pass through

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• clf_sensitivities: Stores sensitivities of the proxied classifier
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

Sensitivity

Initialize instance of ProxyClassifierSensitivityAnalyzer

Parameters:

• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled

analyzer

untrain()

Sensitivity

class mvpa.measures.base.Sensitivity(clf, force_training=True, **kwargs)

Bases: mvpa.measures.base.FeaturewiseDatasetMeasure

No documentation found. Sorry!

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

FeaturewiseDatasetMeasure

Initialize the analyzer with the classifier it shall use.

Parameters:

• clf (Classifier) – classifier to use.
• force_training (Bool) – if classifier was already trained – do not retrain
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled
• combiner (Functor) – The combiner is only applied if the computed featurewise dataset measure is more than one-dimensional. This is different from a transformer, which is always applied. By default, the sum of absolute values along the second axis is computed.
• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.

clf

feature_ids

Return feature_ids used by the underlying classifier

untrain()

Untrain corresponding classifier for Sensitivity

SplitFeaturewiseDatasetMeasure

class mvpa.measures.base.SplitFeaturewiseDatasetMeasure(splitter, analyzer, insplit_index=0, combiner=None, **kwargs)

Bases: mvpa.measures.base.FeaturewiseDatasetMeasure

Compute measures across splits for a specific analyzer

Note

Available state variables:

• base_sensitivities: Stores basic sensitivities if the sensitivity relies on combining multiple ones
• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm
• sensitivities: Sensitivities produced for each split
• splits: Store the actual splits of the data. Can be memory expensive

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

FeaturewiseDatasetMeasure

Initialize SplitFeaturewiseDatasetMeasure

Parameters:

• splitter (Splitter) – Splitter to use to split the dataset
• analyzer (DatasetMeasure) – Measure to be used. Could be analyzer as well (XXX)
• insplit_index (int) – splitter generates tuples of dataset on each iteration (usually 0th for training, 1st for testing). On what split index in that tuple to operate.
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled
• combiner (Functor) – The combiner is only applied if the computed featurewise dataset measure is more than one-dimensional. This is different from a transformer, which is always applied. By default, the sum of absolute values along the second axis is computed.
• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.

untrain()

Untrain SplitFeaturewiseDatasetMeasure

StaticDatasetMeasure

class mvpa.measures.base.StaticDatasetMeasure(measure=None, bias=None, *args, **kwargs)

Bases: mvpa.measures.base.DatasetMeasure

A static (assigned) sensitivity measure.

Since implementation is generic it might be per feature or per whole dataset

Note

Available state variables:

• null_prob+: State variable
• null_t: State variable
• raw_results: Computed results before applying any transformation algorithm

(States enabled by default are listed with +)

See also

Please refer to the documentation of the base class for more information:

DatasetMeasure

Initialize.

Parameters:

• measure – actual sensitivity to be returned
• bias – optionally available bias
• enable_states (None or list of basestring) – Names of the state variables which should be enabled additionally to default ones
• disable_states (None or list of basestring) – Names of the state variables which should be disabled
• transformer (Functor) – This functor is called in __call__() to perform a final processing step on the to be returned dataset measure. If None, nothing is called
• null_dist (instance of distribution estimator) – The estimated distribution is used to assign a probability for a certain value of the computed measure.

bias