gov.sandia.cognition.statistics

Class KullbackLeiblerDivergence<DomainType>

java.lang.Object

gov.sandia.cognition.statistics.KullbackLeiblerDivergence<DomainType>

Type Parameters:

DomainType - The type for the domain of the two distributions which this class is comparing.

@PublicationReference(author="Wikipedia",
                      title="Kullback--Leibler Divergence",
                      type=WebPage,
                      year=2014,
                      url="http://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence")
public class KullbackLeiblerDivergence<DomainType>
extends java.lang.Object

A class used for measuring how similar two distributions are using Kullback--Leibler Divergence.

Since:

3.4.2

Author:

trbroun

Constructor Summary

Constructors

Constructor and Description
KullbackLeiblerDivergence(DiscreteDistribution<DomainType> firstDistribution, DiscreteDistribution<DomainType> secondDistribution) Basic constructor to find the Kullback--Leibler Divergence between the two supplied distributions.

Method Summary

Modifier and Type	Method and Description
double	compute() Computes the Kullback--Leibler Divergence.
java.util.Set<? extends DomainType>	getDomain() Gets the domain of the distributions.

Methods inherited from class java.lang.Object

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

Constructor Detail

KullbackLeiblerDivergence

public KullbackLeiblerDivergence(DiscreteDistribution<DomainType> firstDistribution,
                                 DiscreteDistribution<DomainType> secondDistribution)

Basic constructor to find the Kullback--Leibler Divergence between the two supplied distributions. The two distributions must be over the same domain.

Parameters:

firstDistribution -

secondDistribution -

Method Detail

getDomain

public java.util.Set<? extends DomainType> getDomain()

Gets the domain of the distributions.

Returns:

The domain of the distributions.

compute

public double compute()

Computes the Kullback--Leibler Divergence.

Returns:

The divergence value.