RankLoss (jaicore-ml 0.2.5 API)

java.lang.Object
- ai.libs.jaicore.ml.classification.loss.dataset.APredictionPerformanceMeasure<int[],org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification>
- - ai.libs.jaicore.ml.classification.multilabel.evaluation.loss.AMultiLabelClassificationMeasure
  - - ai.libs.jaicore.ml.classification.multilabel.evaluation.loss.RankLoss

All Implemented Interfaces:

org.api4.java.ai.ml.classification.multilabel.evaluation.loss.IMultiLabelClassificationPredictionPerformanceMeasure, org.api4.java.ai.ml.core.evaluation.supervised.loss.IDeterministicPredictionPerformanceMeasure<int[],org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification>
```
public class RankLoss
extends AMultiLabelClassificationMeasure
```

Constructor Summary

Constructors
Constructor Description

RankLoss()

RankLoss(double tieLoss)
Create a Ranking Loss measure instance.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method	Description
`double`	`loss(java.util.List<? extends int[]> expected, java.util.List<? extends org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification> predicted)`	If this performance measure is originally a score function its score is transformed into a loss by multiplying the score with -1.

Methods inherited from class ai.libs.jaicore.ml.classification.multilabel.evaluation.loss.AMultiLabelClassificationMeasure
getThreshold, getThresholdedPredictionAsSet, listToMatrix, listToRelevanceMatrix, listToThresholdedRelevanceMatrix

Methods inherited from class ai.libs.jaicore.ml.classification.loss.dataset.APredictionPerformanceMeasure
averageInstanceWiseLoss, averageInstanceWiseScore, checkConsistency, loss, score, score

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.api4.java.ai.ml.core.evaluation.supervised.loss.IDeterministicPredictionPerformanceMeasure
loss, score, score

- Constructor Detail
  - RankLoss
```
public RankLoss()
```
  - RankLoss
```
public RankLoss(double tieLoss)
```
    Create a Ranking Loss measure instance.
    
    Parameters:
    
    tieLoss - The loss [0,1] which is accounted for a tie of the predicted relevant and irrelevant label's relevance score.
- Method Detail
  - loss
```
public double loss(java.util.List<? extends int[]> expected,
                   java.util.List<? extends org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification> predicted)
```
    Description copied from class: APredictionPerformanceMeasure
    
    If this performance measure is originally a score function its score is transformed into a loss by multiplying the score with -1. (loss=-score).
    
    Specified by:
    
    loss in interface org.api4.java.ai.ml.core.evaluation.supervised.loss.IDeterministicPredictionPerformanceMeasure<int[],org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification>
    
    Overrides:
    
    loss in class APredictionPerformanceMeasure<int[],org.api4.java.ai.ml.classification.multilabel.evaluation.IMultiLabelClassification>