com.yahoo.labs.samoa.moa.clusterers.clustream

Class WithKmeans

java.lang.Object

com.yahoo.labs.samoa.moa.AbstractMOAObject

com.yahoo.labs.samoa.moa.options.AbstractOptionHandler

com.yahoo.labs.samoa.moa.clusterers.AbstractClusterer

com.yahoo.labs.samoa.moa.clusterers.clustream.WithKmeans

All Implemented Interfaces:

com.github.javacliparser.Configurable, Clusterer, MOAObject, OptionHandler, Serializable

public class WithKmeans
extends AbstractClusterer

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
com.github.javacliparser.IntOption	kernelRadiFactorOption
com.github.javacliparser.IntOption	kOption
com.github.javacliparser.IntOption	maxNumKernelsOption
com.github.javacliparser.IntOption	timeWindowOption

Fields inherited from class com.yahoo.labs.samoa.moa.clusterers.AbstractClusterer

evaluateMicroClusteringOption

Constructor Summary

Constructors

Constructor and Description
WithKmeans()

Method Summary

Methods

Modifier and Type	Method and Description
Clustering	getClusteringResult()
Clustering	getClusteringResult(Clustering gtClustering)
Clustering	getMicroClusteringResult()
void	getModelDescription(StringBuilder out, int indent)
String	getName()
double[]	getVotesForInstance(Instance inst)
boolean	implementsMicroClusterer() Miscellaneous
boolean	isRandomizable()
static Clustering	kMeans_gta(int k, Clustering clustering, Clustering gtClustering) k-means of (micro)clusters, with ground-truth-aided initialization.
static Clustering	kMeans_rand(int k, Clustering clustering) k-means of (micro)clusters, with randomized initialization.
void	resetLearningImpl()
void	trainOnInstanceImpl(Instance instance)

Methods inherited from class com.yahoo.labs.samoa.moa.clusterers.AbstractClusterer

contextIsCompatible, copy, getAttributeNameString, getClassLabelString, getClassNameString, getDescription, getModelContext, getModelMeasurements, getNominalValueString, getPurposeString, getSubClusterers, keepClassLabel, prepareForUseImpl, resetLearning, setModelContext, setRandomSeed, trainingHasStarted, trainingWeightSeenByModel, trainOnInstance

Methods inherited from class com.yahoo.labs.samoa.moa.options.AbstractOptionHandler

getCLICreationString, getOptions, prepareForUse, prepareForUse

Methods inherited from class com.yahoo.labs.samoa.moa.AbstractMOAObject

copy, measureByteSize, measureByteSize, toString

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface com.yahoo.labs.samoa.moa.options.OptionHandler

getCLICreationString, getOptions, prepareForUse, prepareForUse

Methods inherited from interface com.yahoo.labs.samoa.moa.MOAObject

measureByteSize

Field Detail

timeWindowOption

public com.github.javacliparser.IntOption timeWindowOption

maxNumKernelsOption

public com.github.javacliparser.IntOption maxNumKernelsOption

kernelRadiFactorOption

public com.github.javacliparser.IntOption kernelRadiFactorOption

kOption

public com.github.javacliparser.IntOption kOption

Constructor Detail

WithKmeans

public WithKmeans()

Method Detail

resetLearningImpl

public void resetLearningImpl()

Specified by:

resetLearningImpl in class AbstractClusterer

trainOnInstanceImpl

public void trainOnInstanceImpl(Instance instance)

Specified by:

trainOnInstanceImpl in class AbstractClusterer

getMicroClusteringResult

public Clustering getMicroClusteringResult()

Specified by:

getMicroClusteringResult in interface Clusterer

Overrides:

getMicroClusteringResult in class AbstractClusterer

getClusteringResult

public Clustering getClusteringResult()

getClusteringResult

public Clustering getClusteringResult(Clustering gtClustering)

getName

public String getName()

kMeans_gta

public static Clustering kMeans_gta(int k,
                    Clustering clustering,
                    Clustering gtClustering)

k-means of (micro)clusters, with ground-truth-aided initialization. (to produce best results)

Parameters:

k -

data -

Returns:

(macro)clustering - CFClusters

kMeans_rand

public static Clustering kMeans_rand(int k,
                     Clustering clustering)

k-means of (micro)clusters, with randomized initialization.

Parameters:

k -

data -

Returns:

(macro)clustering - CFClusters

implementsMicroClusterer

public boolean implementsMicroClusterer()

Miscellaneous

Specified by:

implementsMicroClusterer in interface Clusterer

Overrides:

implementsMicroClusterer in class AbstractClusterer

isRandomizable

public boolean isRandomizable()

getVotesForInstance

public double[] getVotesForInstance(Instance inst)

getModelDescription

public void getModelDescription(StringBuilder out,
                       int indent)

Specified by:

getModelDescription in class AbstractClusterer