jsat.clustering

Interface KClusterer

All Superinterfaces:

Clusterer, Serializable

All Known Implementing Classes:

CLARA, DivisiveGlobalClusterer, DivisiveLocalClusterer, ElkanKernelKMeans, ElkanKMeans, EMGaussianMixture, GapStatistic, GMeans, HamerlyKMeans, KClustererBase, KernelKMeans, KMeans, KMeansPDN, LloydKernelKMeans, MiniBatchKMeans, NaiveKMeans, NNChainHAC, PAM, PriorityHAC, SimpleHAC, XMeans

public interface KClusterer
extends Clusterer

Defines a clustering method that requires the number of clusters in the data set to be known before hand.

Author:

Edward Raff

Method Summary

Modifier and Type	Method and Description
KClusterer	clone()
List<List<DataPoint>>	cluster(DataSet dataSet, int clusters) Performs clustering on the given data set.
List<List<DataPoint>>	cluster(DataSet dataSet, int clusters, ExecutorService threadpool) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int clusters, ExecutorService threadpool, int[] designations)
List<List<DataPoint>>	cluster(DataSet dataSet, int lowK, int highK) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int clusters, int[] designations)
List<List<DataPoint>>	cluster(DataSet dataSet, int lowK, int highK, ExecutorService threadpool) Performs clustering on the given data set.
int[]	cluster(DataSet dataSet, int lowK, int highK, ExecutorService threadpool, int[] designations)
int[]	cluster(DataSet dataSet, int lowK, int highK, int[] designations)

Methods inherited from interface jsat.clustering.Clusterer

cluster, cluster, cluster, cluster, supportsWeightedData

Method Detail

cluster

List<List<DataPoint>> cluster(DataSet dataSet,
                              int clusters,
                              ExecutorService threadpool)

Performs clustering on the given data set.

Parameters:

dataSet - the data points to perform clustering on

clusters - the number of clusters to assume

threadpool - a source of threads to run tasks

Returns:

A list of DataSets, where each DataSet contains the data points for one cluster in the group

cluster

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

cluster

List<List<DataPoint>> cluster(DataSet dataSet,
                              int clusters)

Performs clustering on the given data set.

Parameters:

dataSet - the data points to perform clustering on

clusters - the number of clusters to assume

Returns:

A list of DataSets, where each DataSet contains the data points for one cluster in the group

cluster

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

cluster

List<List<DataPoint>> cluster(DataSet dataSet,
                              int lowK,
                              int highK,
                              ExecutorService threadpool)

Performs clustering on the given data set. The implementation will attempt to determine the best number of clusters for the given data.

Parameters:

dataSet - the data points to perform clustering on

lowK - the lower bound, inclusive, of the range to search

highK - the upper bound, inclusive, of the range to search

threadpool - a source of threads to run tasks

Returns:

A list of DataSets, where each DataSet contains the data points for one cluster in the group

cluster

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

cluster

List<List<DataPoint>> cluster(DataSet dataSet,
                              int lowK,
                              int highK)

Performs clustering on the given data set. The implementation will attempt to determine the best number of clusters for the given data.

Parameters:

dataSet - the data points to perform clustering on

lowK - the lower bound, inclusive, of the range to search

highK - the upper bound, inclusive, of the range to search

Returns:

A list of DataSets, where each DataSet contains the data points for one cluster in the group

cluster

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

clone

KClusterer clone()

Specified by:

clone in interface Clusterer