Meta classifiers – classifiers which use other classifiers or preprocessing

Meta Classifiers can be grouped according to their function as

group BoostedClassifiers:
 CombinedClassifier MulticlassClassifier SplitClassifier
group ProxyClassifiers:
 ProxyClassifier BinaryClassifier MappedClassifier FeatureSelectionClassifier
group PredictionsCombiners for CombinedClassifier:
 PredictionsCombiner MaximalVote MeanPrediction
Inheritance diagram of mvpa2.clfs.meta


asobjarray(x) Generates numpy.ndarray with dtype object from an iterable
cartesian_distance(a, b) Return Cartesian distance between a and b
first_axis_mean(x) Mean computed along the first axis.
get_samples_by_attr(dataset, attr, values[, ...]) Return indices of samples given a list of attributes
group_kwargs(prefixes[, assign, passthrough]) Decorator function to join parts of kwargs together
is_sequence_type isSequenceType(a) – Return True if a has a sequence type, False otherwise.


AttributeMap([map, mapnumeric, ...]) Map to translate literal values to numeric ones (and back).
BinaryClassifier(clf, poslabels, neglabels, ...) ProxyClassifier which maps set of two labels into +1 and -1
BinaryClassifierSensitivityAnalyzer(\*args_, ...) Set sensitivity analyzer output to have proper labels
BoostedClassifier([clfs, propagate_ca]) Classifier containing the farm of other classifiers.
BoostedClassifierSensitivityAnalyzer(\*args_, ...) Set sensitivity analyzers to be merged into a single output
ClassWithCollections([descr]) Base class for objects which contain any known collection
Classifier([space]) Abstract classifier class to be inherited by all classifiers
ClassifierCombiner(clf[, variables]) Provides a decision using training a classifier on predictions/estimates
CombinedClassifier([clfs, combiner]) BoostedClassifier which combines predictions using some
ConditionalAttribute([enabled]) Simple container intended to conditionally store the value
Dataset(samples[, sa, fa, a]) Generic storage class for datasets with multiple attributes.
FeatureSelectionClassifier(clf, mapper, \*\*kwargs) This is nothing but a MappedClassifier.


MappedClassifier(clf, mapper, \*\*kwargs) ProxyClassifier which uses some mapper prior training/testing.
MappedClassifierSensitivityAnalyzer(\*args_, ...) Set sensitivity analyzer output be reverse mapped using mapper of the
MaximalVote(\*\*kwargs) Provides a decision using maximal vote rule
MeanPrediction([descr]) Provides a decision by taking mean of the results
MulticlassClassifier(clf[, bclf_type]) Perform multiclass classification using a list of binary classifiers.
NFoldPartitioner([cvtype]) Generic N-fold data partitioner.
Parameter(default[, constraints, ro, index, ...]) This class shall serve as a representation of a parameter.
PredictionsCombiner([descr]) Base class for combining decisions of multiple classifiers
ProxyClassifier(clf, \*\*kwargs) Classifier which decorates another classifier
ProxyClassifierSensitivityAnalyzer(\*args_, ...) Set sensitivity analyzer output just to pass through
RegressionAsClassifier(clf[, centroids, ...]) Allows to use arbitrary regression for classification.
RegressionAsClassifierSensitivityAnalyzer(...) Set sensitivity analyzer output to have proper labels
SplitClassifier(clf[, partitioner, splitter]) BoostedClassifier to work on splits of the data
Splitter(attr[, attr_values, count, ...]) Generator node for dataset splitting.
TreeClassifier(clf, groups, \*\*kwargs) TreeClassifier which allows to create hierarchy of classifiers