boone.training

Class HebbTrainer

java.lang.Object

boone.Trainer

boone.training.HebbTrainer

All Implemented Interfaces:

Storable, java.lang.Cloneable

public class HebbTrainer
extends Trainer

The Hebbian trainer is of more theoretical importance, as it simply uses the basic Hebb update rule to change the weights and biases of the net. As this is purely unsupervised there is no explicit feedback to training.

Author:

August Mayer, Helmut A. Mayer

Field Summary

Fields inherited from class boone.Trainer

currentEpoch, epochs, learnRate, net, shuffle, stepMode, testData, trainingData, trainingSignalGenerator

Constructor Summary

Constructors

Constructor and Description
HebbTrainer()

Method Summary

Modifier and Type	Method and Description
protected double	calcAdaptation(BrainPart part) Not used.
protected void	endBatch() Does nothing.
protected void	endTrain() Does nothing.
void	fromXML(org.jdom2.Element element) Reads attributes from XML.
protected void	reset() Does nothing.
void	train(java.util.List<java.lang.Double> input, java.util.List<java.lang.Double> target) Hebbian learning of a single pattern.

Methods inherited from class boone.Trainer

clone, getBatchSize, getEpochs, getLearnRate, getNetwork, getRank, getTestData, getTrainingData, getTrainingSignalGenerator, getWinningNeuron, setBatchSize, setEpochs, setLearnRate, setNetwork, setShuffle, setStepMode, setTestData, setTrainingData, setTrainingSignalGenerator, test, test, toString, toXML, train, trainPattern, updateLinks, updateNeurons

Methods inherited from class java.lang.Object

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

Constructor Detail

HebbTrainer

public HebbTrainer()

Method Detail

train

public void train(java.util.List<java.lang.Double> input,
                  java.util.List<java.lang.Double> target)

Hebbian learning of a single pattern. The implementation does this:

• apply input to net
• go through all neurons in forward tick list order and apply Hebb learning

Specified by:

train in class Trainer

Parameters:

input - an input pattern

target - not used

reset

protected void reset()

Does nothing.

Specified by:

reset in class Trainer

endBatch

protected void endBatch()

Does nothing.

Overrides:

endBatch in class Trainer

calcAdaptation

protected double calcAdaptation(BrainPart part)

Not used.

Specified by:

calcAdaptation in class Trainer

Parameters:

part - the brain part to be adapted

Returns:

the adaptation value

endTrain

protected void endTrain()

Does nothing.

Specified by:

endTrain in class Trainer

fromXML

public void fromXML(org.jdom2.Element element)

Description copied from class: Trainer

Reads attributes from XML.

Specified by:

fromXML in interface Storable

Overrides:

fromXML in class Trainer

Parameters:

element - a JDOM element