BNW using Octave instead of Matlab.

This version of BNW should perform the same as the original version. The only difference is that it uses Octave instead of Matlab when running BayesNet Toolbox during parameter learning.

I am calling this BNW_1.02. It can be accessed at:
compbio.uthsc.edu/BNW_1.02

1 files changed, 29 insertions, 0 deletions

diff --git a/sourcecodes/bnt-master/netlab3.3/neterr.m b/sourcecodes/bnt-master/netlab3.3/neterr.m
new file mode 100644
index 00000000..6679630c
--- /dev/null
+++ b/sourcecodes/bnt-master/netlab3.3/neterr.m
@@ -0,0 +1,29 @@
+function [e, varargout] = neterr(w, net, x, t)
+%NETERR	Evaluate network error function for generic optimizers
+%
+%	Description
+%
+%	E = NETERR(W, NET, X, T) takes a weight vector W and a network data
+%	structure NET, together with the matrix X of input vectors and the
+%	matrix T of target vectors, and returns the value of the error
+%	function evaluated at W.
+%
+%	[E, VARARGOUT] = NETERR(W, NET, X, T) also returns any additional
+%	return values from the error function.
+%
+%	See also
+%	NETGRAD, NETHESS, NETOPT
+%
+
+%	Copyright (c) Ian T Nabney (1996-2001)
+
+errstr = [net.type, 'err'];
+net = netunpak(net, w);
+
+[s{1:nargout}] = feval(errstr, net, x, t);
+e = s{1};
+if nargout > 1
+  for i = 2:nargout
+    varargout{i-1} = s{i};
+  end
+end