Package weka.classifiers.evaluation

Interface StandardEvaluationMetric

  • public interface StandardEvaluationMetric
    Primarily a marker interface for a "standard" evaluation metric - i.e. one that would be part of the normal output in Weka without having to turn specific display options.
    Version:
    $Revision: 9320 $
    Author:
    Mark Hall (mhall{[at]}pentaho{[dot]}com)

    • Method Summary

      All Methods Instance Methods Abstract Methods 
      Modifier and Type Method Description
      java.lang.String toSummaryString()
      Return a formatted string (suitable for displaying in console or GUI output) containing all the statistics that this metric computes.
      void updateStatsForClassifier​(double[] predictedDistribution, Instance instance)
      Updates the statistics about a classifiers performance for the current test instance.
      void updateStatsForPredictor​(double predictedValue, Instance instance)
      Updates the statistics about a predictors performance for the current test instance.

    • Method Detail

      • toSummaryString

        java.lang.String toSummaryString()
        Return a formatted string (suitable for displaying in console or GUI output) containing all the statistics that this metric computes.
        Returns:
        a formatted string containing all the computed statistics

      • updateStatsForClassifier

        void updateStatsForClassifier​(double[] predictedDistribution,
                              Instance instance)
                       throws java.lang.Exception
        Updates the statistics about a classifiers performance for the current test instance. Gets called when the class is nominal. Implementers need only implement this method if it is not possible to compute their statistics from what is stored in the base Evaluation object.
        Parameters:
        predictedDistribution - the probabilities assigned to each class
        instance - the instance to be classified
        Throws:
        java.lang.Exception - if the class of the instance is not set

      • updateStatsForPredictor

        void updateStatsForPredictor​(double predictedValue,
                             Instance instance)
                      throws java.lang.Exception
        Updates the statistics about a predictors performance for the current test instance. Gets called when the class is numeric. Implementers need only implement this method if it is not possible to compute their statistics from what is stored in the base Evaluation object.
        Parameters:
        predictedValue - the numeric value the classifier predicts
        instance - the instance to be classified
        Throws:
        java.lang.Exception - if the class of the instance is not set