ai.libs.jaicore.ml.core.evaluation.measure

Class LossScoreTransformer<I>

java.lang.Object

ai.libs.jaicore.ml.core.evaluation.measure.ADecomposableMeasure<I,java.lang.Double>

ai.libs.jaicore.ml.core.evaluation.measure.ADecomposableDoubleMeasure<I>

ai.libs.jaicore.ml.core.evaluation.measure.LossScoreTransformer<I>

Type Parameters:

I - The input domain of the two measures.

All Implemented Interfaces:

IMeasure<I,java.lang.Double>

Direct Known Subclasses:

AutoMEKAGGPFitnessMeasure, ExactMatchAccuracy, F1MacroAverageLLoss, HammingAccuracy, InstanceWiseF1AsLoss, JaccardLoss, RankScore

public class LossScoreTransformer<I>
extends ADecomposableDoubleMeasure<I>

This transformer transforms a decomposable double measure from a scoring function to a loss or vice versa. It is required that the measured values lie in the interval [0,1]. Another requirement is that the input domains of the two measures are the same.

Constructor Summary

Constructors

Constructor and Description
LossScoreTransformer(ADecomposableDoubleMeasure<I> measure) Constructor for setting the measure to be transformed from loss to score or vice versa.

Method Summary

Modifier and Type	Method and Description
java.lang.Double	calculateMeasure(I actual, I expected) Computes the measure for a measured input actual and the expected outcome expected.

Methods inherited from class ai.libs.jaicore.ml.core.evaluation.measure.ADecomposableDoubleMeasure

calculateAvgMeasure

Methods inherited from class ai.libs.jaicore.ml.core.evaluation.measure.ADecomposableMeasure

calculateMeasure, calculateMeasure

Methods inherited from class java.lang.Object

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

Constructor Detail

LossScoreTransformer

public LossScoreTransformer(ADecomposableDoubleMeasure<I> measure)

Constructor for setting the measure to be transformed from loss to score or vice versa.

Parameters:

measure - The measure to be transformed into the opposite.

Method Detail

calculateMeasure

public java.lang.Double calculateMeasure(I actual,
                                         I expected)

Description copied from interface: IMeasure

Computes the measure for a measured input actual and the expected outcome expected.

Parameters:

actual - The actually available values.

expected - The expected values to compare the actual values with.

Returns:

The return value of the measure.