Package elki.clustering.kmeans.spherical

Class SphericalHamerlyKMeans.Instance

java.lang.Object

elki.clustering.kmeans.AbstractKMeans.Instance

elki.clustering.kmeans.spherical.SphericalKMeans.Instance

elki.clustering.kmeans.spherical.SphericalHamerlyKMeans.Instance

Enclosing class:

SphericalHamerlyKMeans<V extends NumberVector>

protected static class SphericalHamerlyKMeans.Instance
extends SphericalKMeans.Instance

Inner instance, storing state for a single data set.

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field	Description
(package private) double[]	csim	Cluster self-similarity.
(package private) WritableDoubleDataStore	lsim	Similarity lower bound.
(package private) double[][]	newmeans	Scratch space for new means.
(package private) double[][]	sums	Sum aggregate for the new mean.
(package private) WritableDoubleDataStore	usim	Similarity upper bound.

Fields inherited from class elki.clustering.kmeans.AbstractKMeans.Instance

assignment, clusters, diststat, isSquared, k, key, means, relation, varsum

Constructor Summary

Constructors

Constructor	Description
Instance(Relation<? extends NumberVector> relation, double[][] means)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected int	assignToNearestCluster()	Assign each object to the nearest cluster.
protected Logging	getLogger()	Get the class logger.
protected int	initialAssignToNearestCluster()	Perform initial cluster assignment.
int	iterate(int iteration)	Main loop function.
protected void	recomputeSeperation(double[] csim)	Recompute the separation of cluster means.
protected void	updateBounds(double[] msim)	Update the bounds for k-means.

Methods inherited from class elki.clustering.kmeans.spherical.SphericalKMeans.Instance

distance, distance, distance, initialSeparation, means, meansFromSums, movedSimilarity, recomputeVariance, similarity, similarity, sqrtdistance, sqrtdistance

Methods inherited from class elki.clustering.kmeans.AbstractKMeans.Instance

buildResult, buildResult, computeSquaredSeparation, copyMeans, initialSeperation, movedDistance, recomputeSeperation, run, sqrtdistance

Methods inherited from class java.lang.Object

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

Field Detail

sums

double[][] sums

Sum aggregate for the new mean.

newmeans

double[][] newmeans

Scratch space for new means.

lsim

WritableDoubleDataStore lsim

Similarity lower bound.

usim

WritableDoubleDataStore usim

Similarity upper bound.

csim

double[] csim

Cluster self-similarity.

Constructor Detail

Instance

public Instance(Relation<? extends NumberVector> relation,
                double[][] means)

Constructor.

Parameters:

relation - Relation

means - Initial means

Method Detail

iterate

public int iterate(int iteration)

Description copied from class: AbstractKMeans.Instance

Main loop function.

Overrides:

iterate in class SphericalKMeans.Instance

Parameters:

iteration - Iteration number (beginning at 1)

Returns:

Number of reassigned points

initialAssignToNearestCluster

protected int initialAssignToNearestCluster()

Perform initial cluster assignment.

Returns:

Number of changes (i.e., relation size)

recomputeSeperation

protected void recomputeSeperation(double[] csim)

Recompute the separation of cluster means.

Used by Hamerly.

Parameters:

csim - Output array of separation

assignToNearestCluster

protected int assignToNearestCluster()

Description copied from class: AbstractKMeans.Instance

Assign each object to the nearest cluster.

Overrides:

assignToNearestCluster in class SphericalKMeans.Instance

Returns:

number of objects reassigned

updateBounds

protected void updateBounds(double[] msim)

Update the bounds for k-means.

Parameters:

msim - Similarity movement of centers

getLogger

protected Logging getLogger()

Description copied from class: AbstractKMeans.Instance

Get the class logger.

Overrides:

getLogger in class SphericalKMeans.Instance

Returns:

Logger