boone.training

Class TrainingSignalGenerator

java.lang.Object

boone.training.TrainingSignalGenerator

All Implemented Interfaces:

Storable, java.lang.Cloneable

Direct Known Subclasses:

CrossEntropy, SquareError

public abstract class TrainingSignalGenerator
extends java.lang.Object
implements java.lang.Cloneable, Storable

This generator computes error signals for back-propagation and network errors.

Since:

12. Juni 2003, 23:14

Version:

$Id$

Author:

August Mayer, Helmut A. Mayer

Constructor Summary

Constructors

Constructor and Description
TrainingSignalGenerator()

Method Summary

Modifier and Type	Method and Description
TrainingSignalGenerator	clone() Return a deep copy.
abstract double	computeError(NeuralNet net, java.util.List<java.lang.Double> input, java.util.List<java.lang.Double> target) Computes the error at the output layer for the given input.
abstract void	computeErrorSignal(NeuralNet net, java.util.List<java.lang.Double> input, java.util.List<java.lang.Double> target) Computes the error signal at the output layer for back-propagation.
void	fromXML(org.jdom2.Element element) Reads attributes from XML tree.
org.jdom2.Element	toXML(org.jdom2.Element element) Writes attributes to XML tree.

Methods inherited from class java.lang.Object

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

Constructor Detail

TrainingSignalGenerator

public TrainingSignalGenerator()

Method Detail

computeErrorSignal

public abstract void computeErrorSignal(NeuralNet net,
                                        java.util.List<java.lang.Double> input,
                                        java.util.List<java.lang.Double> target)

Computes the error signal at the output layer for back-propagation.

Parameters:

net - the neural network

input - the input pattern

target - the target pattern, may be null, if this is e.g. unsupervised learning.

computeError

public abstract double computeError(NeuralNet net,
                                    java.util.List<java.lang.Double> input,
                                    java.util.List<java.lang.Double> target)

Computes the error at the output layer for the given input.

Parameters:

net - the neural network

input - the input pattern

target - the target pattern

Returns:

the error

clone

public TrainingSignalGenerator clone()

Return a deep copy.

Overrides:

clone in class java.lang.Object

toXML

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

Writes attributes to XML tree.

Specified by:

toXML in interface Storable

Parameters:

element - the parent element

Returns:

the generator element

fromXML

public void fromXML(org.jdom2.Element element)

Description copied from interface: Storable

Reads attributes from XML tree.

Specified by:

fromXML in interface Storable

Parameters:

element - a JDOM element