org.encog.ml.hmm.alog

Class ForwardBackwardScaledCalculator

java.lang.Object

org.encog.ml.hmm.alog.ForwardBackwardCalculator

org.encog.ml.hmm.alog.ForwardBackwardScaledCalculator

public class ForwardBackwardScaledCalculator
extends ForwardBackwardCalculator

The forward-backward algorithm is an inference algorithm for hidden Markov models which computes the posterior marginals of all hidden state variables given a sequence of observations. This version makes use of scaling, and will not generate underflows with long sequences.

Nested Class Summary

Nested classes/interfaces inherited from class org.encog.ml.hmm.alog.ForwardBackwardCalculator

ForwardBackwardCalculator.Computation

Constructor Summary

Constructors

Constructor and Description
ForwardBackwardScaledCalculator(MLDataSet oseq, HiddenMarkovModel hmm)
ForwardBackwardScaledCalculator(MLDataSet oseq, HiddenMarkovModel hmm, java.util.EnumSet<ForwardBackwardCalculator.Computation> flags)

Method Summary

Modifier and Type	Method and Description
double	lnProbability()

Methods inherited from class org.encog.ml.hmm.alog.ForwardBackwardCalculator

alphaElement, betaElement, probability

Methods inherited from class java.lang.Object

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

Constructor Detail

ForwardBackwardScaledCalculator

public ForwardBackwardScaledCalculator(MLDataSet oseq,
                                       HiddenMarkovModel hmm)

ForwardBackwardScaledCalculator

public ForwardBackwardScaledCalculator(MLDataSet oseq,
                                       HiddenMarkovModel hmm,
                                       java.util.EnumSet<ForwardBackwardCalculator.Computation> flags)

Method Detail

lnProbability

public double lnProbability()