jenes.stage.operator

Class Crowder<T extends Chromosome>

java.lang.Object

jenes.stage.AbstractStage<T>

jenes.stage.operator.Operator<T>

jenes.stage.operator.Crowder<T>

Direct Known Subclasses:

DeJongCrowder, DeterministicCrowder, MultiNicheCrowder, SteadyState

public abstract class Crowder<T extends Chromosome>
extends Operator<T>

This class provides abstraction to crowders. A crowder perform preselection before its body processes the population, and controls the replacement after.

Since:

2.0

Version:

2.0

Nested Class Summary

Nested classes/interfaces inherited from class jenes.stage.operator.Operator

Operator.Statistics

Field Summary

Fields

Modifier and Type	Field and Description
protected Sequence<T>	body Body
protected boolean	elitist Elitism

Fields inherited from class jenes.stage.operator.Operator

random, statistics

Fields inherited from class jenes.stage.AbstractStage

biggerIsBetter, fitness, ga

Constructor Summary

Constructors

Constructor and Description
Crowder() Creates a new crowder

Method Summary

Methods

Modifier and Type	Method and Description
Sequence<T>	getBody() Return the crowder body
void	init(GeneticAlgorithm<T> ga) Initializes this stage according to the genetic algorithm that uses it
boolean	isElitist() Says if the crowder is elitist or not.
protected abstract void	preselect(Population<T> in, Population<T> out) Preselection of individuals before processing.
void	process(Population<T> in, Population<T> out) Performs crowding processing, according the the following scheme: pre = preselect(in) evo = body.process(pre) out = replace(evo)
protected abstract void	replace(Population<T> initial, Population<T> preselected, Population<T> evolved, Population<T> out) Implements the replacement policy.
void	setBiggerIsBetter(boolean flag, boolean recursively) Deprecated.
void	setElitist(boolean flag) Sets the crowder elistism.
void	setFitness(Fitness fit, boolean recursively) Change the Fitness to this stage propagating the change recursively according to the flag given as parameter
double	similarity(double[] diff) Computes a degree of similarity given the genetic difference vector between two chromosomes.

Methods inherited from class jenes.stage.operator.Operator

getStatistics, updateStatistics

Methods inherited from class jenes.stage.AbstractStage

dispose, getFitness, isBiggerBetter, isFitnessChanged, setBiggerIsBetter, setFitness

Methods inherited from class java.lang.Object

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

Field Detail

body

protected Sequence<T extends Chromosome> body

Body

elitist

protected boolean elitist

Elitism

Constructor Detail

Crowder

public Crowder()

Creates a new crowder

Method Detail

isElitist

public final boolean isElitist()

Says if the crowder is elitist or not.

Returns:

setElitist

public void setElitist(boolean flag)

Sets the crowder elistism. If true, offspings replace parents only if they are better.

Parameters:

flag -

init

public void init(GeneticAlgorithm<T> ga)

Description copied from class: AbstractStage

Initializes this stage according to the genetic algorithm that uses it

Overrides:

init in class AbstractStage<T extends Chromosome>

Parameters:

ga - the Genetic Algorithm in wchic this stage run

process

public void process(Population<T> in,
           Population<T> out)
             throws StageException

Performs crowding processing, according the the following scheme:

1. pre = preselect(in)
2. evo = body.process(pre)
3. out = replace(evo)

Specified by:

process in class AbstractStage<T extends Chromosome>

Parameters:

in -

out -

Throws:

StageException

preselect

protected abstract void preselect(Population<T> in,
             Population<T> out)

Preselection of individuals before processing.

Parameters:

in - input population

out - output population

replace

protected abstract void replace(Population<T> initial,
           Population<T> preselected,
           Population<T> evolved,
           Population<T> out)

Implements the replacement policy.

Parameters:

initial - - the initial population

preselected - - the population resulting from preselection

evolved - - the part of population that has evolved

out - - the output population

similarity

public double similarity(double[] diff)

Computes a degree of similarity given the genetic difference vector between two chromosomes. By default, similarity is computed as reprocical of eucledian distance between genes. Other definitions are possible by overriding this method in subclasses.

Parameters:

diff - - the difference vector

Returns:

similarity degree

setBiggerIsBetter

@Deprecated
public void setBiggerIsBetter(boolean flag,
                                boolean recursively)

Deprecated.

Overrides:

setBiggerIsBetter in class AbstractStage<T extends Chromosome>

Parameters:

flag - true, if the best individual has the higher fitness

recursively - true, to propagate this setting down, otherwise false.

setFitness

public void setFitness(Fitness fit,
              boolean recursively)

Description copied from class: AbstractStage

Change the Fitness to this stage propagating the change recursively according to the flag given as parameter

Overrides:

setFitness in class AbstractStage<T extends Chromosome>

getBody

public Sequence<T> getBody()

Return the crowder body

Returns:

the body