broad.core.hmm.MarkovModel< T > Class Reference

Classes
class	BackwardResult< T >
class	ForwardResult< T >

Public Member Functions
	MarkovModel (int numberOfStates)
void	addState (MarkovState< T > state)
void	addState (MarkovState< T > state, int stateIdx)
void	setInitialStateTransitionProbability (int stateIdx, double probability)
void	setEndStateTransitionProbability (int stateIdx, double probability)
MarkovState< T >	getState (String stateName)
List< Integer >	emitPath (int numOfSteps)
void	setStateTransitionProbability (int stateIdx1, int stateIdx2, double probability)
double	computePathLogLikelihood (List< T > observationSequence, short[] path, int initialStateIdx, int endStateIdx)
short[]	viterbiMostLikelyEstimation (List< T > observationSequence) throws BadModelException
ForwardResult< T >	runForwardAlgorithm (List< T > observationSequence) throws BadModelException
BackwardResult< T >	runBackwardAlgorithm (List< T > observationSequence, ForwardResult< T > forwardData) throws BadModelException
ForwardResult< T >	runForwardAlgorithmNoRescaling (List< T > observationSequence)

Protected Member Functions
Matrix	getStateTransitionMatrix ()
List< MarkovState< T > >	getStates ()
double[]	getInitialStateTransitionProbabilities ()
List< String >	getConsistencyProblems ()
int	drawInitialState ()
int	drawNextState (int prior)

Constructor & Destructor Documentation

broad.core.hmm.MarkovModel< T >.MarkovModel

(

int

numberOfStates

)

Initialize the Model by specifying the number of states. The probabilities of starting with a given state will be initialized as uniform: All states are equally likely at start

Parameters

numberOfStates

Member Function Documentation

void broad.core.hmm.MarkovModel< T >.addState

(

MarkovState< T >

state

)

void broad.core.hmm.MarkovModel< T >.addState	(	MarkovState< T >	state,
		int	stateIdx
	)

double broad.core.hmm.MarkovModel< T >.computePathLogLikelihood	(	List< T >	observationSequence,
		short[]	path,
		int	initialStateIdx,
		int	endStateIdx
	)

Here is the call graph for this function:

int broad.core.hmm.MarkovModel< T >.drawInitialState ( )

protected

int broad.core.hmm.MarkovModel< T >.drawNextState ( int  prior )

protected

List<Integer> broad.core.hmm.MarkovModel< T >.emitPath

(

int

numOfSteps

)

List<String> broad.core.hmm.MarkovModel< T >.getConsistencyProblems ( )

protected

double [] broad.core.hmm.MarkovModel< T >.getInitialStateTransitionProbabilities ( )

protected

MarkovState<T> broad.core.hmm.MarkovModel< T >.getState

(

String

stateName

)

Here is the call graph for this function:

List<MarkovState<T> > broad.core.hmm.MarkovModel< T >.getStates ( )

protected

Matrix broad.core.hmm.MarkovModel< T >.getStateTransitionMatrix ( )

protected

BackwardResult<T> broad.core.hmm.MarkovModel< T >.runBackwardAlgorithm	(	List< T >	observationSequence,
		ForwardResult< T >	forwardData
	)		throws BadModelException

ForwardResult<T> broad.core.hmm.MarkovModel< T >.runForwardAlgorithm

(

List< T >

observationSequence

)

throws BadModelException

ForwardResult<T> broad.core.hmm.MarkovModel< T >.runForwardAlgorithmNoRescaling

(

List< T >

observationSequence

)

void broad.core.hmm.MarkovModel< T >.setEndStateTransitionProbability	(	int	stateIdx,
		double	probability
	)

void broad.core.hmm.MarkovModel< T >.setInitialStateTransitionProbability	(	int	stateIdx,
		double	probability
	)

void broad.core.hmm.MarkovModel< T >.setStateTransitionProbability	(	int	stateIdx1,
		int	stateIdx2,
		double	probability
	)

short [] broad.core.hmm.MarkovModel< T >.viterbiMostLikelyEstimation

(

List< T >

observationSequence

)

throws BadModelException

---

The documentation for this class was generated from the following file:

• java/broad/core/hmm/MarkovModel.java