Package de.jstacs.sequenceScores.statisticalModels.trainable.hmm.states

The package provides all interfaces and classes for states used in hidden Markov models.

See:
Description

Interface Summary

DifferentiableState	This interface declares a method that allows to evaluate the gradient which is essential for numerical optimization.
SamplingState
State	This interface declares the methods of any state used in a hidden Markov model.
TrainableState	This class implements method that allows to fill a statistic, which is used to estimate the parameters of a state during, for instance, the Baum-Welch training.

Class Summary

SimpleDifferentiableState	This class implements a State based on Emission that allows to reuse Emissions for different States.
SimpleSamplingState	This class implements a state that can be used for a HMM that obtains its parameters from sampling.
SimpleState	This class implements a State based on Emission that allows to reuse Emissions for different States.

Package de.jstacs.sequenceScores.statisticalModels.trainable.hmm.states Description

The package provides all interfaces and classes for states used in hidden Markov models. Since states can share the same emission, the emissions build a subpackage.

Since:

Jstacs 1.5

See Also:

State, de.jstacs.sequenceScores.statisticalModels.trainable.hmm.states.emissions