edu.umass.cs.mallet.base.classify

Class AdaBoostTrainer

java.lang.Object

edu.umass.cs.mallet.base.classify.ClassifierTrainer

edu.umass.cs.mallet.base.classify.AdaBoostTrainer

public class AdaBoostTrainer
extends ClassifierTrainer

This version of AdaBoost should be used only for binary classification. Use AdaBoost.M2 for multi-class problems.

Robert E. Schapire. "A decision-theoretic generalization of on-line learning and an application to boosting" In Journal of Computer and System Sciences http://www.cs.princeton.edu/~schapire/uncompress-papers.cgi/FreundSc95.ps

Author:

Andrew McCallum mccallum@cs.umass.edu

Constructor Summary

Constructors

Constructor and Description
AdaBoostTrainer(ClassifierTrainer weakLearner)
AdaBoostTrainer(ClassifierTrainer weakLearner, int numRounds)

Method Summary

Modifier and Type	Method and Description
Classifier	train(InstanceList trainingList, InstanceList validationList, InstanceList testSet, ClassifierEvaluating evaluator, Classifier initialClassifier) Boosting method that resamples instances using their weights

Methods inherited from class edu.umass.cs.mallet.base.classify.ClassifierTrainer

main, toString, train, train, train, train

Methods inherited from class java.lang.Object

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

Constructor Detail

AdaBoostTrainer

public AdaBoostTrainer(ClassifierTrainer weakLearner,
                       int numRounds)

AdaBoostTrainer

public AdaBoostTrainer(ClassifierTrainer weakLearner)

Method Detail

train

public Classifier train(InstanceList trainingList,
                        InstanceList validationList,
                        InstanceList testSet,
                        ClassifierEvaluating evaluator,
                        Classifier initialClassifier)

Boosting method that resamples instances using their weights

Specified by:

train in class ClassifierTrainer

Parameters:

trainingList - examples used to set parameters.

validationList - examples used to tune meta-parameters. May be null.

testSet - examples not examined at all for training, but passed on to diagnostic routines. May be null.

initialClassifier - training process may start from here. The parameters of the initialClassifier are not modified. May be null.