Mercurial > hg > camir-ismir2012
annotate toolboxes/FullBNT-1.0.7/netlab3.3/netderiv.m @ 0:cc4b1211e677 tip
initial commit to HG from
Changeset:
646 (e263d8a21543) added further path and more save "camirversion.m"
| author | Daniel Wolff |
|---|---|
| date | Fri, 19 Aug 2016 13:07:06 +0200 |
| parents | |
| children |
| rev | line source |
|---|---|
| Daniel@0 | 1 function g = netderiv(w, net, x) |
| Daniel@0 | 2 %NETDERIV Evaluate derivatives of network outputs by weights generically. |
| Daniel@0 | 3 % |
| Daniel@0 | 4 % Description |
| Daniel@0 | 5 % |
| Daniel@0 | 6 % G = NETDERIV(W, NET, X) takes a weight vector W and a network data |
| Daniel@0 | 7 % structure NET, together with the matrix X of input vectors, and |
| Daniel@0 | 8 % returns the gradient of the outputs with respect to the weights |
| Daniel@0 | 9 % evaluated at W. |
| Daniel@0 | 10 % |
| Daniel@0 | 11 % See also |
| Daniel@0 | 12 % NETEVFWD, NETOPT |
| Daniel@0 | 13 % |
| Daniel@0 | 14 |
| Daniel@0 | 15 % Copyright (c) Ian T Nabney (1996-2001) |
| Daniel@0 | 16 |
| Daniel@0 | 17 fstr = [net.type, 'deriv']; |
| Daniel@0 | 18 net = netunpak(net, w); |
| Daniel@0 | 19 g = feval(fstr, net, x); |