Mercurial > hg > camir-aes2014
annotate toolboxes/FullBNT-1.0.7/netlab3.3/netgrad.m @ 0:e9a9cd732c1e tip
first hg version after svn
author | wolffd |
---|---|
date | Tue, 10 Feb 2015 15:05:51 +0000 |
parents | |
children |
rev | line source |
---|---|
wolffd@0 | 1 function g = netgrad(w, net, x, t) |
wolffd@0 | 2 %NETGRAD Evaluate network error gradient for generic optimizers |
wolffd@0 | 3 % |
wolffd@0 | 4 % Description |
wolffd@0 | 5 % |
wolffd@0 | 6 % G = NETGRAD(W, NET, X, T) takes a weight vector W and a network data |
wolffd@0 | 7 % structure NET, together with the matrix X of input vectors and the |
wolffd@0 | 8 % matrix T of target vectors, and returns the gradient of the error |
wolffd@0 | 9 % function evaluated at W. |
wolffd@0 | 10 % |
wolffd@0 | 11 % See also |
wolffd@0 | 12 % MLP, NETERR, NETOPT |
wolffd@0 | 13 % |
wolffd@0 | 14 |
wolffd@0 | 15 % Copyright (c) Ian T Nabney (1996-2001) |
wolffd@0 | 16 |
wolffd@0 | 17 gradstr = [net.type, 'grad']; |
wolffd@0 | 18 |
wolffd@0 | 19 net = netunpak(net, w); |
wolffd@0 | 20 |
wolffd@0 | 21 g = feval(gradstr, net, x, t); |