org.encog.ml.hmm.distributions
Class ContinousDistribution

java.lang.Object
  org.encog.ml.hmm.distributions.ContinousDistribution

All Implemented Interfaces:

Serializable, Cloneable, StateDistribution

public class ContinousDistribution
extends Object
implements StateDistribution

A continuous distribution represents an infinite range of choices between two real numbers. A gaussian distribution is used to distribute the probability.

See Also:

Serialized Form

Constructor Summary

ContinousDistribution(double[] mean, double[][] covariance)
Construct a continuous distribution.

ContinousDistribution(int dimension)
Construct a continuous distribution with the specified number of dimensions.

Method Summary

ContinousDistribution	clone()
void	fit(MLDataSet co) Fit this distribution to the specified data set.
void	fit(MLDataSet co, double[] weights) Fit this distribution to the specified data set, given the specified weights, per element.
MLDataPair	generate() Generate a random data pair, based on the probabilities.
Matrix	getCovariance()
double[]	getMean()
double	probability(MLDataPair o) Determine the probability of the specified data pair.
void	update(double[][] covariance) Update the covariance.

Methods inherited from class java.lang.Object

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

Constructor Detail

ContinousDistribution

public ContinousDistribution(double[] mean,
                             double[][] covariance)

Construct a continuous distribution.

Parameters:

mean - The mean.

covariance - The covariance.

ContinousDistribution

public ContinousDistribution(int dimension)

Construct a continuous distribution with the specified number of dimensions.

Parameters:

dimension - The dimensions.

Method Detail

clone

public ContinousDistribution clone()

Specified by:

clone in interface StateDistribution

Overrides:

clone in class Object

Returns:

A clone of this distribution.

fit

public void fit(MLDataSet co)

Fit this distribution to the specified data set.

Specified by:

fit in interface StateDistribution

Parameters:

co - The data set to fit to.

fit

public void fit(MLDataSet co,
                double[] weights)

Fit this distribution to the specified data set, given the specified weights, per element.

Specified by:

fit in interface StateDistribution

Parameters:

co - The data set to fit to.

weights - The weights.

generate

public MLDataPair generate()

Generate a random data pair, based on the probabilities.

Specified by:

generate in interface StateDistribution

Returns:

A random data pair.

probability

public double probability(MLDataPair o)

Determine the probability of the specified data pair.

Specified by:

probability in interface StateDistribution

Parameters:

o - The pair to consider.

Returns:

The probability.

update

public void update(double[][] covariance)

Update the covariance.

Parameters:

covariance - The new covariance.

getMean

public double[] getMean()

Returns:

The mean for the dimensions of the gaussian curve.

getCovariance

public Matrix getCovariance()

Returns:

The covariance matrix.