Mercurial > hg > camir-aes2014

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

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
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 kmeans
5 </title>
6 </head>
7 <body>
8 <H1> kmeans
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Trains a k means cluster model.
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19 centres = kmeans(centres, data, options)
20 [centres, options] = kmeans(centres, data, options)
21 [centres, options, post, errlog] = kmeans(centres, data, options)
22 </PRE>
23
24
25 <p><h2>
26 Description
27 </h2>
28
29 <CODE>centres = kmeans(centres, data, options)</CODE>
30 uses the batch K-means algorithm to set the centres of a cluster model.
31 The matrix <CODE>data</CODE> represents the data
32 which is being clustered, with each row corresponding to a vector.
33 The sum of squares error function is used. The point at which
34 a local minimum is achieved is returned as <CODE>centres</CODE>. The
35 error value at that point is returned in <CODE>options(8)</CODE>.
36
37 <p><CODE>[centres, options, post, errlog] = kmeans(centres, data, options)</CODE>
38 also returns the cluster number (in a one-of-N encoding) for each data
39 point in <CODE>post</CODE> and a log of the error values after each cycle in
40 <CODE>errlog</CODE>.
41
42 The optional parameters have the following interpretations.
43
44 <p><CODE>options(1)</CODE> is set to 1 to display error values; also logs error
45 values in the return argument <CODE>errlog</CODE>.
46 If <CODE>options(1)</CODE> is set to 0,
47 then only warning messages are displayed. If <CODE>options(1)</CODE> is -1,
48 then nothing is displayed.
49
50 <p><CODE>options(2)</CODE> is a measure of the absolute precision required for the value
51 of <CODE>centres</CODE> at the solution. If the absolute difference between
52 the values of <CODE>centres</CODE> between two successive steps is less than
53 <CODE>options(2)</CODE>, then this condition is satisfied.
54
55 <p><CODE>options(3)</CODE> is a measure of the precision required of the error
56 function at the solution. If the absolute difference between the
57 error functions between two successive steps is less than
58 <CODE>options(3)</CODE>, then this condition is satisfied.
59 Both this and the previous condition must be
60 satisfied for termination.
61
62 <p><CODE>options(14)</CODE> is the maximum number of iterations; default 100.
63
64 <p><h2>
65 Example
66 </h2>
67 <CODE>kmeans</CODE> can be used to initialise the centres of a Gaussian
68 mixture model that is then trained with the EM algorithm.
69 <PRE>
70
71 [priors, centres, var] = gmmunpak(p, md);
72 centres = kmeans(centres, data, options);
73 p = gmmpak(priors, centres, var);
74 p = gmmem(p, md, data, options);
75 </PRE>
76
77
78 <p><h2>
79 See Also
80 </h2>
81 <CODE><a href="gmminit.htm">gmminit</a></CODE>, <CODE><a href="gmmem.htm">gmmem</a></CODE><hr>
82 <b>Pages:</b>
83 <a href="index.htm">Index</a>
84 <hr>
85 <p>Copyright (c) Ian T Nabney (1996-9)
86
87
88 </body>
89 </html>