Package se.hb.jcp.cp.measures
Class ObservedProbabilityLogLoss
- java.lang.Object
-
- se.hb.jcp.cp.measures.ObservedProbabilityLogLoss
-
- All Implemented Interfaces:
IMeasure,IObservedMeasure,IObservedProbabilisticMeasure
public class ObservedProbabilityLogLoss extends java.lang.Object implements IObservedProbabilisticMeasure
The observed log loss measure depends on the prediction made and the true label. See [V. Vovk, I. Petej, "Venn-Abers Predictors", On-line Compression Modelling Project (New Series) Working Paper #7, 2012] for the definitions used here.- Author:
- anders.gidenstam(at)hb.se
-
-
Constructor Summary
Constructors Constructor Description ObservedProbabilityLogLoss()
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description doublecompute(ConformalClassification prediction, double trueLabel)Compute the measure for the supplied conformal prediction and true label.doublecompute(ConformalMultiProbabilisticClassification prediction, double trueLabel)Compute the measure for the supplied conformal prediction and true label.java.lang.StringgetName()Get the name of this measure.
-
-
-
Method Detail
-
compute
public double compute(ConformalMultiProbabilisticClassification prediction, double trueLabel)
Description copied from interface:IObservedProbabilisticMeasureCompute the measure for the supplied conformal prediction and true label.- Specified by:
computein interfaceIObservedProbabilisticMeasure- Parameters:
prediction- a ConformalMultiProbabilisticClassification.trueLabel- the true label of the instance.- Returns:
- the measure for the supplied prediction.
-
compute
public double compute(ConformalClassification prediction, double trueLabel)
Description copied from interface:IObservedMeasureCompute the measure for the supplied conformal prediction and true label.- Specified by:
computein interfaceIObservedMeasure- Parameters:
prediction- a ConformalClassification.trueLabel- the true label of the instance.- Returns:
- the measure for the supplied prediction.
-
-