Package elki.clustering.kmeans

Class KDTreeFilteringKMeans.Instance

java.lang.Object

elki.clustering.kmeans.AbstractKMeans.Instance

elki.clustering.kmeans.KDTreePruningKMeans.Instance

elki.clustering.kmeans.KDTreeFilteringKMeans.Instance

Enclosing class:

KDTreeFilteringKMeans<V extends NumberVector>

protected class KDTreeFilteringKMeans.Instance
extends KDTreePruningKMeans.Instance

Inner instance, storing state for a single data set.

Author:

Cedrik Lüdicke, Erich Schubert

Field Summary

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

clusterSizes, clusterSums, indices, iter, root, sorted

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, NumberVectorDistance<?> df, double[][] means)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected Logging	getLogger()	Get the class logger.
protected int	getNearestCenter(double[] mid, int alive)	Get the nearest (alive) center to a midpoint.
protected boolean	isFarther(double[] z_star, double[] z, double[] mid, double[] halfwidth)	Check if a cluster mean is farther than another.
protected int	pruning(KDTreePruningKMeans.KDNode u, int alive)	The pruning algorithm.

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

buildTreeBoundedMidpoint, buildTreeMedian, buildTreeMidpoint, buildTreeSSQ, getMinMaxDist, iterate, labelSubtree, mindist, run, traversal, traverseLeaf

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

assignToNearestCluster, buildResult, buildResult, computeSquaredSeparation, copyMeans, distance, distance, distance, initialSeperation, meansFromSums, movedDistance, recomputeSeperation, recomputeVariance, sqrtdistance, sqrtdistance, sqrtdistance

Methods inherited from class java.lang.Object

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

Constructor Detail

Instance

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

Constructor.

Parameters:

relation - Relation of data points

df - Distance function

means - Initial means

Method Detail

pruning

protected int pruning(KDTreePruningKMeans.KDNode u,
                      int alive)

Description copied from class: KDTreePruningKMeans.Instance

The pruning algorithm.

Overrides:

pruning in class KDTreePruningKMeans.Instance

Parameters:

u - Current node

alive - Range of alive means

Returns:

Updated range.

getNearestCenter

protected int getNearestCenter(double[] mid,
                               int alive)

Get the nearest (alive) center to a midpoint.

Parameters:

mid - midpoint

alive - Number of alive centers

Returns:

best center

isFarther

protected boolean isFarther(double[] z_star,
                            double[] z,
                            double[] mid,
                            double[] halfwidth)

Check if a cluster mean is farther than another. Optimized version of the comparison suggested by Kanungo.

getLogger

protected Logging getLogger()

Description copied from class: AbstractKMeans.Instance

Get the class logger.

Overrides:

getLogger in class KDTreePruningKMeans.Instance

Returns:

Logger