Class ClassificationMetrics

java.lang.Object
com.knezevic.edaf.v3.problems.grammar.ClassificationMetrics
public final class ClassificationMetrics extends Object
Utility functions for single-label classification metric computation.

  • Method Summary

    Modifier and Type
    Method
    Description
    static double
    accuracy(int[][] confusion)
    Computes plain classification accuracy from confusion matrix.
    static double
    binaryF1(int[][] confusion, int positiveClassIndex)
    Computes F1 score for a selected positive class in one-vs-rest view.
    static double
    macroF1(int[][] confusion)
    Computes macro averaged F1 score across all classes.

    Methods inherited from class java.lang.Object

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

  • Method Details

    • accuracy

      public static double accuracy(int[][] confusion)
      Computes plain classification accuracy from confusion matrix.

    • macroF1

      public static double macroF1(int[][] confusion)
      Computes macro averaged F1 score across all classes.

    • binaryF1

      public static double binaryF1(int[][] confusion, int positiveClassIndex)
      Computes F1 score for a selected positive class in one-vs-rest view.