Modifier and Type	Class and Description
public static enum	KMeansPlusPlusClusterer.EmptyClusterStrategy Strategies to use for replacing an empty cluster.

Modifier and Type	Field and Description
private final KMeansPlusPlusClusterer.EmptyClusterStrategy	emptyStrategy Selected strategy for empty clusters.
private final int	k The number of clusters.
private final int	maxIterations The maximum number of iterations.
private final RandomGenerator	random Random generator for choosing initial centers.

Constructor Summary

Access	Constructor and Description
public	KMeansPlusPlusClusterer(final int the number of clusters to split the data into k) Build a clusterer.
public	KMeansPlusPlusClusterer(final int the number of clusters to split the data into k, final int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. maxIterations) Build a clusterer.
public	KMeansPlusPlusClusterer(final int the number of clusters to split the data into k, final int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. maxIterations, final DistanceMeasure the distance measure to use measure) Build a clusterer.
public	KMeansPlusPlusClusterer(final int the number of clusters to split the data into k, final int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. maxIterations, final DistanceMeasure the distance measure to use measure, final RandomGenerator random generator to use for choosing initial centers random) Build a clusterer.
public	KMeansPlusPlusClusterer(final int the number of clusters to split the data into k, final int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. maxIterations, final DistanceMeasure the distance measure to use measure, final RandomGenerator random generator to use for choosing initial centers random, final KMeansPlusPlusClusterer.EmptyClusterStrategy strategy to use for handling empty clusters that may appear during algorithm iterations emptyStrategy) Build a clusterer.

Method Summary

Modifier and Type	Method and Description
private int	Returns: the number of points assigned to different clusters as the iteration before assignPointsToClusters(final List<CentroidCluster<T>> the `Cluster`s to add the points to clusters, final Collection<T> the points to add to the given `Cluster`s points, final int[] points assignments to clusters assignments) Adds the given points to the closest `Cluster`.
private Clusterable	Returns: the computed centroid for the set of points centroidOf(final Collection<T> the set of points points, final int the point dimension dimension) Computes the centroid for a set of points.
private List<CentroidCluster<T>>	Returns: the initial centers chooseInitialCenters(final Collection<T> the points to choose the initial centers from points) Use K-means++ to choose the initial centers.
public List<CentroidCluster<T>>	Returns: a list of clusters containing the points cluster(final Collection<T> the points to cluster points) Implements abstract org.apache.commons.math3.ml.clustering.Clusterer.cluster. Runs the K-means++ clustering algorithm.
public KMeansPlusPlusClusterer.EmptyClusterStrategy	Returns: the `EmptyClusterStrategy` getEmptyClusterStrategy() Returns the `EmptyClusterStrategy` used by this instance.
private T	Returns: point farthest to its cluster center getFarthestPoint(final Collection<CentroidCluster<T>> the `Cluster`s to search clusters) Get the point farthest to its cluster center
public int	Returns: the number of clusters getK() Return the number of clusters this instance will use.
public int	Returns: the maximum number of iterations, or -1 if no maximum is set getMaxIterations() Returns the maximum number of iterations this instance will use.
private int	Returns: the index of the nearest `Cluster` to the given point getNearestCluster(final Collection<CentroidCluster<T>> the `Cluster`s to search clusters, final T the point to find the nearest `Cluster` for point) Returns the nearest `Cluster` to the given point
private T	Returns: a random point from the selected cluster getPointFromLargestNumberCluster(final Collection<? extends Cluster<T>> the `Cluster`s to search clusters) Get a random point from the `Cluster` with the largest number of points
private T	Returns: a random point from the selected cluster getPointFromLargestVarianceCluster(final Collection<CentroidCluster<T>> the `Cluster`s to search clusters) Get a random point from the `Cluster` with the largest distance variance.
public RandomGenerator	Returns: the random generator getRandomGenerator() Returns the random generator this instance will use.

Inherited from org.apache.commons.math3.ml.clustering.Clusterer:: distance getDistanceMeasure

Field Detail

emptyStrategy	back to summary
private final KMeansPlusPlusClusterer.EmptyClusterStrategy emptyStrategy Selected strategy for empty clusters.

k	back to summary
private final int k The number of clusters.

maxIterations	back to summary
private final int maxIterations The maximum number of iterations.

random	back to summary
private final RandomGenerator random Random generator for choosing initial centers.

