Jet.HMM
Class HMMemitter

java.lang.Object
  Jet.HMM.HMMemitter

Direct Known Subclasses:

BasicHMMemitter, BigramHMMemitter, SuffixHMMemitter, WordFeatureHMMemitter

public abstract class HMMemitter

extends java.lang.Object

this abstract class specifies the methods required for any class which computes the emission probabilities for an HMM.

Field Summary

HMM	hmm
java.lang.String	stateName

Constructor Summary

HMMemitter()

Method Summary

abstract void	computeProbabilities() computate probabilities of emission from counts.
abstract double	getProbability(java.lang.String token, java.lang.String priorToken, FeatureSet fs) returns the probability that the current HMM state will emit token token with FeatureSet fs.
abstract void	print() print the information about emission from this state to System.out.
abstract void	resetForTraining() initialize the emitter prior to training.
void	setCacheCount(java.lang.String type, int n)
abstract void	store(java.io.PrintWriter stream) write the information about emission from this state in a form which can be read by HMM.load(java.io.Reader).
abstract void	trainOnInstances(java.lang.String token, java.lang.String priorToken, int n) update emission counts to indicate that String token appeared n times as an output of the current state.

Methods inherited from class java.lang.Object

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

Field Detail

hmm

public HMM hmm

stateName

public java.lang.String stateName

Constructor Detail

HMMemitter

public HMMemitter()

Method Detail

resetForTraining

public abstract void resetForTraining()

initialize the emitter prior to training. This method should be called before any calls on trainOnInstances(java.lang.String, java.lang.String, int).

trainOnInstances

public abstract void trainOnInstances(java.lang.String token,
                                      java.lang.String priorToken,
                                      int n)

update emission counts to indicate that String token appeared n times as an output of the current state.

computeProbabilities

public abstract void computeProbabilities()

computate probabilities of emission from counts. This method will be called after all calls on trainOnInstances(java.lang.String, java.lang.String, int) and before calls on getProbability(java.lang.String, java.lang.String, Jet.Lisp.FeatureSet).

getProbability

public abstract double getProbability(java.lang.String token,
                                      java.lang.String priorToken,
                                      FeatureSet fs)

returns the probability that the current HMM state will emit token token with FeatureSet fs.

setCacheCount

public void setCacheCount(java.lang.String type,
                          int n)

print

public abstract void print()

print the information about emission from this state to System.out.

store

public abstract void store(java.io.PrintWriter stream)

write the information about emission from this state in a form which can be read by HMM.load(java.io.Reader).