mulan.evaluation.loss
Class BipartitionLossFunctionBase

java.lang.Object
  mulan.evaluation.loss.BipartitionLossFunctionBase

All Implemented Interfaces:

Serializable, BipartitionLossFunction, MultiLabelLossFunction

Direct Known Subclasses:

HammingLoss, HierarchicalLoss

public abstract class BipartitionLossFunctionBase
extends Object
implements BipartitionLossFunction, Serializable

Base class for bipartition loss functions

Version:

2010.11.10

Author:

GrigoriosTsoumakas

See Also:

Serialized Form

Constructor Summary

BipartitionLossFunctionBase()

Method Summary

abstract double	computeLoss(boolean[] bipartition, boolean[] groundTruth) Computes the bipartition loss function
double	computeLoss(MultiLabelOutput prediction, boolean[] groundTruth) Computes the loss function

Methods inherited from class java.lang.Object

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

Methods inherited from interface mulan.evaluation.loss.MultiLabelLossFunction

getName

Constructor Detail

BipartitionLossFunctionBase

public BipartitionLossFunctionBase()

Method Detail

computeLoss

public final double computeLoss(MultiLabelOutput prediction,
                                boolean[] groundTruth)

Description copied from interface: MultiLabelLossFunction

Computes the loss function

Specified by:

computeLoss in interface MultiLabelLossFunction

Parameters:

prediction - the prediction of the learner for an example

groundTruth - the ground truth of the example

Returns:

the value of the loss function

computeLoss

public abstract double computeLoss(boolean[] bipartition,
                                   boolean[] groundTruth)

Description copied from interface: BipartitionLossFunction

Computes the bipartition loss function

Specified by:

computeLoss in interface BipartitionLossFunction

Parameters:

bipartition - the biprtition of the learner for an example

groundTruth - the ground truth of the example

Returns:

the value of the loss function