Package elki.clustering.kmeans.initialization

Class KMC2.Instance

java.lang.Object

elki.clustering.kmeans.initialization.KMC2.Instance

Direct Known Subclasses:

AFKMC2.Instance

Enclosing class:

KMC2

protected static class KMC2.Instance
extends java.lang.Object

Abstract instance implementing the weight handling.

Author:

Erich Schubert

Field Summary

Fields

Modifier and Type	Field	Description
protected NumberVectorDistance<?>	distance	Distance function
protected long	diststat	Count the number of distance computations.
protected int	m	Number of sampling attempts.
protected java.util.Random	random	Random generator
protected Relation<? extends NumberVector>	relation	Data relation.
protected WritableDoubleDataStore	weights	Weights

Constructor Summary

Constructors

Constructor	Description
Instance(Relation<? extends NumberVector> relation, NumberVectorDistance<?> distance, int m, RandomFactory rnd)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected void	chooseRemaining(int k, java.util.List<NumberVector> means, double weightsum)	Choose remaining means, weighted by distance.
protected double	distance(NumberVector a, DBIDRef b)	Compute the distance of two objects.
protected double	distance(DBIDRef cand, java.util.List<NumberVector> means)	Minimum distance to the current means.
protected Logging	getLogger()	Class logger.
protected double	initialWeights(NumberVector first)	Initialize the weight list.
double[][]	run(int k)	Run k-means++ initialization for number vectors.
protected DBIDRef	sample(double weightsum)	Weighted sampling.

Methods inherited from class java.lang.Object

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

Field Detail

relation

protected Relation<? extends NumberVector> relation

Data relation.

distance

protected NumberVectorDistance<?> distance

Distance function

weights

protected WritableDoubleDataStore weights

Weights

diststat

protected long diststat

Count the number of distance computations.

m

protected int m

Number of sampling attempts.

random

protected java.util.Random random

Random generator

Constructor Detail

Instance

public Instance(Relation<? extends NumberVector> relation,
                NumberVectorDistance<?> distance,
                int m,
                RandomFactory rnd)

Constructor.

Parameters:

relation - Data relation to process

distance - Distance function

m - M parameter

rnd - Random generator

Method Detail

initialWeights

protected double initialWeights(NumberVector first)

Initialize the weight list.

Parameters:

first - Added ID

Returns:

Weight sum

run

public double[][] run(int k)

Run k-means++ initialization for number vectors.

Parameters:

k - K

Returns:

Vectors

distance

protected double distance(NumberVector a,
                          DBIDRef b)

Compute the distance of two objects.

Parameters:

a - First object

b - Second object

Returns:

Distance

chooseRemaining

protected void chooseRemaining(int k,
                               java.util.List<NumberVector> means,
                               double weightsum)

Choose remaining means, weighted by distance.

Parameters:

k - Number of means to choose

means - Means storage

weightsum - Sum of weights

sample

protected DBIDRef sample(double weightsum)

Weighted sampling.

Parameters:

weightsum - Total weight sum.

Returns:

Sampled object

distance

protected double distance(DBIDRef cand,
                          java.util.List<NumberVector> means)

Minimum distance to the current means.

Parameters:

cand - Candidate

means - Current means

Returns:

Minimum distance

getLogger

protected Logging getLogger()

Class logger.

Returns:

Logger