Mercurial > hg > camir-aes2014
annotate toolboxes/FullBNT-1.0.7/netlab3.3/mlpevfwd.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 [y, extra, invhess] = mlpevfwd(net, x, t, x_test, invhess) |
| wolffd@0 | 2 %MLPEVFWD Forward propagation with evidence for MLP |
| wolffd@0 | 3 % |
| wolffd@0 | 4 % Description |
| wolffd@0 | 5 % Y = MLPEVFWD(NET, X, T, X_TEST) takes a network data structure NET |
| wolffd@0 | 6 % together with the input X and target T training data and input test |
| wolffd@0 | 7 % data X_TEST. It returns the normal forward propagation through the |
| wolffd@0 | 8 % network Y together with a matrix EXTRA which consists of error bars |
| wolffd@0 | 9 % (variance) for a regression problem or moderated outputs for a |
| wolffd@0 | 10 % classification problem. The optional argument (and return value) |
| wolffd@0 | 11 % INVHESS is the inverse of the network Hessian computed on the |
| wolffd@0 | 12 % training data inputs and targets. Passing it in avoids recomputing |
| wolffd@0 | 13 % it, which can be a significant saving for large training sets. |
| wolffd@0 | 14 % |
| wolffd@0 | 15 % See also |
| wolffd@0 | 16 % FEVBAYES |
| wolffd@0 | 17 % |
| wolffd@0 | 18 |
| wolffd@0 | 19 % Copyright (c) Ian T Nabney (1996-2001) |
| wolffd@0 | 20 |
| wolffd@0 | 21 [y, z, a] = mlpfwd(net, x_test); |
| wolffd@0 | 22 if nargin == 4 |
| wolffd@0 | 23 [extra, invhess] = fevbayes(net, y, a, x, t, x_test); |
| wolffd@0 | 24 else |
| wolffd@0 | 25 [extra, invhess] = fevbayes(net, y, a, x, t, x_test, invhess); |
| wolffd@0 | 26 end |