weka.classifiers.evaluation

Interface InformationTheoreticEvaluationMetric

public interface InformationTheoreticEvaluationMetric

Primarily a marker interface for information theoretic evaluation metrics to implement. Allows the command line interface to display these metrics or not based on user-supplied options

Version:

$Revision: 9320 $

Author:

Mark Hall (mhall{[at]}pentaho{[dot]}com)

Method Summary

Modifier and Type	Method and 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	updateStatsForConditionalDensityEstimator(ConditionalDensityEstimator classifier, Instance classMissing, double classValue) Updates stats for conditional density estimator based on current test instance.
void	updateStatsForPredictor(double predictedValue, Instance instance) Updates the statistics about a predictors performance for the current test instance.

Method Detail

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

updateStatsForConditionalDensityEstimator

void updateStatsForConditionalDensityEstimator(ConditionalDensityEstimator classifier,
                                               Instance classMissing,
                                               double classValue)
                                        throws java.lang.Exception

Updates stats for conditional density estimator based on current test instance. Gets called when the class is numeric and the classifier is a ConditionalDensityEstimators. 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:

classifier - the conditional density estimator

classMissing - the instance for which density is to be computed, without a class value

classValue - the class value of this instance

Throws:

java.lang.Exception - if density could not be computed successfully

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