Constructor Detail

KMeansPlusPlusClusterer	back to summary
public KMeansPlusPlusClusterer(final int k) Build a clusterer. The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance. The euclidean distance will be used as default distance measure. Parameters k:int the number of clusters to split the data into

KMeansPlusPlusClusterer

back to summary

public KMeansPlusPlusClusterer(final int k)

Build a clusterer.

The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance.

The euclidean distance will be used as default distance measure.

Parameters

k:int: the number of clusters to split the data into

KMeansPlusPlusClusterer	back to summary
public KMeansPlusPlusClusterer(final int k, final int maxIterations) Build a clusterer. The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance. The euclidean distance will be used as default distance measure. Parameters k:int the number of clusters to split the data into maxIterations:int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used.

KMeansPlusPlusClusterer

back to summary

public KMeansPlusPlusClusterer(final int k, final int maxIterations)

Build a clusterer.

The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance.

The euclidean distance will be used as default distance measure.

Parameters

k:int: the number of clusters to split the data into
maxIterations:int: the maximum number of iterations to run the algorithm for. If negative, no maximum will be used.

KMeansPlusPlusClusterer	back to summary
public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure) Build a clusterer. The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance. Parameters k:int the number of clusters to split the data into maxIterations:int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. measure:DistanceMeasure the distance measure to use

KMeansPlusPlusClusterer

back to summary

public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure)

Build a clusterer.

The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance.

Parameters

k:int: the number of clusters to split the data into
maxIterations:int: the maximum number of iterations to run the algorithm for. If negative, no maximum will be used.
measure:DistanceMeasure: the distance measure to use

KMeansPlusPlusClusterer	back to summary
public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure, final RandomGenerator random) Build a clusterer. The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance. Parameters k:int the number of clusters to split the data into maxIterations:int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. measure:DistanceMeasure the distance measure to use random:RandomGenerator random generator to use for choosing initial centers

KMeansPlusPlusClusterer

back to summary

public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure, final RandomGenerator random)

Build a clusterer.

The default strategy for handling empty clusters that may appear during algorithm iterations is to split the cluster with largest distance variance.

Parameters

k:int: the number of clusters to split the data into
maxIterations:int: the maximum number of iterations to run the algorithm for. If negative, no maximum will be used.
measure:DistanceMeasure: the distance measure to use
random:RandomGenerator: random generator to use for choosing initial centers

KMeansPlusPlusClusterer	back to summary
public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure, final RandomGenerator random, final KMeansPlusPlusClusterer.EmptyClusterStrategy emptyStrategy) Build a clusterer. Parameters k:int the number of clusters to split the data into maxIterations:int the maximum number of iterations to run the algorithm for. If negative, no maximum will be used. measure:DistanceMeasure the distance measure to use random:RandomGenerator random generator to use for choosing initial centers emptyStrategy:KMeansPlusPlusClusterer.EmptyClusterStrategy strategy to use for handling empty clusters that may appear during algorithm iterations

KMeansPlusPlusClusterer

back to summary

public KMeansPlusPlusClusterer(final int k, final int maxIterations, final DistanceMeasure measure, final RandomGenerator random, final KMeansPlusPlusClusterer.EmptyClusterStrategy emptyStrategy)

Build a clusterer.

Parameters

k:int: the number of clusters to split the data into
maxIterations:int: the maximum number of iterations to run the algorithm for. If negative, no maximum will be used.
measure:DistanceMeasure: the distance measure to use
random:RandomGenerator: random generator to use for choosing initial centers
emptyStrategy:KMeansPlusPlusClusterer.EmptyClusterStrategy: strategy to use for handling empty clusters that may appear during algorithm iterations

Method Detail

assignPointsToClusters back to summary

assignPointsToClusters	back to summary
private int assignPointsToClusters(final List<CentroidCluster<T>> clusters, final Collection<T> points, final int[] assignments) Adds the given points to the closest `Cluster`. Parameters clusters:List<CentroidCluster<T>> the `Cluster`s to add the points to points:Collection<T> the points to add to the given `Cluster`s assignments:int[] points assignments to clusters Returns:int the number of points assigned to different clusters as the iteration before

private int assignPointsToClusters(final List<CentroidCluster<T>> clusters, final Collection<T> points, final int[] assignments)

Adds the given points to the closest Cluster.

