comparison toolboxes/FullBNT-1.0.7/nethelp3.3/gpcovar.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 gpcovar
5 </title>
6 </head>
7 <body>
8 <H1> gpcovar
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Calculate the covariance for a Gaussian Process.
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19 cov = gpcovar(net, x)
20 [cov, covf] = gpcovar(net, x)
21 </PRE>
22
23
24 <p><h2>
25 Description
26 </h2>
27
28 <p><CODE>cov = gpcovar(net, x)</CODE> takes
29 a Gaussian Process data structure <CODE>net</CODE> together with
30 a matrix <CODE>x</CODE> of input vectors, and computes the covariance
31 matrix <CODE>cov</CODE>. The inverse of this matrix is used when calculating
32 the mean and variance of the predictions made by <CODE>net</CODE>.
33
34 <p><CODE>[cov, covf] = gpcovar(net, x)</CODE> also generates the covariance
35 matrix due to the covariance function specified by <CODE>net.covarfn</CODE>
36 as calculated by <CODE>gpcovarf</CODE>.
37
38 <p><h2>
39 Example
40 </h2>
41 In the following example, the inverse covariance matrix is calculated
42 for a set of training inputs <CODE>x</CODE> and is then
43 passed to <CODE>gpfwd</CODE> so that predictions (with mean <CODE>ytest</CODE> and
44 variance <CODE>sigsq</CODE>) can be made for the test inputs
45 <CODE>xtest</CODE>.
46 <PRE>
47
48 cninv = inv(gpcovar(net, x));
49 [ytest, sigsq] = gpfwd(net, xtest, cninv);
50 </PRE>
51
52
53 <p><h2>
54 See Also
55 </h2>
56 <CODE><a href="gp.htm">gp</a></CODE>, <CODE><a href="gppak.htm">gppak</a></CODE>, <CODE><a href="gpunpak.htm">gpunpak</a></CODE>, <CODE><a href="gpcovarp.htm">gpcovarp</a></CODE>, <CODE><a href="gpcovarf.htm">gpcovarf</a></CODE>, <CODE><a href="gpfwd.htm">gpfwd</a></CODE>, <CODE><a href="gperr.htm">gperr</a></CODE>, <CODE><a href="gpgrad.htm">gpgrad</a></CODE><hr>
57 <b>Pages:</b>
58 <a href="index.htm">Index</a>
59 <hr>
60 <p>Copyright (c) Ian T Nabney (1996-9)
61
62
63 </body>
64 </html>