rdr.mulan
Class MultiLabelClassifierBase

java.lang.Object
  rdr.mulan.MultiLabelClassifierBase

All Implemented Interfaces:

java.io.Serializable, MultiLabelClassifier, weka.core.TechnicalInformationHandler

Direct Known Subclasses:

TransformationBasedMultiLabelClassifier

public abstract class MultiLabelClassifierBase
extends java.lang.Object
implements weka.core.TechnicalInformationHandler, MultiLabelClassifier, java.io.Serializable

Common base class for all multi-label classifiers.

Version:

$Revision: 0.3 $

Author:

Robert Friberg, Jozef Vilcek, Grigorios Tsoumakas

See Also:

Serialized Form

Constructor Summary

MultiLabelClassifierBase(int numLabels)
Creates a MultiLabelClassifierBase instance.

Method Summary

abstract void	buildClassifier(weka.core.Instances instances) Builds the classifier.
boolean	getDebug() Get whether debugging is turned on.
int	getNumLabels() Returns a number of labels the classifier is configured for.
abstract weka.core.TechnicalInformation	getTechnicalInformation()
static MultiLabelClassifier	makeCopy(MultiLabelClassifier model) Creates a deep copy of the given classifier using serialization.
Prediction	predict(weka.core.Instance instance) Computes the prediction of labels for a specified input Instance.
void	setDebug(boolean debug) Set debugging mode.

Methods inherited from class java.lang.Object

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

Constructor Detail

MultiLabelClassifierBase

public MultiLabelClassifierBase(int numLabels)

Creates a MultiLabelClassifierBase instance.

Parameters:

numLabels - the number of labels the classifier should use

Method Detail

getNumLabels

public int getNumLabels()

Returns a number of labels the classifier is configured for. The label attributes are assumed to be the last ones in Instances training data.

Specified by:

getNumLabels in interface MultiLabelClassifier

Returns:

the number of labels

getTechnicalInformation

public abstract weka.core.TechnicalInformation getTechnicalInformation()

Specified by:

getTechnicalInformation in interface weka.core.TechnicalInformationHandler

buildClassifier

public abstract void buildClassifier(weka.core.Instances instances)
                              throws java.lang.Exception

Builds the classifier.

Specified by:

buildClassifier in interface MultiLabelClassifier

Parameters:

instances - set of training data, upon which the classifier should be build

Throws:

java.lang.Exception - if classifier was not created successfully

predict

public final Prediction predict(weka.core.Instance instance)
                         throws java.lang.Exception

Description copied from interface: MultiLabelClassifier

Computes the prediction of labels for a specified input Instance.

Specified by:

predict in interface MultiLabelClassifier

Parameters:

instance - the input instance for which the prediction is made

Returns:

the prediction

Throws:

java.lang.Exception - if prediction was not successful

setDebug

public void setDebug(boolean debug)

Set debugging mode.

Parameters:

debug - true if debug output should be printed

getDebug

public boolean getDebug()

Get whether debugging is turned on.

Returns:

true if debugging output is on

makeCopy

public static MultiLabelClassifier makeCopy(MultiLabelClassifier model)
                                     throws java.lang.Exception

Creates a deep copy of the given classifier using serialization.

Parameters:

model - the classifier to copy

Returns:

a deep copy of the classifier

Throws:

java.lang.Exception - if an error occurs