comparison toolboxes/FullBNT-1.0.7/nethelp3.3/gmminit.htm @ 0:cc4b1211e677 tip

initial commit to HG from Changeset: 646 (e263d8a21543) added further path and more save "camirversion.m"
author Daniel Wolff
date Fri, 19 Aug 2016 13:07:06 +0200
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:cc4b1211e677
1 <html>
2 <head>
3 <title>
4 Netlab Reference Manual gmminit
5 </title>
6 </head>
7 <body>
8 <H1> gmminit
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Initialises Gaussian mixture model from data
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19
20 mix = gmminit(mix, x, options)
21 </PRE>
22
23
24 <p><h2>
25 Description
26 </h2>
27 <CODE>mix = gmminit(mix, x, options)</CODE> uses a dataset <CODE>x</CODE>
28 to initialise the parameters of a Gaussian mixture
29 model defined by the data structure <CODE>mix</CODE>. The k-means algorithm
30 is used to determine the centres. The priors are computed from the
31 proportion of examples belonging to each cluster.
32 The covariance matrices are calculated as the sample covariance of the
33 points associated with (i.e. closest to) the corresponding centres.
34 For a mixture of PPCA model, the PPCA decomposition is calculated
35 for the points closest to a given centre.
36 This initialisation can be used as the starting point for training the
37 model using the EM algorithm.
38
39 <p><h2>
40 Example
41 </h2>
42 <PRE>
43
44 mix = gmm(3, 2);
45 options = foptions;
46 options(14) = 5;
47 mix = gmminit(mix, data, options);
48 </PRE>
49
50 This code sets up a Gaussian mixture model with 3 centres in 2 dimensions, and
51 then initialises the parameters from the data set <CODE>data</CODE> with 5 iterations
52 of the k means algorithm.
53
54 <p><h2>
55 See Also
56 </h2>
57 <CODE><a href="gmm.htm">gmm</a></CODE><hr>
58 <b>Pages:</b>
59 <a href="index.htm">Index</a>
60 <hr>
61 <p>Copyright (c) Ian T Nabney (1996-9)
62
63
64 </body>
65 </html>