org.apache.commons.math3.ml.clustering

Class MultiKMeansPlusPlusClusterer<T extends Clusterable>

java.lang.Object

org.apache.commons.math3.ml.clustering.Clusterer<T>

org.apache.commons.math3.ml.clustering.MultiKMeansPlusPlusClusterer<T>

Type Parameters:

T - type of the points to cluster

public class MultiKMeansPlusPlusClusterer<T extends Clusterable>
extends Clusterer<T>

A wrapper around a k-means++ clustering algorithm which performs multiple trials and returns the best solution.

Since:

3.2

Field Summary

Fields

Modifier and Type	Field and Description
private KMeansPlusPlusClusterer<T>	clusterer The underlying k-means clusterer.
private ClusterEvaluator<T>	evaluator The cluster evaluator to use.
private int	numTrials The number of trial runs.

Constructor Summary

Constructors

Constructor and Description
MultiKMeansPlusPlusClusterer(KMeansPlusPlusClusterer<T> clusterer, int numTrials) Build a clusterer.
MultiKMeansPlusPlusClusterer(KMeansPlusPlusClusterer<T> clusterer, int numTrials, ClusterEvaluator<T> evaluator) Build a clusterer.

Method Summary

Methods

Modifier and Type	Method and Description
java.util.List<CentroidCluster<T>>	cluster(java.util.Collection<T> points) Runs the K-means++ clustering algorithm.
KMeansPlusPlusClusterer<T>	getClusterer() Returns the embedded k-means clusterer used by this instance.
ClusterEvaluator<T>	getClusterEvaluator() Returns the ClusterEvaluator used to determine the "best" clustering.
int	getNumTrials() Returns the number of trials this instance will do.

Methods inherited from class org.apache.commons.math3.ml.clustering.Clusterer

distance, getDistanceMeasure

Methods inherited from class java.lang.Object

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

Field Detail

clusterer

private final KMeansPlusPlusClusterer<T extends Clusterable> clusterer

The underlying k-means clusterer.

numTrials

private final int numTrials

The number of trial runs.

evaluator

private final ClusterEvaluator<T extends Clusterable> evaluator

The cluster evaluator to use.

Constructor Detail

MultiKMeansPlusPlusClusterer

public MultiKMeansPlusPlusClusterer(KMeansPlusPlusClusterer<T> clusterer,
                            int numTrials)

Build a clusterer.

Parameters:

clusterer - the k-means clusterer to use

numTrials - number of trial runs

MultiKMeansPlusPlusClusterer

public MultiKMeansPlusPlusClusterer(KMeansPlusPlusClusterer<T> clusterer,
                            int numTrials,
                            ClusterEvaluator<T> evaluator)

Build a clusterer.

Parameters:

clusterer - the k-means clusterer to use

numTrials - number of trial runs

evaluator - the cluster evaluator to use

Since:

3.3

Method Detail

getClusterer

public KMeansPlusPlusClusterer<T> getClusterer()

Returns the embedded k-means clusterer used by this instance.

Returns:

the embedded clusterer

getNumTrials

public int getNumTrials()

Returns the number of trials this instance will do.

Returns:

the number of trials

getClusterEvaluator

public ClusterEvaluator<T> getClusterEvaluator()

Returns the ClusterEvaluator used to determine the "best" clustering.

Returns:

the used ClusterEvaluator

Since:

3.3

cluster

public java.util.List<CentroidCluster<T>> cluster(java.util.Collection<T> points)
                                                               throws MathIllegalArgumentException,
                                                                      ConvergenceException

Runs the K-means++ clustering algorithm.

Specified by:

cluster in class Clusterer<T extends Clusterable>

Parameters:

points - the points to cluster

Returns:

a list of clusters containing the points

Throws:

MathIllegalArgumentException - if the data points are null or the number of clusters is larger than the number of data points

ConvergenceException - if an empty cluster is encountered and the underlying KMeansPlusPlusClusterer has its KMeansPlusPlusClusterer.EmptyClusterStrategy is set to ERROR.