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
Method Summary |
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, isPausedLearning, learn, learn, onStart, pause, resume, setLearningRate, setMaxIterations |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
UnsupervisedLearning
public UnsupervisedLearning()
- Creates new unsupervised learning rule
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
Copyright © 2012. All Rights Reserved.