Package cc.mallet.classify

Class MaxEntGETrainer

java.lang.Object

cc.mallet.classify.ClassifierTrainer<MaxEnt>

cc.mallet.classify.MaxEntGETrainer

All Implemented Interfaces:

Boostable, ClassifierTrainer.ByOptimization<MaxEnt>, java.io.Serializable

public class MaxEntGETrainer
extends ClassifierTrainer<MaxEnt>
implements ClassifierTrainer.ByOptimization<MaxEnt>, Boostable, java.io.Serializable

Training of MaxEnt models with labeled features using Generalized Expectation Criteria. Based on: "Learning from Labeled Features using Generalized Expectation Criteria" Gregory Druck, Gideon Mann, Andrew McCallum SIGIR 2008

Author:

Gregory Druck gdruck@cs.umass.edu Better explanations of parameters is given in MaxEntOptimizableByGE

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class cc.mallet.classify.ClassifierTrainer

ClassifierTrainer.ByActiveLearning<C extends Classifier>, ClassifierTrainer.ByIncrements<C extends Classifier>, ClassifierTrainer.ByInstanceIncrements<C extends Classifier>, ClassifierTrainer.ByOptimization<C extends Classifier>, ClassifierTrainer.Factory<CT extends ClassifierTrainer<? extends Classifier>>

Field Summary

Fields

Modifier and Type	Field	Description
protected java.util.ArrayList<MaxEntGEConstraint>	constraints

Fields inherited from class cc.mallet.classify.ClassifierTrainer

finishedTraining, validationSet

Constructor Summary

Constructors

Constructor	Description
MaxEntGETrainer()
MaxEntGETrainer(java.util.ArrayList<MaxEntGEConstraint> constraints)
MaxEntGETrainer(java.util.ArrayList<MaxEntGEConstraint> constraints, MaxEnt classifier)

Method Summary

Modifier and Type	Method	Description
MaxEnt	getClassifier()
int	getIteration()
Optimizable.ByGradientValue	getOptimizable(InstanceList trainingList)
Optimizer	getOptimizer()
void	setConstraintsFile(java.lang.String filename)
void	setGaussianPriorVariance(double variance)
void	setL2(boolean flag)
void	setMaxIterations(int iter)	Specifies the maximum number of iterations to run during a single call to train or trainWithFeatureInduction.
void	setNormalize(boolean normalize)
void	setOptimizer(Optimizer opt)
void	setTemperature(double temp)
void	setUseValues(boolean flag)
MaxEnt	train(InstanceList trainingList)
MaxEnt	train(InstanceList train, int maxIterations)

Methods inherited from class cc.mallet.classify.ClassifierTrainer

getValidationInstances, isFinishedTraining, setValidationInstances

Methods inherited from class java.lang.Object

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

Field Detail

constraints

protected java.util.ArrayList<MaxEntGEConstraint> constraints

Constructor Detail

MaxEntGETrainer

public MaxEntGETrainer()

MaxEntGETrainer

public MaxEntGETrainer(java.util.ArrayList<MaxEntGEConstraint> constraints)

MaxEntGETrainer

public MaxEntGETrainer(java.util.ArrayList<MaxEntGEConstraint> constraints,
                       MaxEnt classifier)

Method Detail

setConstraintsFile

public void setConstraintsFile(java.lang.String filename)

setTemperature

public void setTemperature(double temp)

setGaussianPriorVariance

public void setGaussianPriorVariance(double variance)

getClassifier

public MaxEnt getClassifier()

Specified by:

getClassifier in class ClassifierTrainer<MaxEnt>

setUseValues

public void setUseValues(boolean flag)

setL2

public void setL2(boolean flag)

setNormalize

public void setNormalize(boolean normalize)

getOptimizable

public Optimizable.ByGradientValue getOptimizable(InstanceList trainingList)

getOptimizer

public Optimizer getOptimizer()

Specified by:

getOptimizer in interface ClassifierTrainer.ByOptimization<MaxEnt>

setOptimizer

public void setOptimizer(Optimizer opt)

setMaxIterations

public void setMaxIterations(int iter)

Specifies the maximum number of iterations to run during a single call to train or trainWithFeatureInduction.

getIteration

public int getIteration()

Specified by:

getIteration in interface ClassifierTrainer.ByOptimization<MaxEnt>

train

public MaxEnt train(InstanceList trainingList)

Specified by:

train in class ClassifierTrainer<MaxEnt>

train

public MaxEnt train(InstanceList train,
                    int maxIterations)

Specified by:

train in interface ClassifierTrainer.ByOptimization<MaxEnt>