net.sourceforge.javaocr.plugin.cluster
Class AbstractBaseCluster

java.lang.Object
  net.sourceforge.javaocr.plugin.cluster.AbstractBaseCluster

All Implemented Interfaces:

Metric, Cluster

Direct Known Subclasses:

EuclidianDistanceCluster, MahalanobisDistanceCluster

public abstract class AbstractBaseCluster
extends Object
implements Cluster

base cluster functionality. provides mx

Author:

Konstantin Pribluda

Field Summary

private int	amountSamples
private int	dimensions
private double[]	mx
private double[]	sum

Constructor Summary

	AbstractBaseCluster()
protected	AbstractBaseCluster(double[] mx)
protected	AbstractBaseCluster(int dimensions)

Method Summary

double[]	center() lazily calculate and return expectation cluster
int	getAmountSamples()
int	getDimensions()
double[]	getMx()
double[]	getSum()
double	radius(List<double[]> samples) calculate maximal distance for group from center
void	setAmountSamples(int amountSamples)
void	setDimensions(int dimensions)
void	setMx(double[] mx)
void	setSum(double[] sum)
void	train(double[] features) training means coputing sum of values.

Methods inherited from class java.lang.Object

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

Methods inherited from interface net.sourceforge.javaocr.cluster.Metric

distance

Field Detail

sum

private double[] sum

mx

private double[] mx

amountSamples

private int amountSamples

dimensions

private int dimensions

Constructor Detail

AbstractBaseCluster

public AbstractBaseCluster()

AbstractBaseCluster

protected AbstractBaseCluster(int dimensions)

AbstractBaseCluster

protected AbstractBaseCluster(double[] mx)

Method Detail

center

public double[] center()

lazily calculate and return expectation cluster

Specified by:

center in interface Cluster

Returns:

expectation vector

getAmountSamples

public int getAmountSamples()

getMx

public double[] getMx()

setMx

public void setMx(double[] mx)

setDimensions

public void setDimensions(int dimensions)

getDimensions

public int getDimensions()

Specified by:

getDimensions in interface Metric

setAmountSamples

public void setAmountSamples(int amountSamples)

train

public void train(double[] features)

training means coputing sum of values. mx shall be reset and sampler countet up

Specified by:

train in interface Cluster

Parameters:

features -

getSum

public double[] getSum()

setSum

public void setSum(double[] sum)

radius

public double radius(List<double[]> samples)

calculate maximal distance for group from center

Specified by:

radius in interface Cluster

Parameters:

samples - sample group, sample size shall correspond to cluster dimensions

Returns: