mulan.evaluation.loss
Class RankingLossFunctionBase

java.lang.Object
  mulan.evaluation.loss.RankingLossFunctionBase

All Implemented Interfaces:

Serializable, MultiLabelLossFunction, RankingLossFunction

Direct Known Subclasses:

ErrorSetSize, IsError, OneError, OneMinusAveragePrecision

public abstract class RankingLossFunctionBase
extends Object
implements RankingLossFunction, Serializable

Base class for ranking loss functions

Version:

2010.11.10

Author:

GrigoriosTsoumakas

See Also:

Serialized Form

Constructor Summary

RankingLossFunctionBase()

Method Summary

abstract double	computeLoss(int[] ranking, boolean[] groundTruth) Computes the ranking 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

RankingLossFunctionBase

public RankingLossFunctionBase()

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(int[] ranking,
                                   boolean[] groundTruth)

Description copied from interface: RankingLossFunction

Computes the ranking loss function

Specified by:

computeLoss in interface RankingLossFunction

Parameters:

ranking - the ranking of the learner for an example

groundTruth - the ground truth of the example

Returns:

the value of the loss function