comparison toolboxes/FullBNT-1.0.7/nethelp3.3/nethess.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 nethess
5 </title>
6 </head>
7 <body>
8 <H1> nethess
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Evaluate network Hessian
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19 h = neterr(w, net, x, t)
20 [h, varargout] = neterr(w, net, x, t, varargin)
21 </PRE>
22
23
24 <p><h2>
25 Description
26 </h2>
27
28 <p><CODE>h = nethess(w, net, x, t)</CODE> takes a weight vector <CODE>w</CODE> and a network
29 data structure <CODE>net</CODE>, together with the matrix <CODE>x</CODE> of input
30 vectors and the matrix <CODE>t</CODE> of target vectors, and returns the
31 value of the Hessian evaluated at <CODE>w</CODE>.
32
33 <p><CODE>[e, varargout] = nethess(w, net, x, t, varargin)</CODE> also returns any additional
34 return values from the network Hessian function, and passes additional arguments
35 to that function.
36
37 <p><h2>
38 Example
39 </h2>
40
41 <p>In <CODE>evidence</CODE>, this function is called once to compute the
42 data contribution to the Hessian
43 <PRE>
44
45 [h, dh] = nethess(w, net, x, t, dh);
46 </PRE>
47
48 and again to update the Hessian for new values of the hyper-parameters
49 <PRE>
50
51 h = nethess(w, net, x, t, dh);
52 </PRE>
53
54
55 <p><h2>
56 See Also
57 </h2>
58 <CODE><a href="neterr.htm">neterr</a></CODE>, <CODE><a href="netgrad.htm">netgrad</a></CODE>, <CODE><a href="netopt.htm">netopt</a></CODE><hr>
59 <b>Pages:</b>
60 <a href="index.htm">Index</a>
61 <hr>
62 <p>Copyright (c) Ian T Nabney (1996-9)
63
64
65 </body>
66 </html>