public class RankingEvaluator extends Evaluator
Modifier and Type | Class and Description |
---|---|
static class |
RankingEvaluator.RepresentativeMetric |
Constructor and Description |
---|
RankingEvaluator() |
RankingEvaluator(double threshold) |
RankingEvaluator(double threshold,
RankingEvaluator.RepresentativeMetric representative) |
RankingEvaluator(double threshold,
String representative) |
Modifier and Type | Method and Description |
---|---|
double |
auroc()
Returns area under ROC curve.
|
void |
compute(TrainingMap trainingMap)
One of the main computation method.
|
void |
compute(TrainingMap trainingMap,
StandardPredicate predicate)
One of the main computation method.
|
String |
getAllStats()
Get a string that contains the full range of stats that this Evaluator can provide.
|
double |
getRepMetric()
The representative (rep) metric is the metric that was chosen to be the representative for this evaluator.
|
double |
getThreshold() |
boolean |
isHigherRepBetter()
Is a higher value for the current representative metric better?
|
double |
negativeAUPRC()
Returns area under the precision recall curve for the negative class.
|
double |
positiveAUPRC()
Returns area under the precision recall curve.
|
compute, compute, getNormalizedRepMetric
public RankingEvaluator()
public RankingEvaluator(double threshold)
public RankingEvaluator(double threshold, String representative)
public RankingEvaluator(double threshold, RankingEvaluator.RepresentativeMetric representative)
public void compute(TrainingMap trainingMap)
Evaluator
public void compute(TrainingMap trainingMap, StandardPredicate predicate)
Evaluator
public double getRepMetric()
Evaluator
getRepMetric
in class Evaluator
public boolean isHigherRepBetter()
Evaluator
isHigherRepBetter
in class Evaluator
public double getThreshold()
public double positiveAUPRC()
public double negativeAUPRC()
public double auroc()
public String getAllStats()
Evaluator
getAllStats
in class Evaluator
Copyright © 2020 University of California, Santa Cruz. All rights reserved.