org.neuroph.core.learning

Class UnsupervisedLearning

java.lang.Object

org.neuroph.core.learning.LearningRule

org.neuroph.core.learning.IterativeLearning

org.neuroph.core.learning.UnsupervisedLearning

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

CompetitiveLearning, UnsupervisedHebbianLearning

public abstract class UnsupervisedLearning
extends IterativeLearning
implements Serializable

Base class for all unsupervised learning algorithms.

Author:

Zoran Sevarac

See Also:

Serialized Form

Field Summary

Fields inherited from class org.neuroph.core.learning.IterativeLearning

currentIteration, learningRate, stopConditions

Fields inherited from class org.neuroph.core.learning.LearningRule

listeners, neuralNetwork, trainingSet

Constructor Summary

Constructors

Constructor and Description
UnsupervisedLearning() Creates new unsupervised learning rule

Method Summary

Modifier and Type	Method and Description
void	doLearningEpoch(DataSet trainingSet) This method does one learning epoch for the unsupervised learning rules.
protected void	learnPattern(DataSetRow trainingElement) Trains network with the pattern from the specified training element
protected abstract void	updateNetworkWeights() This method implements the weight adjustment

Methods inherited from class org.neuroph.core.learning.IterativeLearning

afterEpoch, beforeEpoch, doOneLearningIteration, getCurrentIteration, getLearningRate, getMaxIterations, hasReachedStopCondition, isIterationsLimited, isPausedLearning, learn, learn, onStart, pause, resume, setLearningRate, setMaxIterations

Methods inherited from class org.neuroph.core.learning.LearningRule

addListener, fireLearningEvent, getNeuralNetwork, getTrainingSet, isStopped, onStop, removeListener, setNeuralNetwork, setTrainingSet, stopLearning

Methods inherited from class java.lang.Object

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

Constructor Detail

UnsupervisedLearning

public UnsupervisedLearning()

Creates new unsupervised learning rule

Method Detail

doLearningEpoch

public void doLearningEpoch(DataSet trainingSet)

This method does one learning epoch for the unsupervised learning rules. It iterates through the training set and trains network weights for each element

Specified by:

doLearningEpoch in class IterativeLearning

Parameters:

trainingSet - training set for training network

learnPattern

protected void learnPattern(DataSetRow trainingElement)

Trains network with the pattern from the specified training element

Parameters:

DataSetItem - unsupervised training element which contains network input

updateNetworkWeights

protected abstract void updateNetworkWeights()

This method implements the weight adjustment