keel.Algorithms.Genetic_Rule_Learning.GAssist

Class PerformanceAgent

java.lang.Object

keel.Algorithms.Genetic_Rule_Learning.GAssist.PerformanceAgent

public class PerformanceAgent
extends java.lang.Object

This class computes several performance measures taken from a fitness computation of an individual

Since:

JDK1.2

Version:

1.1

Author:

Written by Jaume Bacardit (La Salle, Ramón Llull University - Barcelona) 28/03/2004, Modified by Xavi Solé (La Salle, Ramón Llull University - Barcelona) 23/12/2008

Field Summary

Fields

Modifier and Type	Field and Description
static int[][]	confusionMatrix
static int[]	okClass
static int[]	totalClass
static int[]	utilRules

Constructor Summary

Constructors

Constructor and Description
PerformanceAgent()

Method Summary

Modifier and Type	Method and Description
static void	addPrediction(int predicted, int real, int activatedRule) Function used to inform PerformanceAgent of each example classified during the training stage
static void	addPredictionTest(int predicted, int real, int activatedRule) Function used to inform PerformanceAgent of each example classified during the test stage
static int[]	controlBloatRuleDeletion() This function computes the rules of an individual that will be deleted by the rule deletion operator, based on their activity during the previous fitness computation cycle
static void	dumpStats(java.lang.String typeOfTest) This function dumps the test statistics
static double	getAccuracy()
static int	getActivationsOfRule(int rule)
static double	getFitness(Classifier ind) This function returns the fitness formula used
static int	getNumAliveRules()
static void	resetPerformance(int _numRules)
static void	resetPerformanceTest(int _numRules) The test stage computes more statistics.

Methods inherited from class java.lang.Object

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

Field Detail

utilRules

public static int[] utilRules

okClass

public static int[] okClass

totalClass

public static int[] totalClass

confusionMatrix

public static int[][] confusionMatrix

Constructor Detail

PerformanceAgent

public PerformanceAgent()

Method Detail

controlBloatRuleDeletion

public static int[] controlBloatRuleDeletion()

This function computes the rules of an individual that will be deleted by the rule deletion operator, based on their activity during the previous fitness computation cycle

resetPerformance

public static void resetPerformance(int _numRules)

resetPerformanceTest

public static void resetPerformanceTest(int _numRules)

The test stage computes more statistics. The data structures that hold these statistics are initialized here

addPrediction

public static void addPrediction(int predicted,
                                 int real,
                                 int activatedRule)

Function used to inform PerformanceAgent of each example classified during the training stage

addPredictionTest

public static void addPredictionTest(int predicted,
                                     int real,
                                     int activatedRule)

Function used to inform PerformanceAgent of each example classified during the test stage

getAccuracy

public static double getAccuracy()

getNumAliveRules

public static int getNumAliveRules()

getActivationsOfRule

public static int getActivationsOfRule(int rule)

getFitness

public static double getFitness(Classifier ind)

This function returns the fitness formula used

dumpStats

public static void dumpStats(java.lang.String typeOfTest)

This function dumps the test statistics