yawn.nn
Class NeuralNetwork

java.lang.Object
  yawn.nn.NeuralNetwork

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

AppArt, FuzzyArtMap, MultiLayerPerceptron, NetworkCommittee, ParamerFitter, SmartMultiLayerPerceptron

public abstract class NeuralNetwork
extends java.lang.Object
implements java.io.Serializable

An abstract neural network $Id: NeuralNetwork.java,v 1.8 2005/04/07 17:28:28 supermarti Exp $

Version:

$Revision: 1.8 $

Author:

Luis Martí (luis dot marti at uc3m dot es)

See Also:

Serialized Form

Constructor Summary

NeuralNetwork()

Method Summary

protected boolean	getAdapting()
abstract int	getInputSize()
abstract java.lang.String	getNeuralNetworkName() Returns a human readable
abstract int	getOutputSize()
StatisticsFacility	getStatisticsFacility()
protected boolean	isAdapting()
abstract void	oneLearningStep(Pattern input, Pattern output) Performs one learning iteration.
abstract Pattern	predict(Pattern input) Computes a network prediction
protected void	setAdapting(boolean adapting) If set to true puts the network in adapting (training) mode; false puts the network in static or frozen (testing) mode.
void	setStatisticsFacility(StatisticsFacility statisticsFacility)
abstract void	setup(NeuralNetworkConfig config)
abstract void	train(InputOutputPattern[] iop) Trains the network until the stop criteria is met.
abstract NeuralNetworkConfig	yieldConfiguration()

Methods inherited from class java.lang.Object

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

Constructor Detail

NeuralNetwork

public NeuralNetwork()

Method Detail

oneLearningStep

public abstract void oneLearningStep(Pattern input,
                                     Pattern output)

Performs one learning iteration.

Parameters:

input - The network input.

output - The desired output.

predict

public abstract Pattern predict(Pattern input)

Computes a network prediction

Parameters:

input - The input to propagate.

Returns:

the network output.

train

public abstract void train(InputOutputPattern[] iop)

Trains the network until the stop criteria is met.

Parameters:

iop - The training set to be learned.

getInputSize

public abstract int getInputSize()

getOutputSize

public abstract int getOutputSize()

setup

public abstract void setup(NeuralNetworkConfig config)
                    throws ConfigurationException

Throws:

ConfigurationException

yieldConfiguration

public abstract NeuralNetworkConfig yieldConfiguration()

getStatisticsFacility

public StatisticsFacility getStatisticsFacility()

getNeuralNetworkName

public abstract java.lang.String getNeuralNetworkName()

Returns a human readable

Returns:

The name

isAdapting

protected boolean isAdapting()

Returns:

true if the network is adapting itself (training) or is frozen

getAdapting

protected boolean getAdapting()

Returns:

true if the network is adapting itself (training) or is frozen

setAdapting

protected void setAdapting(boolean adapting)

If set to true puts the network in adapting (training) mode; false puts the network in static or frozen (testing) mode.

Parameters:

adapting -

setStatisticsFacility

public void setStatisticsFacility(StatisticsFacility statisticsFacility)

Parameters:

statisticsFacility - The statisticsFacility to set.