org.encog.neural.pattern
Class RSOMPattern

java.lang.Object
  org.encog.neural.pattern.RSOMPattern

All Implemented Interfaces:

NeuralNetworkPattern

public class RSOMPattern
extends java.lang.Object
implements NeuralNetworkPattern

A recurrent self organizing map is a self organizing map that has a recurrent context connection on the hidden layer. This type of neural network is adept at classifying temporal data.

Author:

jheaton

Constructor Summary

RSOMPattern()

Method Summary

void	addHiddenLayer(int count) Add a hidden layer.
BasicNetwork	generate() Generate the RSOM network.
void	setActivationFunction(ActivationFunction activation) Set the activation function.
void	setInputNeurons(int count) Set the input neuron count.
void	setOutputNeurons(int count) Set the output neuron count.

Methods inherited from class java.lang.Object

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

Constructor Detail

RSOMPattern

public RSOMPattern()

Method Detail

addHiddenLayer

public void addHiddenLayer(int count)

Add a hidden layer. SOM networks do not have hidden layers, so this will throw an error.

Specified by:

addHiddenLayer in interface NeuralNetworkPattern

Parameters:

count - The number of hidden neurons.

generate

public BasicNetwork generate()

Generate the RSOM network.

Specified by:

generate in interface NeuralNetworkPattern

Returns:

The neural network.

setActivationFunction

public void setActivationFunction(ActivationFunction activation)

Set the activation function. A SOM uses a linear activation function, so this method throws an error.

Specified by:

setActivationFunction in interface NeuralNetworkPattern

Parameters:

activation - The activation function to use.

setInputNeurons

public void setInputNeurons(int count)

Set the input neuron count.

Specified by:

setInputNeurons in interface NeuralNetworkPattern

Parameters:

count - The number of neurons.

setOutputNeurons

public void setOutputNeurons(int count)

Set the output neuron count.

Specified by:

setOutputNeurons in interface NeuralNetworkPattern

Parameters:

count - The number of neurons.