Package moa.classifiers.meta

Class RCD

java.lang.Object

moa.AbstractMOAObject

moa.options.AbstractOptionHandler

moa.classifiers.AbstractClassifier

moa.classifiers.drift.DriftDetectionMethodClassifier

moa.classifiers.drift.SingleClassifierDrift

moa.classifiers.meta.RCD

All Implemented Interfaces:

Configurable, java.io.Serializable, CapabilitiesHandler, Classifier, MultiClassClassifier, AWTRenderable, Learner<Example<Instance>>, MOAObject, OptionHandler

public class RCD
extends SingleClassifierDrift

Creates a set of classifiers, each one representing a different context. Reuses classifier associating to each one a sample of data and compares new data to old ones using a multivariate non-parametric statistical test. Tests are performed in parallel and classifiers are stored based on their accuracy and stored time. 1) Parameterized number of classifiers to store. 2) Classifiers are stored removing the older ones if the set is full. 3) Classifier with higher significance value is selected. Based on: Gonçalves Jr, Paulo Mauricio, and Roberto Souto Maior De Barros. "RCD: A recurring concept drift framework." Pattern Recognition Letters 34.9 (2013): 1018-1025.

Author:

Paulo Goncalves (paulogoncalves at recife dot ifpe dot edu dot br)

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
protected int	bufferSize
IntOption	bufferSizeOption
IntOption	classifiersSizeOption
protected java.util.List<Instance>	currentChunk
protected java.util.List<Instance>	currentChunk2
protected int	index
protected int	previousState
IntOption	quantityClassifiersTestOption
FloatOption	similarityBetweenDistributionsOption
ClassOption	statisticalTestOption
protected java.util.List<Instance>	testChunk
IntOption	testFrequencyOption
IntOption	threadSizeOption

Fields inherited from class moa.classifiers.AbstractClassifier

classifierRandom, modelContext, randomSeed, randomSeedOption, trainingWeightSeenByModel

Fields inherited from class moa.options.AbstractOptionHandler

config

Fields inherited from class moa.classifiers.drift.DriftDetectionMethodClassifier

baseLearnerOption, changeDetected, classifier, DDM_INCONTROL_LEVEL, DDM_OUTCONTROL_LEVEL, DDM_WARNING_LEVEL, ddmLevel, driftDetectionMethod, driftDetectionMethodOption, newclassifier, newClassifierReset, warningDetected

Constructor Summary

Constructors

Constructor	Description
RCD()

Method Summary

Modifier and Type	Method	Description
double[]	getVotesForInstance(Instance inst)	Predicts the class memberships for a given instance.
void	resetLearningImpl()	Resets this classifier.
void	trainOnInstanceImpl(Instance inst)	Trains this classifier incrementally using the given instance. The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods.

Methods inherited from class moa.classifiers.AbstractClassifier

contextIsCompatible, copy, correctlyClassifies, getAttributeNameString, getAWTRenderer, getClassLabelString, getClassNameString, getDescription, getModel, getModelContext, getModelMeasurements, getNominalValueString, getPredictionForInstance, getPredictionForInstance, getSubClassifiers, getSublearners, getVotesForInstance, modelAttIndexToInstanceAttIndex, modelAttIndexToInstanceAttIndex, prepareForUseImpl, resetLearning, setModelContext, setRandomSeed, trainingHasStarted, trainingWeightSeenByModel, trainOnInstance, trainOnInstance

Methods inherited from class moa.AbstractMOAObject

copy, measureByteSize, measureByteSize, toString

Methods inherited from class moa.options.AbstractOptionHandler

getCLICreationString, getOptions, getPreparedClassOption, prepareClassOptions, prepareForUse, prepareForUse

Methods inherited from interface moa.capabilities.CapabilitiesHandler

getCapabilities

Methods inherited from class moa.classifiers.drift.DriftDetectionMethodClassifier

defineImmutableCapabilities, getModelDescription, getModelMeasurementsImpl, getPurposeString, isChangeDetected, isRandomizable, isWarningDetected

Methods inherited from interface moa.MOAObject

measureByteSize

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface moa.options.OptionHandler

getCLICreationString, getOptions, prepareForUse, prepareForUse

Field Detail

statisticalTestOption

public ClassOption statisticalTestOption

similarityBetweenDistributionsOption

public FloatOption similarityBetweenDistributionsOption

bufferSizeOption

public IntOption bufferSizeOption

testFrequencyOption

public IntOption testFrequencyOption

classifiersSizeOption

public IntOption classifiersSizeOption

threadSizeOption

public IntOption threadSizeOption

quantityClassifiersTestOption

public IntOption quantityClassifiersTestOption

currentChunk

protected java.util.List<Instance> currentChunk

currentChunk2

protected java.util.List<Instance> currentChunk2

testChunk

protected java.util.List<Instance> testChunk

bufferSize

protected int bufferSize

previousState

protected int previousState

index

protected int index

Constructor Detail

RCD

public RCD()

Method Detail

resetLearningImpl

public void resetLearningImpl()

Description copied from class: AbstractClassifier

Resets this classifier. It must be similar to starting a new classifier from scratch.

The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods. Note that this will produce compiler errors if not overridden.

Overrides:

resetLearningImpl in class DriftDetectionMethodClassifier

trainOnInstanceImpl

public void trainOnInstanceImpl(Instance inst)

Description copied from class: AbstractClassifier

Trains this classifier incrementally using the given instance.

The reason for ...Impl methods: ease programmer burden by not requiring them to remember calls to super in overridden methods. Note that this will produce compiler errors if not overridden.

Overrides:

trainOnInstanceImpl in class DriftDetectionMethodClassifier

Parameters:

inst - the instance to be used for training

getVotesForInstance

public double[] getVotesForInstance(Instance inst)

Description copied from interface: Classifier

Predicts the class memberships for a given instance. If an instance is unclassified, the returned array elements must be all zero.

Specified by:

getVotesForInstance in interface Classifier

Overrides:

getVotesForInstance in class DriftDetectionMethodClassifier

Parameters:

inst - the instance to be classified

Returns:

an array containing the estimated membership probabilities of the test instance in each class