boone.training

Class LVQTrainer

java.lang.Object

boone.Trainer

boone.training.SOMTrainer

boone.training.LVQTrainer

All Implemented Interfaces:

Storable, java.lang.Cloneable

public class LVQTrainer
extends SOMTrainer

The Learning Vector Quantization training.

Author:

ghowa, Helmut A. Mayer

Field Summary

Fields

Modifier and Type	Field and Description
static int	LVQ1 Simple LVQ with constant learn rate.
static int	OLVQ1 Optimized LVQ with neuron-specific adapting learn rate (default).

Fields inherited from class boone.training.SOMTrainer

deltaRadius, radius, startRadius, topology

Fields inherited from class boone.Trainer

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

Constructor Summary

Constructors

Constructor and Description
LVQTrainer(Topology topology) Creates the LVQ Trainer.

Method Summary

Modifier and Type	Method and Description
LVQTrainer	clone() A deep clone.
void	fromXML(org.jdom2.Element element) Reads attributes from XML.
protected void	reset() Resets the trainer by setting the output neuron's error signal to 'learnRate'.
void	setFlavor(int flavor)
org.jdom2.Element	toXML(org.jdom2.Element element) Writes specific attributes to XML tree.
void	train(PatternSet set, int epochs) Trains the SOM depending on the 'supervised' flag.

Methods inherited from class boone.training.SOMTrainer

calcAdaptation, endBatch, endTrain, getStartRadius, getTopology, setNetwork, setStartRadius, test, train, train

Methods inherited from class boone.Trainer

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

Methods inherited from class java.lang.Object

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

Field Detail

LVQ1

public static final int LVQ1

Simple LVQ with constant learn rate.

See Also:

Constant Field Values

OLVQ1

public static final int OLVQ1

Optimized LVQ with neuron-specific adapting learn rate (default).

See Also:

Constant Field Values

Constructor Detail

LVQTrainer

public LVQTrainer(Topology topology)

Creates the LVQ Trainer. At this time the associated net is not known. Could be improved, but maybe too complicated.

Parameters:

topology - the SOM topology

Method Detail

clone

public LVQTrainer clone()

A deep clone.

Overrides:

clone in class SOMTrainer

reset

protected void reset()

Resets the trainer by setting the output neuron's error signal to 'learnRate'. The error signal is used here as a neuron specific learn rate.

Overrides:

reset in class SOMTrainer

setFlavor

public void setFlavor(int flavor)

train

public void train(PatternSet set,
                  int epochs)

Trains the SOM depending on the 'supervised' flag. If not supervised, a basic SOM training is conducted. If supervised, the basic OLVQ1 algorithm is used assigning specific neurons to specific clusters.

Parameters:

set - the complete training set

epochs - the number of epochs

toXML

public org.jdom2.Element toXML(org.jdom2.Element element)

Writes specific attributes to XML tree.

Specified by:

toXML in interface Storable

Overrides:

toXML in class SOMTrainer

Parameters:

element - the parent element

Returns:

the trainer element

fromXML

public void fromXML(org.jdom2.Element element)

Description copied from class: SOMTrainer

Reads attributes from XML.

Specified by:

fromXML in interface Storable

Overrides:

fromXML in class SOMTrainer

Parameters:

element - a JDOM element