Package mvpa :: Package clfs :: Module smlr :: Class SMLRWeights
[hide private]
[frames] | no frames]

Class SMLRWeights

source code


SensitivityAnalyzer that reports the weights SMLR trained on a given Dataset.

By default SMLR provides multiple weights per feature (one per label in training dataset). By default, all weights are combined into a single sensitivity value. Please, see the FeaturewiseDatasetMeasure constructor arguments how to custmize this behavior.

Nested Classes [hide private]

Inherited from misc.state.ClassWithCollections: __metaclass__

Instance Methods [hide private]
 
_call(self, dataset=None)
Extract weights from SMLR classifier.
source code

Inherited from measures.base.Sensitivity: __call__, __init__, __repr__, feature_ids, untrain

Inherited from measures.base.Sensitivity (private): _setClassifier

Inherited from measures.base.FeaturewiseDatasetMeasure: combiner

Inherited from measures.base.FeaturewiseDatasetMeasure (private): _postcall

Inherited from measures.base.DatasetMeasure: null_dist, transformer

Inherited from misc.state.ClassWithCollections: __getattribute__, __new__, __setattr__, __str__, reset

Inherited from object: __delattr__, __format__, __hash__, __reduce__, __reduce_ex__, __sizeof__, __subclasshook__

Class Variables [hide private]
  biases = StateVariable(enabled= True, doc= "A 1-d ndarray of b...
  _LEGAL_CLFS = [SMLR]
If Sensitivity is classifier specific, classes of classifiers should be listed in the list

Inherited from measures.base.Sensitivity: clf

Inherited from measures.base.FeaturewiseDatasetMeasure: base_sensitivities

Inherited from measures.base.DatasetMeasure: __doc__, null_prob, null_t, raw_results

Inherited from misc.state.ClassWithCollections: _DEV__doc__, descr

Instance Variables [hide private]

Inherited from measures.base.Sensitivity (private): _force_training

Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

_call(self, dataset=None)

source code 

Extract weights from SMLR classifier.

SMLR always has weights available, so nothing has to be computed here.

Overrides: measures.base.DatasetMeasure._call

Class Variable Details [hide private]

biases

Value:
StateVariable(enabled= True, doc= "A 1-d ndarray of biases")