net.sourceforge.cilib.neuralnetwork.basicFFNN
Class FFNNTopology

java.lang.Object
  net.sourceforge.cilib.neuralnetwork.basicFFNN.FFNNTopology

All Implemented Interfaces:

Initializable, NeuralNetworkTopology

public class FFNNTopology
extends Object
implements NeuralNetworkTopology

This is a basic Feedforward Neural network implementation. Its sole purpose is to illustrate the working of the CIlib Neural network framework and to show how it can be used to create a custom NN.

Author:

Stefan van der Stockt, Wiehann Matthysen

Field Summary

protected double[]

weights

Constructor Summary

FFNNTopology(int inputNr, int hiddenNr, int outputNr, double learn, double moment)

Method Summary

TypeList	evaluate(NNPattern p)
Vector	getWeights()
void	initialize()
void	setWeights(Vector w)
void	train()

Methods inherited from class java.lang.Object

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

Field Detail

weights

protected double[] weights

Constructor Detail

FFNNTopology

public FFNNTopology(int inputNr,
                    int hiddenNr,
                    int outputNr,
                    double learn,
                    double moment)

Method Detail

evaluate

public TypeList evaluate(NNPattern p)

Specified by:

evaluate in interface NeuralNetworkTopology

getWeights

public Vector getWeights()

Specified by:

getWeights in interface NeuralNetworkTopology

setWeights

public void setWeights(Vector w)

Specified by:

setWeights in interface NeuralNetworkTopology

train

public void train()

initialize

public void initialize()

Specified by:

initialize in interface Initializable