error.MSE

error.LMLS

error.TAO

deltaE.MSE

deltaE.LMLS

deltaE.TAO

The error functions calculate the goodness of fit of a neural network according to certain criterium:
<item>MSE: Mean Squared Error. Least Mean Squares minimization.
<item>LMLS: Least Mean Log Squares minimization. 
<item>TAO: TAO error minimization.
The deltaE functions calculate the corresponding influence functions.</item><usage>error.MSE(arguments)
error.LMLS(arguments)
error.TAO(arguments)
deltaE.MSE(arguments)
deltaE.LMLS(arguments)
deltaE.TAO(arguments)</usage>
<arguments><item>arguments</item>{List of arguments to pass to the functions.
 <item>The first element is the prediction of the neural network.
 <item>The second element is the target value.
 <item>A third element is needed for the TAO method containing the value of the S parameter.</item></item></item></arguments><value>This functions return the error and influence function criteria.</value>
[object Object],[object Object],[object Object]<references>Pernia Espinoza, A.V. TAO-robust backpropagation learning algorithm. Neural Networks. In press. Simon Haykin. Neural Networks. A comprehensive foundation. 2nd Edition.</references><seealso><code><a href="/link/train?package=AMORE&version=0.1.1" rd-options="" data-mini-rdoc="AMORE::train">train</a></code>, <code><a href="/link/train.compare?package=AMORE&version=0.1.1" rd-options="" data-mini-rdoc="AMORE::train.compare">train.compare</a></code></seealso><keyword>neural</keyword></item></item>

This package was born to release the TAO robust neural network algorithm to the R users. It has grown and I think it can be of interest for the users wanting to implement their own training algorithms as well as for those others whose needs lye only in the "user space".

error.TAO: Neural network training error criteria.

Description

Arguments

encoding