Package elki.clustering.kmeans.parallel

Class KMeansProcessor.Instance<V extends NumberVector>

java.lang.Object

elki.clustering.kmeans.parallel.KMeansProcessor.Instance<V>

All Implemented Interfaces:

Processor.Instance

Enclosing class:

KMeansProcessor<V extends NumberVector>

public static class KMeansProcessor.Instance<V extends NumberVector>
extends java.lang.Object
implements Processor.Instance

Instance to process part of the data set, for a single iteration.

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field	Description
private WritableIntegerDataStore	assignment	Cluster assignment storage.
private double[][]	centroids	Updated cluster centroids
private boolean	changed	Changed flag.
private NumberVectorDistance<? super V>	distance	Distance function.
private double[][]	means	Current mean vectors.
private Relation<V>	relation	Data relation.
private int[]	sizes	(Partial) cluster sizes
private double[]	varsum	Variance sum.

Constructor Summary

Constructors

Constructor	Description
Instance(Relation<V> relation, NumberVectorDistance<? super V> distance, WritableIntegerDataStore assignment, double[][] means)	Constructor.

Method Summary

Modifier and Type	Method	Description
void	map(DBIDRef id)	Process ("map") a single object

Methods inherited from class java.lang.Object

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

Field Detail

relation

private Relation<V extends NumberVector> relation

Data relation.

distance

private NumberVectorDistance<? super V extends NumberVector> distance

Distance function.

assignment

private WritableIntegerDataStore assignment

Cluster assignment storage.

means

private double[][] means

Current mean vectors.

centroids

private double[][] centroids

Updated cluster centroids

sizes

private int[] sizes

(Partial) cluster sizes

varsum

private double[] varsum

Variance sum.

changed

private boolean changed

Changed flag.

Constructor Detail

Instance

public Instance(Relation<V> relation,
                NumberVectorDistance<? super V> distance,
                WritableIntegerDataStore assignment,
                double[][] means)

Constructor.

Parameters:

relation - Data relation

distance - Distance function

assignment - Current assignment

means - Previous mean vectors

Method Detail

map

public void map(DBIDRef id)

Description copied from interface: Processor.Instance

Process ("map") a single object

Specified by:

map in interface Processor.Instance

Parameters:

id - Object to map.