keel.Algorithms.Neural_Networks.NNEP_Clas.algorithm.classification

Class CCRElitistNeuralNetAlgorithm<I extends <any>>

java.lang.Object

<any>

keel.Algorithms.Neural_Networks.NNEP_Common.algorithm.NeuralNetAlgorithm<I>

keel.Algorithms.Neural_Networks.NNEP_Clas.algorithm.classification.CCRElitistNeuralNetAlgorithm<I>

Type Parameters:

I - Object Type which should extend from AbstractIndividual

public class CCRElitistNeuralNetAlgorithm<I extends <any>>
extends NeuralNetAlgorithm<I>

Base implementation for all classification neural net algorithms.

Since:

JDK1.5

Version:

0.1

Author:

Written by Pedro Antonio Gutierrez Penia (University of Cordoba) 16/7/2007

Field Summary

Fields

Modifier and Type	Field and Description
protected I	bestCCRIndividual Best individual of last generation

Fields inherited from class keel.Algorithms.Neural_Networks.NNEP_Common.algorithm.NeuralNetAlgorithm

bestIndividual, bset, cratio, cset, currentBest, currentMean, evaluator, fitDif, generation, individualsComparator, maxOfGenerations, mogbest, mogmean, mutator1, mutator2, nofselSecondMutator, nogbest, nogmean, percentageSecondMutator, populationSize, previousBest, previousMean, provider, pset1, pset2, randGenFactory, rset, species

Constructor Summary

Constructors

Constructor and Description
CCRElitistNeuralNetAlgorithm() Empty constructor

Method Summary

Modifier and Type	Method and Description
void	doInit() Create individuals in population, evaluating before start rest of evolution
protected void	doUpdate() Implementation for the Update task
I	getBestCCRIndividual() Returns the best individual of the population
protected I	obtainBestClassifier(java.util.List<I> individuals, IEvaluator evaluator) Returns best individual of a list of individuals using the MSEErrorFunction

Methods inherited from class keel.Algorithms.Neural_Networks.NNEP_Common.algorithm.NeuralNetAlgorithm

configure, createRandGen, doControl, doGeneration, doIterate, doReplacement, doSelection, doSort, equals, getBestIndividual, getCratio, getCurrentBest, getCurrentMean, getEvaluator, getFitDif, getGeneration, getInhabitants, getMaxOfGenerations, getMogbest, getMogmean, getMutator1, getMutator2, getNogbest, getNogmean, getPercentageSecondMutator, getPopulationSize, getPreviousBest, getPreviousMean, getProvider, getRandGenFactory, getSpecies, setCratio, setEvaluator, setFitDif, setGeneration, setInhabitants, setMaxOfGenerations, setMogbest, setMogmean, setMutator1, setMutator2, setPercentageSecondMutator, setPopulationSize, setProvider, setRandGenFactory, setSpecies

Methods inherited from class java.lang.Object

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

Field Detail

bestCCRIndividual

protected I extends <any> bestCCRIndividual

Best individual of last generation

Constructor Detail

CCRElitistNeuralNetAlgorithm

public CCRElitistNeuralNetAlgorithm()

Empty constructor

Method Detail

getBestCCRIndividual

public I getBestCCRIndividual()

Returns the best individual of the population

Returns:

I Best individual of the population

doInit

public void doInit()

Create individuals in population, evaluating before start rest of evolution

Overrides:

doInit in class NeuralNetAlgorithm<I extends <any>>

doUpdate

protected void doUpdate()

Implementation for the Update task

Overrides:

doUpdate in class NeuralNetAlgorithm<I extends <any>>

obtainBestClassifier

protected I obtainBestClassifier(java.util.List<I> individuals,
                                 IEvaluator evaluator)

Returns best individual of a list of individuals using the MSEErrorFunction

Parameters:

individuals - list of individuals to obtain best

evaluator - NeuralNetEvaluator to use in individual evaluation

Returns:

I Best classifier of the ensemble