Class LabelSequenceEvaluation

java.lang.Object

org.tribuo.classification.sequence.LabelSequenceEvaluation

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.provenance.Provenancable<EvaluationProvenance>, SequenceEvaluation<Label>

public class LabelSequenceEvaluation
extends Object
implements SequenceEvaluation<Label>

A class that can be used to evaluate a sequence label classification model element wise on a given set of data.

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	LabelSequenceEvaluation(Map<MetricID<Label>, Double> results, LabelMetric.Context ctx, EvaluationProvenance provenance)

Method Summary

Modifier and Type	Method	Description
double	accuracy()
double	accuracy(Label label)
Map<MetricID<Label>, Double>	asMap()	Get a map of all the metrics stored in this evaluation.
double	balancedErrorRate()
double	confusion(Label predictedLabel, Label trueLabel)	Note: confusion is not stored in the underlying map, so it won't show up in aggregation.
double	f1(Label label)
double	fn()
double	fn(Label label)
double	fp()
double	fp(Label label)
ConfusionMatrix<Label>	getConfusionMatrix()	Gets the confusion matrix backing this evaluation.
List<Prediction<Label>>	getPredictions()	Gets the flattened predictions.
EvaluationProvenance	getProvenance()
double	macroAveragedF1()
double	macroAveragedPrecision()
double	macroAveragedRecall()
double	macroFN()
double	macroFP()
double	macroTN()
double	macroTP()
double	microAveragedF1()
double	microAveragedPrecision()
double	microAveragedRecall()
double	precision(Label label)
double	recall(Label label)
double	tn()
double	tn(Label label)
String	toString()
double	tp()
double	tp(Label label)

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.tribuo.sequence.SequenceEvaluation

get

Constructor Details

LabelSequenceEvaluation

protected LabelSequenceEvaluation(Map<MetricID<Label>, Double> results,
 LabelMetric.Context ctx,
 EvaluationProvenance provenance)

Method Details

getPredictions

public List<Prediction<Label>> getPredictions()

Gets the flattened predictions.

Returns:

The flattened predictions.

getConfusionMatrix

public ConfusionMatrix<Label> getConfusionMatrix()

Gets the confusion matrix backing this evaluation.

Returns:

The confusion matrix.

asMap

public Map<MetricID<Label>, Double> asMap()

Description copied from interface: SequenceEvaluation

Get a map of all the metrics stored in this evaluation. The keys are metric id's and the values are their corresponding computed results.

Specified by:

asMap in interface SequenceEvaluation<Label>

Returns:

a map of all stored results

confusion

public double confusion(Label predictedLabel,
 Label trueLabel)

Note: confusion is not stored in the underlying map, so it won't show up in aggregation.

Parameters:

predictedLabel - The predicted label.

trueLabel - The true label.

Returns:

The number of times that predictedLabel was predicted for trueLabel.

tp

public double tp(Label label)

tp

public double tp()

macroTP

public double macroTP()

fp

public double fp(Label label)

fp

public double fp()

macroFP

public double macroFP()

tn

public double tn(Label label)

tn

public double tn()

macroTN

public double macroTN()

fn

public double fn(Label label)

fn

public double fn()

macroFN

public double macroFN()

precision

public double precision(Label label)

microAveragedPrecision

public double microAveragedPrecision()

macroAveragedPrecision

public double macroAveragedPrecision()

recall

public double recall(Label label)

microAveragedRecall

public double microAveragedRecall()

macroAveragedRecall

public double macroAveragedRecall()

f1

public double f1(Label label)

microAveragedF1

public double microAveragedF1()

macroAveragedF1

public double macroAveragedF1()

accuracy

public double accuracy()

accuracy

public double accuracy(Label label)

balancedErrorRate

public double balancedErrorRate()

getProvenance

public EvaluationProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<EvaluationProvenance>

toString

public String toString()

Overrides:

toString in class Object