ttr3.m

来自「书籍“Regularization tools for training lar」· M 代码 · 共 21 行

function [w1,b1,w2,b2,w3,b3,tr,rq] = ttr3(w1,b1,f1,w2,b2,f2,w3,b3,f3,...
					xc,P,T,VA,VAT,TE,TET,TP)

%TTR3 Trains a feed-forward network with one hidden layer
%using the Gauss-Newton method on a Tikhonov regularized problem.
%
%The general design of TTR3 is much the same as the design of the 
%functions in the Neural Network (NN) Toolbox from MathWorks even though the
%calling sequences differ. See Neural Network Toolbox User's Guide from the
%MathWorks Inc.
%
%TTR3 is intended for problems so small that explicit storing of
%the corresponding Jacobian matrix is possible.
%For lager problems where the Jacobian can't be stored in main memory, there
%is a function LTTR3 that uses a truncated Gauss-Newton method
%on a Tikhonov regularized problem. However, even for smaller problems LTTR3
%is often to be prefered to TTR3 (not to mention the Levenberg-Marquardt
%implementations in the NN Toolbox) since it is normally much faster.
%
%You can find more details in two papers by Eriksson J., Gulliksson M.,
%Lindstr鰉 P. and Wedin P-