cc.mallet.classify

Class RankMaxEntTrainer

java.lang.Object

cc.mallet.classify.ClassifierTrainer<MaxEnt>

cc.mallet.classify.MaxEntTrainer

cc.mallet.classify.RankMaxEntTrainer

All Implemented Interfaces:

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

public class RankMaxEntTrainer
extends MaxEntTrainer

The trainer for a RankMaxEnt classifier. Expects Instance data to be a FeatureVectorSequence, and the target to be a String representation of the index of the true best FeatureVectorSequence. Note that the Instance target may be a Labels to indicate a tie for the best Instance.

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>>

Constructor Summary

Constructors

Constructor and Description
RankMaxEntTrainer()
RankMaxEntTrainer(double gaussianPriorVariance) Constructs a trainer with a parameter to avoid overtraining.

Method Summary

Modifier and Type	Method and Description
Optimizable.ByGradientValue	getMaximizableTrainer(InstanceList ilist)
java.lang.String	toString() Like the other version of trainWithFeatureInduction, but allows some default options to be changed.
MaxEnt	train(InstanceList trainingSet)

Methods inherited from class cc.mallet.classify.MaxEntTrainer

getClassifier, getIteration, getOptimizable, getOptimizable, getOptimizable, getOptimizer, getOptimizer, setClassifier, setGaussianPriorVariance, setL1Weight, setNumIterations, train

Methods inherited from class cc.mallet.classify.ClassifierTrainer

getValidationInstances, isFinishedTraining, setValidationInstances

Methods inherited from class java.lang.Object

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

Constructor Detail

RankMaxEntTrainer

public RankMaxEntTrainer()

RankMaxEntTrainer

public RankMaxEntTrainer(double gaussianPriorVariance)

Constructs a trainer with a parameter to avoid overtraining. 1.0 is usually a reasonable default value.

Method Detail

getMaximizableTrainer

public Optimizable.ByGradientValue getMaximizableTrainer(InstanceList ilist)

train

public MaxEnt train(InstanceList trainingSet)

Overrides:

train in class MaxEntTrainer

toString

public java.lang.String toString()

Like the other version of trainWithFeatureInduction, but allows some default options to be changed.

Overrides:

toString in class MaxEntTrainer

Parameters:

maxent - An initial partially-trained classifier (default null). This classifier may be modified during training.

gainName - The estimate of gain (log-likelihood increase) we want our chosen features to maximize. Should be one of MaxEntTrainer.EXP_GAIN, MaxEntTrainer.GRADIENT_GAIN, or MaxEntTrainer.INFORMATION_GAIN (default EXP_GAIN).

Returns:

The trained MaxEnt classifier