org.apache.ignite.ml.selection.scoring.evaluator.aggregator

Class BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>

java.lang.Object

org.apache.ignite.ml.selection.scoring.evaluator.aggregator.BinaryClassificationPointwiseMetricStatsAggregator<L>

All Implemented Interfaces:

Serializable, MetricStatsAggregator<L,BinaryClassificationEvaluationContext<L>,BinaryClassificationPointwiseMetricStatsAggregator<L>>

Direct Known Subclasses:

BinaryClassificationPointwiseMetricStatsAggregator.WithCustomLabelsAggregator

public class BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>
extends Object
implements MetricStatsAggregator<L,BinaryClassificationEvaluationContext<L>,BinaryClassificationPointwiseMetricStatsAggregator<L>>

This class represents statistics for pointwise metrics evaluation for binary classification like TruePositive, FalsePositive, TrueNegative and FalseNegative.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	BinaryClassificationPointwiseMetricStatsAggregator.WithCustomLabelsAggregator<L extends Serializable> Class represents already initialized aggregator.

Constructor Summary

Constructors

Constructor and Description
BinaryClassificationPointwiseMetricStatsAggregator() Creates an instance of BinaryClassificationPointwiseMetricStatsAggregator.
BinaryClassificationPointwiseMetricStatsAggregator(L falseLabel, L truthLabel, int truePositive, int falsePositive, int trueNegative, int falseNegative) Creates an instance of BinaryClassificationPointwiseMetricStatsAggregator.

Method Summary

Modifier and Type	Method and Description
void	aggregate(IgniteModel<Vector,L> mdl, LabeledVector<L> vector) Aggregates statistics for metric computation given model and vector with answer.
BinaryClassificationEvaluationContext<L>	createInitializedContext() Returns initialized context.
L	getFalseLabel() Returns false label.
int	getFalseNegative() Returns false negatives count.
int	getFalsePositive() Returns false positives count.
int	getN() Returns number of elements in dataset.
int	getTrueNegative() Returns true negatives count.
int	getTruePositive() Returns true positives count.
L	getTruthLabel() Returns truth label.
void	initByContext(BinaryClassificationEvaluationContext<L> ctx) Inits this aggtegator by evaluation context.
BinaryClassificationPointwiseMetricStatsAggregator<L>	mergeWith(BinaryClassificationPointwiseMetricStatsAggregator other) Merges statistics of two aggregators to new aggreagator.

Methods inherited from class java.lang.Object

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

Constructor Detail

BinaryClassificationPointwiseMetricStatsAggregator

public BinaryClassificationPointwiseMetricStatsAggregator()

Creates an instance of BinaryClassificationPointwiseMetricStatsAggregator.

BinaryClassificationPointwiseMetricStatsAggregator

public BinaryClassificationPointwiseMetricStatsAggregator(L falseLabel,
                                                          L truthLabel,
                                                          int truePositive,
                                                          int falsePositive,
                                                          int trueNegative,
                                                          int falseNegative)

Creates an instance of BinaryClassificationPointwiseMetricStatsAggregator.

Parameters:

falseLabel - False label.

truthLabel - Truth label.

truePositive - True positives amount.

falsePositive - False positives amount.

trueNegative - True negatives amount.

falseNegative - False negatives amount.

Method Detail

aggregate

public void aggregate(IgniteModel<Vector,L> mdl,
                      LabeledVector<L> vector)

Aggregates statistics for metric computation given model and vector with answer.

Specified by:

aggregate in interface MetricStatsAggregator<L extends Serializable,BinaryClassificationEvaluationContext<L extends Serializable>,BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>>

Parameters:

mdl - Model.

vector - Vector.

mergeWith

public BinaryClassificationPointwiseMetricStatsAggregator<L> mergeWith(BinaryClassificationPointwiseMetricStatsAggregator other)

Merges statistics of two aggregators to new aggreagator.

Specified by:

mergeWith in interface MetricStatsAggregator<L extends Serializable,BinaryClassificationEvaluationContext<L extends Serializable>,BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>>

Parameters:

other - Other aggregator.

Returns:

New aggregator.

createInitializedContext

public BinaryClassificationEvaluationContext<L> createInitializedContext()

Returns initialized context.

Specified by:

createInitializedContext in interface MetricStatsAggregator<L extends Serializable,BinaryClassificationEvaluationContext<L extends Serializable>,BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>>

Returns:

initialized evaluation context.

initByContext

public void initByContext(BinaryClassificationEvaluationContext<L> ctx)

Inits this aggtegator by evaluation context.

Specified by:

initByContext in interface MetricStatsAggregator<L extends Serializable,BinaryClassificationEvaluationContext<L extends Serializable>,BinaryClassificationPointwiseMetricStatsAggregator<L extends Serializable>>

Parameters:

ctx - Evaluation context.

getFalseLabel

public L getFalseLabel()

Returns false label.

Returns:

False label.

getTruthLabel

public L getTruthLabel()

Returns truth label.

Returns:

Truth label.

getTruePositive

public int getTruePositive()

Returns true positives count.

Returns:

True positives count.

getFalsePositive

public int getFalsePositive()

Returns false positives count.

Returns:

False positives count.

getTrueNegative

public int getTrueNegative()

Returns true negatives count.

Returns:

True negatives count.

getFalseNegative

public int getFalseNegative()

Returns false negatives count.

Returns:

False negatives count.

getN

public int getN()

Returns number of elements in dataset.

Returns:

Number of elements in dataset.