edu.gtts.sautrela.wfsa.models
Class Gaussian

java.lang.Object
  edu.gtts.sautrela.wfsa.models.Gaussian

public class Gaussian
extends java.lang.Object

Gaussian Probability Density Function

Author:

mpenagar

Constructor Summary

Gaussian(java.lang.String name, double[] mean, double[] var, double[] varFloor, boolean trainMean, boolean trainVar)
Creates a new instance of Gaussian

Method Summary

void	addTrain(double[] x) Equivalent to addTrain(x,1.0)
void	addTrain(double[] x, double count)
Gaussian	clone() Returns a copy of this Gaussian.
void	dump()
void	dump(Gaussian[] clones)
boolean	equals(Gaussian other) Indicates whether some other Gaussian is "equal to" this one.
double	getLogNorma()
double[]	getMean()
java.lang.String	getName()
double	getProb(double[] x) Returns the log probability of an array.
double[]	getRandomData() Returns a random vector.
double[]	getVar()
double[]	getVarFloor()
void	initTrain()
void	initTrain(double initCount)
boolean	isTrainMean()
boolean	isTrainVar()
Gaussian	randomClone() Returns a new Gaussian with random mean, but same var, varFloor & trainFlags.
void	resize(int newdim)
void	setName(java.lang.String name)
void	setTrainMean(boolean flag)
void	setTrainVar(boolean flag)
Gaussian	split() Returns a new Gaussian with random mean, but same var, varFloor & trainFlags.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

Gaussian

public Gaussian(java.lang.String name,
                double[] mean,
                double[] var,
                double[] varFloor,
                boolean trainMean,
                boolean trainVar)

Creates a new instance of Gaussian

Method Detail

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

getName

public java.lang.String getName()

setName

public void setName(java.lang.String name)

getMean

public double[] getMean()

getVar

public double[] getVar()

getVarFloor

public double[] getVarFloor()

getLogNorma

public double getLogNorma()

isTrainMean

public boolean isTrainMean()

setTrainMean

public void setTrainMean(boolean flag)

isTrainVar

public boolean isTrainVar()

setTrainVar

public void setTrainVar(boolean flag)

resize

public void resize(int newdim)

getProb

public double getProb(double[] x)

Returns the log probability of an array.

Parameters:

x - the data

Returns:

the log probability

getRandomData

public double[] getRandomData()

Returns a random vector.

Returns:

the log probability

clone

public Gaussian clone()

Returns a copy of this Gaussian.

Overrides:

clone in class java.lang.Object

Returns:

a clone of this Gaussian

randomClone

public Gaussian randomClone()

Returns a new Gaussian with random mean, but same var, varFloor & trainFlags.

Returns:

a new Gaussian with random mean, but same var, varFloor & trainFlags

split

public Gaussian split()

Returns a new Gaussian with random mean, but same var, varFloor & trainFlags. Random means are obtained using the varFloor instead of the variance

Returns:

a new Gaussian with random mean, but same var, varFloor & trainFlags

initTrain

public void initTrain()

initTrain

public void initTrain(double initCount)

addTrain

public void addTrain(double[] x)

Equivalent to addTrain(x,1.0)

addTrain

public void addTrain(double[] x,
                     double count)

dump

public void dump()

dump

public void dump(Gaussian[] clones)

equals

public boolean equals(Gaussian other)

Indicates whether some other Gaussian is "equal to" this one. Compares the mean and variances of both Gaussian (weigths are not considered).

Parameters:

other - - the reference Gaussian with which to compare

Returns:

true if this object is the same as the other argument; false otherwise.