art.ga
Class GaManager

java.lang.Object
  art.ga.GaManager

Direct Known Subclasses:

CsManager

public class GaManager

extends java.lang.Object

This is the only class that have to by accessed directly by the user in order to easily manage ART as a black box; it hides the internal structure of the genetic algorithm.

Author:

Marco Lamieri and Gianluigi Ferraris

See Also:

GaSolution, Fitness

Constructor Summary

GaManager(java.lang.String ConfigurationFileName, java.lang.String project)
Constructor of the class

Method Summary

GaSolution	computeBestSolution(Fitness aFitnessFunction)
int	getGenerationNumber() Get the generation currently computed by the ga.
GaSolution	getNextSolution() Get the next GaSolution generated by the genetic algorithm.
void	setFitness(GaSolution gaStrategy, double fitness) Set the fitness (reward) for a given GaSolution.

Methods inherited from class java.lang.Object

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

Constructor Detail

GaManager

public GaManager(java.lang.String ConfigurationFileName,
                 java.lang.String project)

Constructor of the class

Parameters:

ConfigurationFileName - Parameter's file

project - Parameter's project

Method Detail

computeBestSolution

public GaSolution computeBestSolution(Fitness aFitnessFunction)

getGenerationNumber

public int getGenerationNumber()

Get the generation currently computed by the ga.

Returns:

Current generation number

getNextSolution

public GaSolution getNextSolution()

Get the next GaSolution generated by the genetic algorithm. This method is used in combination with setFitness(GaSolution, double) in AB simulations or in dynamic problems evolving over time for which is not possible define a fitness function implementing the Fitness interface.

Returns:

A solution that have to be evaluated

setFitness

public void setFitness(GaSolution gaStrategy,
                       double fitness)

Set the fitness (reward) for a given GaSolution. This method is used in combination with getNextSolution() in AB simulations or in dynamic problems evolving over time for wich is not possible define a fitness function implementing the Fitness interface.

Parameters:

gaStrategy - Startegy evaluated

fitness - Reward of the startegy