Parameters

clusters:List<CentroidCluster<T>>: the Clusters to add the points to
points:Collection<T>: the points to add to the given Clusters
assignments:int[]: points assignments to clusters

Returns:int

the number of points assigned to different clusters as the iteration before

centroidOf	back to summary
private Clusterable centroidOf(final Collection<T> points, final int dimension) Computes the centroid for a set of points. Parameters points:Collection<T> the set of points dimension:int the point dimension Returns:Clusterable the computed centroid for the set of points

centroidOf

back to summary

private Clusterable centroidOf(final Collection<T> points, final int dimension)

Computes the centroid for a set of points.

Parameters

points:Collection<T>: the set of points
dimension:int: the point dimension

Returns:Clusterable

the computed centroid for the set of points

chooseInitialCenters	back to summary
private List<CentroidCluster<T>> chooseInitialCenters(final Collection<T> points) Use K-means++ to choose the initial centers. Parameters points:Collection<T> the points to choose the initial centers from Returns:List<CentroidCluster<T>> the initial centers

chooseInitialCenters

back to summary

private List<CentroidCluster<T>> chooseInitialCenters(final Collection<T> points)

Use K-means++ to choose the initial centers.

Parameters

points:Collection<T>: the points to choose the initial centers from

Returns:List<CentroidCluster<T>>

the initial centers

cluster back to summary

cluster	back to summary
public List<CentroidCluster<T>> cluster(final Collection<T> points) throws MathIllegalArgumentException, ConvergenceException Implements abstract org.apache.commons.math3.ml.clustering.Clusterer.cluster. Runs the K-means++ clustering algorithm. Parameters points:Collection<T> the points to cluster Returns:List<CentroidCluster<T>> a list of clusters containing the points Annotations @Override Exceptions 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 `emptyStrategy` is set to `ERROR`

public List<CentroidCluster<T>> cluster(final Collection<T> points) throws MathIllegalArgumentException, ConvergenceException

Implements abstract org.apache.commons.math3.ml.clustering.Clusterer.cluster.

Runs the K-means++ clustering algorithm.

Parameters

points:Collection<T>: the points to cluster

Returns:List<CentroidCluster<T>>

a list of clusters containing the points

Annotations

@Override

Exceptions

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 emptyStrategy is set to ERROR

getEmptyClusterStrategy back to summary

getEmptyClusterStrategy	back to summary
public KMeansPlusPlusClusterer.EmptyClusterStrategy getEmptyClusterStrategy() Returns the `EmptyClusterStrategy` used by this instance. Returns:KMeansPlusPlusClusterer.EmptyClusterStrategy the `EmptyClusterStrategy`

public KMeansPlusPlusClusterer.EmptyClusterStrategy getEmptyClusterStrategy()

Returns the EmptyClusterStrategy used by this instance.

Returns:KMeansPlusPlusClusterer.EmptyClusterStrategy: the EmptyClusterStrategy

getFarthestPoint back to summary

getFarthestPoint	back to summary
private T getFarthestPoint(final Collection<CentroidCluster<T>> clusters) throws ConvergenceException Get the point farthest to its cluster center Parameters clusters:Collection<CentroidCluster<T>> the `Cluster`s to search Returns:T point farthest to its cluster center Exceptions ConvergenceException: if clusters are all empty

private T getFarthestPoint(final Collection<CentroidCluster<T>> clusters) throws ConvergenceException

Get the point farthest to its cluster center

Parameters

clusters:Collection<CentroidCluster<T>>: the Clusters to search

Returns:T

point farthest to its cluster center

Exceptions

ConvergenceException:: if clusters are all empty

getK	back to summary
public int getK() Return the number of clusters this instance will use. Returns:int the number of clusters

getK

back to summary

public int getK()

Return the number of clusters this instance will use.

Returns:int: the number of clusters

getMaxIterations	back to summary
public int getMaxIterations() Returns the maximum number of iterations this instance will use. Returns:int the maximum number of iterations, or -1 if no maximum is set

getMaxIterations

back to summary

public int getMaxIterations()

Returns the maximum number of iterations this instance will use.

