Package org.tribuo.classification.ensemble

Class ClassificationEnsembleOptions

java.lang.Object

org.tribuo.classification.ensemble.ClassificationEnsembleOptions

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Options

public class ClassificationEnsembleOptions
extends Object
implements com.oracle.labs.mlrg.olcut.config.Options

Options for building a classification ensemble.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static enum	ClassificationEnsembleOptions.EnsembleType	The type of ensemble.

Field Summary

Fields

Modifier and Type	Field	Description
int	ensembleSize	Number of base learners in the ensemble.
long	seed	RNG seed.
ClassificationEnsembleOptions.EnsembleType	type	Ensemble method, options are {ADABOOST, BAGGING, RF}.

Fields inherited from interface com.oracle.labs.mlrg.olcut.config.Options

header

Constructor Summary

Constructors

Constructor	Description
ClassificationEnsembleOptions()

Method Summary

Modifier and Type	Method	Description
Trainer<Label>	wrapTrainer(Trainer<Label> trainer)	Wraps the supplied trainer using the ensemble trainer described by these options.

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.oracle.labs.mlrg.olcut.config.Options

getOptionsDescription

Field Details

type

@Option(longName="ensemble-type",
        usage="Ensemble method, options are {ADABOOST, BAGGING, RF}.")
public ClassificationEnsembleOptions.EnsembleType type

Ensemble method, options are {ADABOOST, BAGGING, RF}.

ensembleSize

@Option(longName="ensemble-size",
        usage="Number of base learners in the ensemble.")
public int ensembleSize

Number of base learners in the ensemble.

seed

@Option(longName="ensemble-seed",
        usage="RNG seed.")
public long seed

RNG seed.

Constructor Details

ClassificationEnsembleOptions

public ClassificationEnsembleOptions()

Method Details

wrapTrainer

public Trainer<Label> wrapTrainer(Trainer<Label> trainer)

Wraps the supplied trainer using the ensemble trainer described by these options.

Parameters:

trainer - The trainer to wrap.

Returns:

An ensemble trainer.