keel.Algorithms.Shared.ClassicalOptim

Class SquaresErrorNN

java.lang.Object

keel.Algorithms.Shared.ClassicalOptim.FUN

keel.Algorithms.Shared.ClassicalOptim.SquaresErrorNN

public class SquaresErrorNN
extends FUN

Derived class from FUN that implements the error for a perceptron trained with conjugated gradient.

Since:

JDK1.4

Version:

1.0

Author:

Written by Luciano Sánchez (University of Oviedo) 27/02/2004, Modified by Enrique A. de la Cal (University of Oviedo) 13/12/2008

Field Summary

Fields

Modifier and Type	Field and Description
ConjGradNN	Net Neural network container.

Constructor Summary

Constructors

Constructor and Description
SquaresErrorNN(ConjGradNN vNet) Constructor of an error calculator for neural network based on the conjugated gradient.

Method Summary

Modifier and Type	Method and Description
double	evaluate(double[][][] x) Returns the training mean square error for a perceptron with weights x

Methods inherited from class java.lang.Object

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

Field Detail

Net

public ConjGradNN Net

Neural network container.

Constructor Detail

SquaresErrorNN

public SquaresErrorNN(ConjGradNN vNet)

Constructor of an error calculator for neural network based on the conjugated gradient.

Parameters:

vNet - container with a perceptron.

Method Detail

evaluate

public double evaluate(double[][][] x)

Returns the training mean square error for a perceptron with weights x

Specified by:

evaluate in class FUN

Parameters:

x - the weights of a perceptron.

Returns:

the training mean square error of a perceptron with weights x.