Package gov.sandia.cognition.learning.algorithm.hmm

Provides hidden Markov model (HMM) algorithms.

See:
          Description

Class Summary
AbstractBaumWelchAlgorithm<ObservationType,DataType> Partial implementation of the Baum-Welch algorithm.
BaumWelchAlgorithm<ObservationType> Implements the Baum-Welch algorithm, also known as the "forward-backward algorithm", the expectation-maximization algorithm, etc for Hidden Markov Models (HMMs).
HiddenMarkovModel<ObservationType> A discrete-state Hidden Markov Model (HMM) with either continuous or discrete observations.
MarkovChain A Markov chain is a random process that has a finite number of states with random transition probabilities between states at discrete time steps.
ParallelBaumWelchAlgorithm<ObservationType> A Parallelized implementation of some of the methods of the Baum-Welch Algorithm.
ParallelBaumWelchAlgorithm.DistributionEstimatorTask<ObservationType> Re-estimates the PDF from the gammas.
ParallelHiddenMarkovModel<ObservationType> A Hidden Markov Model with parallelized processing.
ParallelHiddenMarkovModel.ComputeTransitionsTask Calls the computeTransitions method.
ParallelHiddenMarkovModel.NormalizeTransitionTask Calls the normalizeTransitionMatrix method.
ParallelHiddenMarkovModel.ObservationLikelihoodTask<ObservationType> Calls the computeObservationLikelihoods() method.
ParallelHiddenMarkovModel.StateObservationLikelihoodTask Calls the computeStateObservationLikelihood() method.
 

Package gov.sandia.cognition.learning.algorithm.hmm Description

Provides hidden Markov model (HMM) algorithms.

Since:
3.0
Author:
Kevin R. Dixon