jsat.clustering

Class PAM

java.lang.Object

jsat.clustering.ClustererBase

jsat.clustering.KClustererBase

jsat.clustering.PAM

All Implemented Interfaces:

Serializable, Clusterer, KClusterer

Direct Known Subclasses:

CLARA

public class PAM
extends KClustererBase

Author:

Edward Raff

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected DistanceMetric	dm
protected int	iterLimit
protected int[]	medoids
protected Random	rand
protected int	repeats
protected SeedSelectionMethods.SeedSelection	seedSelection
protected boolean	storeMedoids

Constructor Summary

Constructors

Constructor and Description
PAM()
PAM(DistanceMetric dm)
PAM(DistanceMetric dm, Random rand)
PAM(DistanceMetric dm, Random rand, SeedSelectionMethods.SeedSelection seedSelection)
PAM(PAM toCopy) Copy constructor

Method Summary

Modifier and Type	Method and Description
PAM	clone()
protected double	cluster(DataSet data, boolean doInit, int[] medioids, int[] assignments, List<Double> cacheAccel) Performs the actual work of PAM.
int[]	cluster(DataSet dataSet, ExecutorService threadpool, int[] designations) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int[] designations) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int clusters, ExecutorService threadpool, int[] designations)
int[]	cluster(DataSet dataSet, int clusters, int[] designations)
int[]	cluster(DataSet dataSet, int lowK, int highK, ExecutorService threadpool, int[] designations)
int[]	cluster(DataSet dataSet, int lowK, int highK, int[] designations)
int	getMaxIterations()
int[]	getMedoids() Returns the raw array of indices that indicate which data point acted as the center for each cluster.
SeedSelectionMethods.SeedSelection	getSeedSelection()
void	setMaxIterations(int iterLimit)
void	setSeedSelection(SeedSelectionMethods.SeedSelection seedSelection) Sets the method of seed selection used by this algorithm
void	setStoreMedoids(boolean storeMedoids) If set to true the computed medoids will be stored after clustering is completed, and can then be retrieved using getMedoids().

Methods inherited from class jsat.clustering.KClustererBase

cluster, cluster, cluster, cluster

Methods inherited from class jsat.clustering.ClustererBase

cluster, cluster, createClusterListFromAssignmentArray, getDatapointsFromCluster, supportsWeightedData

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.clustering.Clusterer

cluster, cluster, supportsWeightedData

Field Detail

dm

protected DistanceMetric dm

rand

protected Random rand

seedSelection

protected SeedSelectionMethods.SeedSelection seedSelection

repeats

protected int repeats

iterLimit

protected int iterLimit

medoids

protected int[] medoids

storeMedoids

protected boolean storeMedoids

Constructor Detail

PAM

public PAM(DistanceMetric dm,
           Random rand,
           SeedSelectionMethods.SeedSelection seedSelection)

PAM

public PAM(DistanceMetric dm,
           Random rand)

PAM

public PAM(DistanceMetric dm)

PAM

public PAM()

PAM

public PAM(PAM toCopy)

Copy constructor

Parameters:

toCopy - the object to copy

Method Detail

setMaxIterations

public void setMaxIterations(int iterLimit)

Parameters:

iterLimit - the maximum number of iterations of the algorithm to perform

getMaxIterations

public int getMaxIterations()

Returns:

the maximum number of iterations of the algorithm to perform

setStoreMedoids

public void setStoreMedoids(boolean storeMedoids)

If set to true the computed medoids will be stored after clustering is completed, and can then be retrieved using getMedoids().

Parameters:

storeMedoids - true if the medoids should be stored for later, false to discard them once clustering is complete.

getMedoids

public int[] getMedoids()

Returns the raw array of indices that indicate which data point acted as the center for each cluster.

Returns:

the array of medeoid indices

setSeedSelection

public void setSeedSelection(SeedSelectionMethods.SeedSelection seedSelection)

Sets the method of seed selection used by this algorithm

Parameters:

seedSelection - the method of seed selection to used

getSeedSelection

public SeedSelectionMethods.SeedSelection getSeedSelection()

Returns:

the method of seed selection used by this algorithm

cluster

protected double cluster(DataSet data,
                         boolean doInit,
                         int[] medioids,
                         int[] assignments,
                         List<Double> cacheAccel)

Performs the actual work of PAM.

Parameters:

data - the data set to apply PAM to

doInit - true if the initialization procedure of training the distance metric, initiating its cache, and selecting he seeds, should be done.

medioids - the array to store the indices that get chosen as the medoids. The length of the array indicates how many medoids should be obtained.

assignments - an array of the same length as data, each value indicating what cluster that point belongs to.

cacheAccel - the pre-computed distance acceleration cache. May be null.

Returns:

the sum of the squared distance from each point to its closest medoids

cluster

public int[] cluster(DataSet dataSet,
                     int[] designations)

Description copied from interface: Clusterer

Performs clustering on the given data set. Parameters may be estimated by the method, or other heuristics performed.

Parameters:

dataSet - the data set to perform clustering on

designations - the array which will contain the designated values. The array will be altered and returned by the function. If null is given, a new array will be created and returned.

Returns:

an array indicating for each value indicating the cluster designation. This is the same array as designations, or a new one if the input array was null

cluster

public int[] cluster(DataSet dataSet,
                     ExecutorService threadpool,
                     int[] designations)

Description copied from interface: Clusterer

Performs clustering on the given data set. Parameters may be estimated by the method, or other heuristics performed.

Parameters:

dataSet - the data set to perform clustering on

threadpool - a source of threads to run tasks

designations - the array which will contain the designated values. The array will be altered and returned by the function. If null is given, a new array will be created and returned.

Returns:

an array indicating for each value indicating the cluster designation. This is the same array as designations, or a new one if the input array was null

cluster

public int[] cluster(DataSet dataSet,
                     int clusters,
                     ExecutorService threadpool,
                     int[] designations)

cluster

public int[] cluster(DataSet dataSet,
                     int clusters,
                     int[] designations)

cluster

public int[] cluster(DataSet dataSet,
                     int lowK,
                     int highK,
                     int[] designations)

clone

public PAM clone()

Specified by:

clone in interface Clusterer

Specified by:

clone in interface KClusterer

Specified by:

clone in class KClustererBase

cluster

public int[] cluster(DataSet dataSet,
                     int lowK,
                     int highK,
                     ExecutorService threadpool,
                     int[] designations)