jsat.classifiers.boosting

Class LogitBoost

java.lang.Object

jsat.classifiers.boosting.LogitBoost

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Classifier, Parameterized

Direct Known Subclasses:

LogitBoostPL

public class LogitBoost
extends java.lang.Object
implements Classifier, Parameterized

An implementation of the original 2 class LogitBoost algorithm. While there is a multi-class description in the original paper, its implementation is congruent with the result of using LogitBoost with OneVSAll classifier.
LogitBoost differs from its predecessors in that it boosts regression models to create a powerful classifier.

Paper: Special Invited Paper Additive Logistic Regression: A Statistical View of Boosting, By Jerome Friedman, Trevor Hastie and Robert Tibshirani. The Annals of Statistics 2000, Vol. 28, No. 2, 337–407

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
LogitBoost(int M) Creates a new LogitBoost using the standard MultipleLinearRegression .
LogitBoost(Regressor baseLearner, int M) Creates a new LogitBoost using the given base learner.

Method Summary

Modifier and Type	Method and Description
CategoricalResults	classify(DataPoint data) Performs classification on the given data point.
LogitBoost	clone()
int	getMaxIterations() The maximum number of iterations of boosting that may occur.
java.util.List<Regressor>	getModels()
double	getzMax() Returns the maximum miss-classification penalty used by the algorithm.
void	setMaxIterations(int maxIterations) Sets the maximum number of iterations of boosting that can occur, giving the maximum number of base learners that may be trained
void	setzMax(double zMax) Sets the penalty bound for miss-classification of results.
boolean	supportsWeightedData() Indicates whether the model knows how to train using weighted data points.
void	train(ClassificationDataSet dataSet, boolean parallel) Trains the classifier and constructs a model for classification using the given data set.

Methods inherited from class java.lang.Object

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

Methods inherited from interface jsat.classifiers.Classifier

train

Methods inherited from interface jsat.parameters.Parameterized

getParameter, getParameters

Constructor Detail

LogitBoost

public LogitBoost(int M)

Creates a new LogitBoost using the standard MultipleLinearRegression .

Parameters:

M - the maximum number of iterations.

LogitBoost

public LogitBoost(Regressor baseLearner,
                  int M)

Creates a new LogitBoost using the given base learner.

Parameters:

baseLearner - the weak learner to build an ensemble out of.

M - the maximum number of iterations.

Method Detail

getModels

public java.util.List<Regressor> getModels()

Returns:

a list of the models that are in this ensemble.

setMaxIterations

public void setMaxIterations(int maxIterations)

Sets the maximum number of iterations of boosting that can occur, giving the maximum number of base learners that may be trained

Parameters:

maxIterations - the maximum number of iterations

getMaxIterations

public int getMaxIterations()

The maximum number of iterations of boosting that may occur.

Returns:

maximum number of iterations of boosting that may occur.

setzMax

public void setzMax(double zMax)

Sets the penalty bound for miss-classification of results. This also provides numerical stability to the algorithm. The results are not sensitive to this value. The recommended value range is in [2, 4]

Parameters:

zMax - the penalty bound

Throws:

java.lang.ArithmeticException - if the value is not in (0, Double.MAX_VALUE]

getzMax

public double getzMax()

Returns the maximum miss-classification penalty used by the algorithm.

Returns:

the maximum miss-classification

classify

public CategoricalResults classify(DataPoint data)

Description copied from interface: Classifier

Performs classification on the given data point.

Specified by:

classify in interface Classifier

Parameters:

data - the data point to classify

Returns:

the results of the classification.

train

public void train(ClassificationDataSet dataSet,
                  boolean parallel)

Description copied from interface: Classifier

Trains the classifier and constructs a model for classification using the given data set. If the training method knows how, it will used the threadPool to conduct training in parallel. This method will block until the training has completed.

Specified by:

train in interface Classifier

Parameters:

dataSet - the data set to train on

parallel - true if multiple threads should be used to train the model. false if it should be done in a single threaded manner.

supportsWeightedData

public boolean supportsWeightedData()

Description copied from interface: Classifier

Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.

Specified by:

supportsWeightedData in interface Classifier

Returns:

true if the model supports weighted data, false otherwise

clone

public LogitBoost clone()

Specified by:

clone in interface Classifier

Overrides:

clone in class java.lang.Object