Package mvpa :: Package clfs :: Module base
[hide private]
[frames] | no frames]

Module base

source code

Base classes for all classifiers.

Base Classifiers can be grouped according to their function as

Classes [hide private]
  ClassifierCombiner
Provides a decision using training a classifier on predictions/values
    Basic Classifiers
  Classifier
Abstract classifier class to be inherited by all classifiers
  BoostedClassifier
Classifier containing the farm of other classifiers.
  ProxyClassifier
Classifier which decorates another classifier
    BoostedClassifiers
  CombinedClassifier
BoostedClassifier which combines predictions using some PredictionsCombiner functor.
  MulticlassClassifier
CombinedClassifier to perform multiclass using a list of BinaryClassifier.
  SplitClassifier
BoostedClassifier to work on splits of the data
    ProxyClassifiers
  BinaryClassifier
ProxyClassifier which maps set of two labels into +1 and -1
  MappedClassifier
ProxyClassifier which uses some mapper prior training/testing.
  FeatureSelectionClassifier
ProxyClassifier which uses some FeatureSelection prior training.
    PredictionsCombiners for CombinedClassifier
  PredictionsCombiner
Base class for combining decisions of multiple classifiers
  MaximalVote
Provides a decision using maximal vote rule
  MeanPrediction
Provides a decision by taking mean of the results
Functions [hide private]
 
_deepcopyclf(clf)
Deepcopying of a classifier.
source code

Imports: operator, sys, N, deepcopy, time, Set, group_kwargs, idhash, MaskMapper, NFoldSplitter, StateVariable, Stateful, Harvestable, Parametrized, Parameter, ConfusionMatrix, RegressionStatistics, FirstAxisMean, SecondAxisSumOfAbs, BoostedClassifierSensitivityAnalyzer, ProxyClassifierSensitivityAnalyzer, MappedClassifierSensitivityAnalyzer, warning, traceback, debug


Function Details [hide private]

_deepcopyclf(clf)

source code 

Deepcopying of a classifier.

If deepcopy fails -- tries to untrain it first so that there is no swig bindings attached