org.encog.neural.networks.training.neat
Class NEATNeuronGene

java.lang.Object
  org.encog.solve.genetic.genes.BasicGene
      org.encog.neural.networks.training.neat.NEATNeuronGene

All Implemented Interfaces:

Comparable<Gene>, Gene

public class NEATNeuronGene
extends BasicGene

Implements a NEAT neuron gene. NeuroEvolution of Augmenting Topologies (NEAT) is a genetic algorithm for the generation of evolving artificial neural networks. It was developed by Ken Stanley while at The University of Texas at Austin. http://www.cs.ucf.edu/~kstanley/

Constructor Summary

NEATNeuronGene()
The default constructor.

NEATNeuronGene(NEATNeuronType type, long id, double splitY, double splitX)
Construct a gene.

NEATNeuronGene(NEATNeuronType type, long id, double splitY, double splitX, boolean recurrent, double act)
Construct a neuron gene.

Method Summary

void	copy(Gene gene) Copy another gene to this one.
double	getActivationResponse()
NEATNeuronType	getNeuronType()
double	getSplitX()
double	getSplitY()
boolean	isRecurrent()
void	setActivationResponse(double activationResponse) Set the activation response.
void	setNeuronType(NEATNeuronType neuronType) Set the neuron type.
void	setRecurrent(boolean recurrent) Set if this is a recurrent neuron.
void	setSplitX(double splitX) Set the split x.
void	setSplitY(double splitY) Set the split y.

Methods inherited from class org.encog.solve.genetic.genes.BasicGene

compareTo, getId, getInnovationId, isEnabled, setEnabled, setId, setInnovationId

Methods inherited from class java.lang.Object

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

Constructor Detail

NEATNeuronGene

public NEATNeuronGene()

The default constructor.

NEATNeuronGene

public NEATNeuronGene(NEATNeuronType type,
                      long id,
                      double splitY,
                      double splitX)

Construct a gene.

Parameters:

type - The type of neuron.

id - The id of this gene.

splitY - The split y.

splitX - The split x.

NEATNeuronGene

public NEATNeuronGene(NEATNeuronType type,
                      long id,
                      double splitY,
                      double splitX,
                      boolean recurrent,
                      double act)

Construct a neuron gene.

Parameters:

type - The type of neuron.

id - The id of this gene.

splitY - The split y.

splitX - The split x.

recurrent - True if this is a recurrent link.

act - The activation response.

Method Detail

copy

public void copy(Gene gene)

Copy another gene to this one.

Parameters:

gene - The other gene.

getActivationResponse

public double getActivationResponse()

Returns:

The activation response.

getNeuronType

public NEATNeuronType getNeuronType()

Returns:

The type for this neuron.

getSplitX

public double getSplitX()

Returns:

The split x value.

getSplitY

public double getSplitY()

Returns:

The split y value.

isRecurrent

public boolean isRecurrent()

Returns:

True if this is recurrent.

setActivationResponse

public void setActivationResponse(double activationResponse)

Set the activation response.

Parameters:

activationResponse - The activation response.

setNeuronType

public void setNeuronType(NEATNeuronType neuronType)

Set the neuron type.

Parameters:

neuronType - The neuron type.

setRecurrent

public void setRecurrent(boolean recurrent)

Set if this is a recurrent neuron.

Parameters:

recurrent - True if this is a recurrent neuron.

setSplitX

public void setSplitX(double splitX)

Set the split x.

Parameters:

splitX - The split x.

setSplitY

public void setSplitY(double splitY)

Set the split y.

Parameters:

splitY - The split y.