org.apache.ignite.ml.selection.scoring.metric.classification

Class BinaryClassificationMetrics

java.lang.Object

org.apache.ignite.ml.selection.scoring.metric.AbstractMetrics<BinaryClassificationMetricValues>

org.apache.ignite.ml.selection.scoring.metric.classification.BinaryClassificationMetrics

All Implemented Interfaces:

Metric<Double>

public class BinaryClassificationMetrics
extends AbstractMetrics<BinaryClassificationMetricValues>

Binary classification metrics calculator. It could be used in two ways: to caculate all binary classification metrics or specific metric.

Field Summary

Fields inherited from class org.apache.ignite.ml.selection.scoring.metric.AbstractMetrics

metric

Constructor Summary

Constructors

Constructor and Description
BinaryClassificationMetrics()

Method Summary

Modifier and Type	Method and Description
boolean	isROCAUCenabled()
String	name() Returns the metric's name.
double	negativeClsLb()
double	positiveClsLb()
BinaryClassificationMetricValues	scoreAll(Iterator<LabelPair<Double>> iter) Calculates binary metrics values.
BinaryClassificationMetrics	withEnablingROCAUC(boolean enableROCAUC)
BinaryClassificationMetrics	withNegativeClsLb(double negativeClsLb)
BinaryClassificationMetrics	withPositiveClsLb(double positiveClsLb)

Methods inherited from class org.apache.ignite.ml.selection.scoring.metric.AbstractMetrics

score, withMetric

Methods inherited from class java.lang.Object

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

Constructor Detail

BinaryClassificationMetrics

public BinaryClassificationMetrics()

Method Detail

scoreAll

public BinaryClassificationMetricValues scoreAll(Iterator<LabelPair<Double>> iter)

Calculates binary metrics values.

Specified by:

scoreAll in class AbstractMetrics<BinaryClassificationMetricValues>

Parameters:

iter - Iterator that supplies pairs of truth values and predicated.

Returns:

Scores for all binary metrics.

positiveClsLb

public double positiveClsLb()

withPositiveClsLb

public BinaryClassificationMetrics withPositiveClsLb(double positiveClsLb)

negativeClsLb

public double negativeClsLb()

withNegativeClsLb

public BinaryClassificationMetrics withNegativeClsLb(double negativeClsLb)

withEnablingROCAUC

public BinaryClassificationMetrics withEnablingROCAUC(boolean enableROCAUC)

isROCAUCenabled

public boolean isROCAUCenabled()

name

public String name()

Returns the metric's name. NOTE: Should be unique to calculate multiple metrics correctly.

Returns:

String name representation.