org.encog.neural.neat
Class NEATNetwork

java.lang.Object
  org.encog.ml.BasicML
      org.encog.neural.neat.NEATNetwork

All Implemented Interfaces:

Serializable, MLContext, MLError, MLInput, MLInputOutput, MLMethod, MLOutput, MLProperties, MLRegression

public class NEATNetwork
extends BasicML
implements MLContext, MLRegression, MLError

Implements a NEAT network as a synapse between two layers. In Encog, a NEAT network is created by using a NEATSynapse between an input and output layer. NEAT networks only have an input and an output layer. There are no actual hidden layers. Rather this synapse will evolve many hidden neurons that have connections that are not easily defined by layers. Connections can be feedforward, recurrent, or self-connected. NEAT networks relieve the programmer of the need to define the hidden layer structure of the neural network. The output from the neural network can be calculated normally or using a snapshot. The snapshot mode is slower, but it can be more accurate. The snapshot handles recurrent layers better, as it takes the time to loop through the network multiple times to "flush out" the recurrent links. 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/

See Also:

Serialized Form

Field Summary

static String	PROPERTY_LINKS
static String	PROPERTY_NETWORK_DEPTH
static String	PROPERTY_SNAPSHOT

Constructor Summary

NEATNetwork()
Default constructor.

NEATNetwork(int inputCount, int outputCount)
Construct a NEAT network.

NEATNetwork(int inputCount, int outputCount, List<NEATNeuron> neurons, ActivationFunction activationFunction, ActivationFunction outputActivationFunction, int networkDepth)
Construct a NEAT synapse.

Method Summary

double	calculateError(MLDataSet data) Calculate the error for this neural network.
void	clearContext() Clear any context from previous runs.
MLData	compute(MLData input) Compute the output from this synapse.
ActivationFunction	getActivationFunction()
int	getInputCount()
int	getNetworkDepth()
List<NEATNeuron>	getNeurons()
ActivationFunction	getOutputActivationFunction()
int	getOutputCount()
boolean	isSnapshot()
void	setActivationFunction(ActivationFunction activationFunction) Set the activation function.
void	setInputCount(int i)
void	setNetworkDepth(int i)
void	setOutputActivationFunction(ActivationFunction outputActivationFunction)
void	setOutputCount(int i)
void	setSnapshot(boolean snapshot) Sets if snapshot is used.
void	updateProperties() Update any objeccts when a property changes.

Methods inherited from class org.encog.ml.BasicML

getProperties, getPropertyDouble, getPropertyLong, getPropertyString, setProperty, setProperty, setProperty

Methods inherited from class java.lang.Object

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

Field Detail

PROPERTY_NETWORK_DEPTH

public static final String PROPERTY_NETWORK_DEPTH

See Also:

Constant Field Values

PROPERTY_LINKS

public static final String PROPERTY_LINKS

See Also:

Constant Field Values

PROPERTY_SNAPSHOT

public static final String PROPERTY_SNAPSHOT

See Also:

Constant Field Values

Constructor Detail

NEATNetwork

public NEATNetwork()

Default constructor.

NEATNetwork

public NEATNetwork(int inputCount,
                   int outputCount,
                   List<NEATNeuron> neurons,
                   ActivationFunction activationFunction,
                   ActivationFunction outputActivationFunction,
                   int networkDepth)

Construct a NEAT synapse.

Parameters:

inputCount - The number of input neurons.

outputCount - The number of output neurons.

neurons - The neurons in this synapse.

activationFunction - The activation function to use.

networkDepth - The depth of the network.

NEATNetwork

public NEATNetwork(int inputCount,
                   int outputCount)

Construct a NEAT network.

Parameters:

inputCount - The input count.

outputCount - The output count.

Method Detail

clearContext

public void clearContext()

Clear any context from previous runs. This sets the activation of all neurons to zero.

Specified by:

clearContext in interface MLContext

compute

public MLData compute(MLData input)

Compute the output from this synapse.

Specified by:

compute in interface MLRegression

Parameters:

input - The input to this synapse.

Returns:

The output from this synapse.

getActivationFunction

public ActivationFunction getActivationFunction()

Returns:

The activation function.

getNetworkDepth

public int getNetworkDepth()

Returns:

The network depth.

getNeurons

public List<NEATNeuron> getNeurons()

Returns:

The NEAT neurons.

isSnapshot

public boolean isSnapshot()

Returns:

True if snapshot is being used.

setActivationFunction

public void setActivationFunction(ActivationFunction activationFunction)

Set the activation function.

Parameters:

activationFunction - The activation function.

setSnapshot

public void setSnapshot(boolean snapshot)

Sets if snapshot is used.

Parameters:

snapshot - True if snapshot is used.

getInputCount

public int getInputCount()

Specified by:

getInputCount in interface MLInput

Returns:

The input.

getOutputCount

public int getOutputCount()

Specified by:

getOutputCount in interface MLOutput

Returns:

The output count.

updateProperties

public void updateProperties()

Description copied from interface: MLProperties

Update any objeccts when a property changes.

Specified by:

updateProperties in interface MLProperties

Specified by:

updateProperties in class BasicML

setInputCount

public void setInputCount(int i)

setOutputCount

public void setOutputCount(int i)

setNetworkDepth

public void setNetworkDepth(int i)

getOutputActivationFunction

public ActivationFunction getOutputActivationFunction()

Returns:

the outputActivationFunction

setOutputActivationFunction

public void setOutputActivationFunction(ActivationFunction outputActivationFunction)

Parameters:

outputActivationFunction - the outputActivationFunction to set

calculateError

public double calculateError(MLDataSet data)

Calculate the error for this neural network.

Specified by:

calculateError in interface MLError

Parameters:

data - The training set.

Returns:

The error percentage.