trainableSegmentation.metrics

Class ClassificationStatistics

java.lang.Object

trainableSegmentation.metrics.ClassificationStatistics

public class ClassificationStatistics
extends Object

This class stores statistics from a classification

Field Summary

Fields

Modifier and Type	Field and Description
double	falseNegatives number of false negatives
double	falsePositives number of false positives
double	fScore F-score, harmonic mean of precision and recall
double	metricValue value of the classification metric
double	precision precision: true positives / ( true positives + false positives )
double	recall recall (also called sensitivity of hit rate): true positives / ( true positives + false negatives )
double	specificity specificity, also called true negative rate (TNR): true negatives / (true negatives + false negatives)
double	trueNegatives number of true negatives
double	truePositives number of true positives

Constructor Summary

Constructors

Constructor and Description
ClassificationStatistics(double truePositives, double trueNegatives, double falsePositives, double falseNegatives, double metricValue) Create classification statistics

Method Summary

Methods inherited from class java.lang.Object

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

Field Detail

truePositives

public double truePositives

number of true positives

trueNegatives

public double trueNegatives

number of true negatives

falsePositives

public double falsePositives

number of false positives

falseNegatives

public double falseNegatives

number of false negatives

metricValue

public double metricValue

value of the classification metric

precision

public double precision

precision: true positives / ( true positives + false positives )

recall

public double recall

recall (also called sensitivity of hit rate): true positives / ( true positives + false negatives )

fScore

public double fScore

F-score, harmonic mean of precision and recall

specificity

public double specificity

specificity, also called true negative rate (TNR): true negatives / (true negatives + false negatives)

Constructor Detail

ClassificationStatistics

public ClassificationStatistics(double truePositives,
                                double trueNegatives,
                                double falsePositives,
                                double falseNegatives,
                                double metricValue)

Create classification statistics

Parameters:

truePositives - number of true positives

trueNegatives - number of true negatives

falsePositives - number of false positives

falseNegatives - number of false negatives

metricValue - value of the classification metric