Returns:int: the maximum number of iterations, or -1 if no maximum is set

getNearestCluster back to summary

getNearestCluster	back to summary
private int getNearestCluster(final Collection<CentroidCluster<T>> clusters, final T point) Returns the nearest `Cluster` to the given point Parameters clusters:Collection<CentroidCluster<T>> the `Cluster`s to search point:T the point to find the nearest `Cluster` for Returns:int the index of the nearest `Cluster` to the given point

private int getNearestCluster(final Collection<CentroidCluster<T>> clusters, final T point)

Returns the nearest Cluster to the given point

Parameters

clusters:Collection<CentroidCluster<T>>: the Clusters to search
point:T: the point to find the nearest Cluster for

Returns:int

the index of the nearest Cluster to the given point

getPointFromLargestNumberCluster back to summary

getPointFromLargestNumberCluster	back to summary
private T getPointFromLargestNumberCluster(final Collection<? extends Cluster<T>> clusters) throws ConvergenceException Get a random point from the `Cluster` with the largest number of points Parameters clusters:Collection<? extends Cluster<T>> the `Cluster`s to search Returns:T a random point from the selected cluster Exceptions ConvergenceException: if clusters are all empty

private T getPointFromLargestNumberCluster(final Collection<? extends Cluster<T>> clusters) throws ConvergenceException

Get a random point from the Cluster with the largest number of points

Parameters

clusters:Collection<? extends Cluster<T>>: the Clusters to search

Returns:T

a random point from the selected cluster

Exceptions

ConvergenceException:: if clusters are all empty

getPointFromLargestVarianceCluster back to summary

getPointFromLargestVarianceCluster	back to summary
private T getPointFromLargestVarianceCluster(final Collection<CentroidCluster<T>> clusters) throws ConvergenceException Get a random point from the `Cluster` with the largest distance variance. Parameters clusters:Collection<CentroidCluster<T>> the `Cluster`s to search Returns:T a random point from the selected cluster Exceptions ConvergenceException: if clusters are all empty

private T getPointFromLargestVarianceCluster(final Collection<CentroidCluster<T>> clusters) throws ConvergenceException

Get a random point from the Cluster with the largest distance variance.

Parameters

clusters:Collection<CentroidCluster<T>>: the Clusters to search

Returns:T

a random point from the selected cluster

Exceptions

ConvergenceException:: if clusters are all empty

getRandomGenerator	back to summary
public RandomGenerator getRandomGenerator() Returns the random generator this instance will use. Returns:RandomGenerator the random generator

getRandomGenerator

back to summary

public RandomGenerator getRandomGenerator()

Returns the random generator this instance will use.

Returns:RandomGenerator: the random generator

org.apache.commons.math3.ml.clustering back to summary

public final Enum KMeansPlusPlusClusterer.EmptyClusterStrategy

Class Inheritance

extends Enum<KMeansPlusPlusClusterer.EmptyClusterStrategy>

java.lang.Object
java.lang.Enum
org.apache.commons.math3.ml.clustering.KMeansPlusPlusClusterer.EmptyClusterStrategy

Strategies to use for replacing an empty cluster.

Field Summary

Modifier and Type	Field and Description
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy	ERROR Generate an error.
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy	FARTHEST_POINT Create a cluster around the point farthest from its centroid.
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy	LARGEST_POINTS_NUMBER Split the cluster with largest number of points.
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy	LARGEST_VARIANCE Split the cluster with largest distance variance.

Constructor Summary

Access	Constructor and Description
private	EmptyClusterStrategy()

Method Summary

Modifier and Type	Method and Description
public static KMeansPlusPlusClusterer.EmptyClusterStrategy	valueOf(String name)
public static KMeansPlusPlusClusterer.EmptyClusterStrategy[]	values()

Inherited from java.lang.Enum:: clone compareTo describeConstable equals finalize getDeclaringClass hashCode name ordinal toString valueOf

Field Detail

ERROR	back to summary
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy ERROR Generate an error.

FARTHEST_POINT	back to summary
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy FARTHEST_POINT Create a cluster around the point farthest from its centroid.

LARGEST_POINTS_NUMBER	back to summary
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy LARGEST_POINTS_NUMBER Split the cluster with largest number of points.

LARGEST_VARIANCE	back to summary
public static final KMeansPlusPlusClusterer.EmptyClusterStrategy LARGEST_VARIANCE Split the cluster with largest distance variance.

Constructor Detail

EmptyClusterStrategy	back to summary
private EmptyClusterStrategy()

Method Detail

valueOf	back to summary
public static KMeansPlusPlusClusterer.EmptyClusterStrategy valueOf(String name)

values	back to summary
public static KMeansPlusPlusClusterer.EmptyClusterStrategy[] values()

public Class KMeansPlusPlusClusterer<T extends Clusterable>

Nested and Inner Type Summary

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail

public final Enum KMeansPlusPlusClusterer.EmptyClusterStrategy

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail