comparison toolboxes/FullBNT-1.0.7/nethelp3.3/rbfgrad.htm @ 0:e9a9cd732c1e tip

first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:e9a9cd732c1e
1 <html>
2 <head>
3 <title>
4 Netlab Reference Manual rbfgrad
5 </title>
6 </head>
7 <body>
8 <H1> rbfgrad
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Evaluate gradient of error function for RBF network.
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19
20 g = rbfgrad(net, x, t)
21 [g, gdata, gprior] = rbfgrad(net, x, t)
22 </PRE>
23
24
25 <p><h2>
26 Description
27 </h2>
28 <CODE>g = rbfgrad(net, x, t)</CODE> takes a network data structure <CODE>net</CODE>
29 together with a matrix <CODE>x</CODE> of input
30 vectors and a matrix <CODE>t</CODE> of target vectors, and evaluates the gradient
31 <CODE>g</CODE> of the error function with respect to the network weights (i.e.
32 including the hidden unit parameters). The error
33 function is sum of squares.
34 Each row of <CODE>x</CODE> corresponds to one
35 input vector and each row of <CODE>t</CODE> contains the corresponding target vector.
36 If the output function is <CODE>'neuroscale'</CODE> then the gradient is only
37 computed for the output layer weights and biases.
38
39 <p><CODE>[g, gdata, gprior] = rbfgrad(net, x, t)</CODE> also returns separately
40 the data and prior contributions to the gradient. In the case of
41 multiple groups in the prior, <CODE>gprior</CODE> is a matrix with a row
42 for each group and a column for each weight parameter.
43
44 <p><h2>
45 See Also
46 </h2>
47 <CODE><a href="rbf.htm">rbf</a></CODE>, <CODE><a href="rbffwd.htm">rbffwd</a></CODE>, <CODE><a href="rbferr.htm">rbferr</a></CODE>, <CODE><a href="rbfpak.htm">rbfpak</a></CODE>, <CODE><a href="rbfunpak.htm">rbfunpak</a></CODE>, <CODE><a href="rbfbkp.htm">rbfbkp</a></CODE><hr>
48 <b>Pages:</b>
49 <a href="index.htm">Index</a>
50 <hr>
51 <p>Copyright (c) Ian T Nabney (1996-9)
52
53
54 </body>
55 </html>