IObservedMeasure (JCP API)

All Superinterfaces:

IMeasure

All Known Subinterfaces:

IObservedProbabilisticMeasure

All Known Implementing Classes:

ObservedAccuracy, ObservedExcessCriterion, ObservedFuzzinessCriterion, ObservedMultipleCriterion, ObservedOneCCriterion, ObservedProbabilityLogLoss, ObservedProbabilitySquareLoss, ObservedUnconfidenceCriterion
```
public interface IObservedMeasure
extends IMeasure
```
An observed measure depends on the prediction made and the true label. See [V. Vovk, V. Fedorova, I. Nouretdinov and A. Gammerman, "Criteria of Efficiency for Conformal Prediction", COPA 2016, LNAI 9653, pp. 23-39, 2016] for the definitions used here.

Author:

anders.gidenstam(at)hb.se

Method Summary

All Methods Instance Methods Abstract Methods
Modifier and Type	Method	Description
`double`	`compute(ConformalClassification prediction, double trueLabel)`	Compute the measure for the supplied conformal prediction and true label.

Methods inherited from interface se.hb.jcp.cp.measures.IMeasure
getName

- Method Detail
  - compute
```
double compute(ConformalClassification prediction,
               double trueLabel)
```
    Compute the measure for the supplied conformal prediction and true label.
    
    Parameters:
    
    prediction - a ConformalClassification.
    
    trueLabel - the true label of the instance.
    
    Returns:
    
    the measure for the supplied prediction.