Package opennlp.tools.ml

Class BeamSearch<T>

java.lang.Object

opennlp.tools.ml.BeamSearch<T>

All Implemented Interfaces:

SequenceClassificationModel<T>

public class BeamSearch<T>
extends Object
implements SequenceClassificationModel<T>

Performs k-best search over sequence. This is based on the description in Ratnaparkhi (1998), PhD diss, Univ. of Pennsylvania.

See Also:

Sequence, SequenceValidator, BeamSearchContextGenerator

Field Summary

Fields

Modifier and Type	Field	Description
static String	BEAM_SIZE_PARAMETER

Constructor Summary

Constructors

Constructor	Description
BeamSearch(int size, MaxentModel model)	Creates new search object.
BeamSearch(int size, MaxentModel model, int cacheSize)

Method Summary

Modifier and Type	Method	Description
Sequence	bestSequence(T[] sequence, Object[] additionalContext, BeamSearchContextGenerator<T> cg, SequenceValidator<T> validator)	Finds the sequence with the highest probability.
Sequence[]	bestSequences(int numSequences, T[] sequence, Object[] additionalContext, double minSequenceScore, BeamSearchContextGenerator<T> cg, SequenceValidator<T> validator)	Returns the best sequence of outcomes based on model for this object.
Sequence[]	bestSequences(int numSequences, T[] sequence, Object[] additionalContext, BeamSearchContextGenerator<T> cg, SequenceValidator<T> validator)	Finds the n most probable sequences.
String[]	getOutcomes()	Returns all possible outcomes.

Methods inherited from class java.lang.Object

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

Field Detail

BEAM_SIZE_PARAMETER

public static final String BEAM_SIZE_PARAMETER

See Also:

Constant Field Values

Constructor Detail

BeamSearch

public BeamSearch(int size,
                  MaxentModel model)

Creates new search object.

Parameters:

size - The size of the beam (k).

model - the model for assigning probabilities to the sequence outcomes.

BeamSearch

public BeamSearch(int size,
                  MaxentModel model,
                  int cacheSize)

Method Detail

bestSequences

public Sequence[] bestSequences(int numSequences,
                                T[] sequence,
                                Object[] additionalContext,
                                double minSequenceScore,
                                BeamSearchContextGenerator<T> cg,
                                SequenceValidator<T> validator)

Returns the best sequence of outcomes based on model for this object.

Specified by:

bestSequences in interface SequenceClassificationModel<T>

Parameters:

sequence - The input sequence.

additionalContext - An Object[] of additional context. This is passed to the context generator blindly with the assumption that the context are appropiate.

Returns:

The top ranked sequence of outcomes or null if no sequence could be found

bestSequences

public Sequence[] bestSequences(int numSequences,
                                T[] sequence,
                                Object[] additionalContext,
                                BeamSearchContextGenerator<T> cg,
                                SequenceValidator<T> validator)

Description copied from interface: SequenceClassificationModel

Finds the n most probable sequences.

Specified by:

bestSequences in interface SequenceClassificationModel<T>

Returns:

bestSequence

public Sequence bestSequence(T[] sequence,
                             Object[] additionalContext,
                             BeamSearchContextGenerator<T> cg,
                             SequenceValidator<T> validator)

Description copied from interface: SequenceClassificationModel

Finds the sequence with the highest probability.

Specified by:

bestSequence in interface SequenceClassificationModel<T>

Returns:

getOutcomes

public String[] getOutcomes()

Description copied from interface: SequenceClassificationModel

Returns all possible outcomes.

Specified by:

getOutcomes in interface SequenceClassificationModel<T>

Returns: