org.vizzini.ai.geneticalgorithm
Class ChromosomeNeuralNetwork

java.lang.Object
  org.vizzini.ai.geneticalgorithm.AbstractChromosome
      org.vizzini.ai.geneticalgorithm.ChromosomeDouble
          org.vizzini.ai.geneticalgorithm.ChromosomeNeuralNetwork

All Implemented Interfaces:

Cloneable, Comparable, IChromosome

public class ChromosomeNeuralNetwork
extends ChromosomeDouble

Provides a chromosome which contains DefaultNeuralNetwork s.

Since:

v0.1

Version:

v0.3

Author:

Jeffrey M. Thompson

Field Summary

Fields inherited from interface org.vizzini.ai.geneticalgorithm.IChromosome

UNKNOWN_FITNESS

Constructor Summary

ChromosomeNeuralNetwork(INeuralNetwork network)
Construct this object using the given parameter.

Method Summary

static void	chromosomeToNetwork(ChromosomeDouble chromosome, INeuralNetwork network) Set the neural network's weights from the chromosome.
static int	computeSize(INeuralNetwork network) Compute the chromosome size based upon the given neural network.
void	copy(IChromosome another) Copy the given chromosome into this.
static int	getIndex(INeuralNetwork network, int layerIndex, int i, int j) Return the gene index of the given neural network coordinates.
INeuralNetwork	getNeuralNetwork() Return the neural network.
boolean	isDirty() Return true if this chromosome is out of sync with the neural network.
static void	networkToChromosome(INeuralNetwork network, ChromosomeDouble chromosome) Set the chromosome's genes from the neural network weights.
void	setFitnessUnknown() Set the fitness to unknown.
String	toString() Return a string representation of this object.

Methods inherited from class org.vizzini.ai.geneticalgorithm.ChromosomeDouble

clone, equals, get, hashCode, mix, mix, mutate, randomlyInitialize, set, setRandomMax, setRandomMin, size, splice

Methods inherited from class org.vizzini.ai.geneticalgorithm.AbstractChromosome

addFitness, compareTo, getDescription, getFitness, isFitnessUnknown, setDescription, setFitness

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

ChromosomeNeuralNetwork

public ChromosomeNeuralNetwork(INeuralNetwork network)

Construct this object using the given parameter.

Since:

v0.1

Method Detail

chromosomeToNetwork

public static void chromosomeToNetwork(ChromosomeDouble chromosome,
                                       INeuralNetwork network)

Set the neural network's weights from the chromosome.

Since:

v0.1

computeSize

public static int computeSize(INeuralNetwork network)

Compute the chromosome size based upon the given neural network.

Since:

v0.1

getIndex

public static int getIndex(INeuralNetwork network,
                           int layerIndex,
                           int i,
                           int j)

Return the gene index of the given neural network coordinates.

Since:

v0.1

networkToChromosome

public static void networkToChromosome(INeuralNetwork network,
                                       ChromosomeDouble chromosome)

Set the chromosome's genes from the neural network weights.

Since:

v0.1

copy

public void copy(IChromosome another)

Copy the given chromosome into this. Extend the super method to clone the neural network.

Specified by:

copy in interface IChromosome

Overrides:

copy in class ChromosomeDouble

Since:

v0.1

getNeuralNetwork

public INeuralNetwork getNeuralNetwork()

Return the neural network.

Since:

v0.1

isDirty

public boolean isDirty()

Return true if this chromosome is out of sync with the neural network.

Since:

v0.1

setFitnessUnknown

public void setFitnessUnknown()

Set the fitness to unknown. Extend the super method to set the dirty flag.

Specified by:

setFitnessUnknown in interface IChromosome

Overrides:

setFitnessUnknown in class AbstractChromosome

Since:

v0.1

toString

public String toString()

Return a string representation of this object. Override the super method to avoid printing the genes.

Overrides:

toString in class ChromosomeDouble

Since:

v0.1