|
wolffd@0
|
1 <html>
|
|
wolffd@0
|
2 <head>
|
|
wolffd@0
|
3 <title>
|
|
wolffd@0
|
4 Netlab Reference Manual gtmem
|
|
wolffd@0
|
5 </title>
|
|
wolffd@0
|
6 </head>
|
|
wolffd@0
|
7 <body>
|
|
wolffd@0
|
8 <H1> gtmem
|
|
wolffd@0
|
9 </H1>
|
|
wolffd@0
|
10 <h2>
|
|
wolffd@0
|
11 Purpose
|
|
wolffd@0
|
12 </h2>
|
|
wolffd@0
|
13 EM algorithm for Generative Topographic Mapping.
|
|
wolffd@0
|
14
|
|
wolffd@0
|
15 <p><h2>
|
|
wolffd@0
|
16 Synopsis
|
|
wolffd@0
|
17 </h2>
|
|
wolffd@0
|
18 <PRE>
|
|
wolffd@0
|
19
|
|
wolffd@0
|
20 [net, options, errlog] = gtmem(net, t, options)
|
|
wolffd@0
|
21 </PRE>
|
|
wolffd@0
|
22
|
|
wolffd@0
|
23
|
|
wolffd@0
|
24 <p><h2>
|
|
wolffd@0
|
25 Description
|
|
wolffd@0
|
26 </h2>
|
|
wolffd@0
|
27 <CODE>[net, options, errlog] = gtmem(net, t, options)</CODE> uses the Expectation
|
|
wolffd@0
|
28 Maximization algorithm to estimate the parameters of
|
|
wolffd@0
|
29 a GTM defined by a data structure <CODE>net</CODE>.
|
|
wolffd@0
|
30 The matrix <CODE>t</CODE> represents the data whose expectation
|
|
wolffd@0
|
31 is maximized, with each row corresponding to a vector. It is assumed
|
|
wolffd@0
|
32 that the latent data <CODE>net.X</CODE> has been set following a call to
|
|
wolffd@0
|
33 <CODE>gtminit</CODE>, for example.
|
|
wolffd@0
|
34
|
|
wolffd@0
|
35 The optional parameters have the following interpretations.
|
|
wolffd@0
|
36
|
|
wolffd@0
|
37 <p><CODE>options(1)</CODE> is set to 1 to display error values; also logs error
|
|
wolffd@0
|
38 values in the return argument <CODE>errlog</CODE>.
|
|
wolffd@0
|
39 If <CODE>options(1)</CODE> is set to 0,
|
|
wolffd@0
|
40 then only warning messages are displayed. If <CODE>options(1)</CODE> is -1,
|
|
wolffd@0
|
41 then nothing is displayed.
|
|
wolffd@0
|
42
|
|
wolffd@0
|
43 <p><CODE>options(3)</CODE> is a measure of the absolute precision required of the error
|
|
wolffd@0
|
44 function at the solution. If the change in log likelihood between two steps of
|
|
wolffd@0
|
45 the EM algorithm is less than this value, then the function terminates.
|
|
wolffd@0
|
46
|
|
wolffd@0
|
47 <p><CODE>options(14)</CODE> is the maximum number of iterations; default 100.
|
|
wolffd@0
|
48
|
|
wolffd@0
|
49 <p>The optional return value <CODE>options</CODE> contains the final error value
|
|
wolffd@0
|
50 (i.e. data log likelihood) in
|
|
wolffd@0
|
51 <CODE>options(8)</CODE>.
|
|
wolffd@0
|
52
|
|
wolffd@0
|
53 <p><h2>
|
|
wolffd@0
|
54 Examples
|
|
wolffd@0
|
55 </h2>
|
|
wolffd@0
|
56 The following code fragment sets up a GTM, initialises
|
|
wolffd@0
|
57 the latent data sample and RBF
|
|
wolffd@0
|
58 parameters from the data, sets the options and trains the model.
|
|
wolffd@0
|
59 <PRE>
|
|
wolffd@0
|
60
|
|
wolffd@0
|
61 % Create and initialise GTM model
|
|
wolffd@0
|
62 net = gtm(latentdim, nlatent, datadim, numrbfcentres, ...
|
|
wolffd@0
|
63 'gaussian', 0.1);
|
|
wolffd@0
|
64
|
|
wolffd@0
|
65 <p>options = foptions;
|
|
wolffd@0
|
66 options(1) = -1;
|
|
wolffd@0
|
67 options(7) = 1; % Set width factor of RBF
|
|
wolffd@0
|
68 net = gtminit(net, options, data, 'regular', latentshape, [4 4]);
|
|
wolffd@0
|
69
|
|
wolffd@0
|
70 <p>options = foptions;
|
|
wolffd@0
|
71 options(14) = 30;
|
|
wolffd@0
|
72 options(1) = 1;
|
|
wolffd@0
|
73 [net, options] = gtmem(net, data, options);
|
|
wolffd@0
|
74 </PRE>
|
|
wolffd@0
|
75
|
|
wolffd@0
|
76
|
|
wolffd@0
|
77 <p><h2>
|
|
wolffd@0
|
78 See Also
|
|
wolffd@0
|
79 </h2>
|
|
wolffd@0
|
80 <CODE><a href="gtm.htm">gtm</a></CODE>, <CODE><a href="gtminit.htm">gtminit</a></CODE><hr>
|
|
wolffd@0
|
81 <b>Pages:</b>
|
|
wolffd@0
|
82 <a href="index.htm">Index</a>
|
|
wolffd@0
|
83 <hr>
|
|
wolffd@0
|
84 <p>Copyright (c) Ian T Nabney (1996-9)
|
|
wolffd@0
|
85
|
|
wolffd@0
|
86
|
|
wolffd@0
|
87 </body>
|
|
wolffd@0
|
88 </html> |
