|
wolffd@0
|
1 <html>
|
|
wolffd@0
|
2 <head>
|
|
wolffd@0
|
3 <title>
|
|
wolffd@0
|
4 Netlab Reference Manual gmmem
|
|
wolffd@0
|
5 </title>
|
|
wolffd@0
|
6 </head>
|
|
wolffd@0
|
7 <body>
|
|
wolffd@0
|
8 <H1> gmmem
|
|
wolffd@0
|
9 </H1>
|
|
wolffd@0
|
10 <h2>
|
|
wolffd@0
|
11 Purpose
|
|
wolffd@0
|
12 </h2>
|
|
wolffd@0
|
13 EM algorithm for Gaussian mixture model.
|
|
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 [mix, options, errlog] = gmmem(mix, x, 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>[mix, options, errlog] = gmmem(mix, x, options)</CODE> uses the Expectation
|
|
wolffd@0
|
28 Maximization algorithm of Dempster et al. to estimate the parameters of
|
|
wolffd@0
|
29 a Gaussian mixture model defined by a data structure <CODE>mix</CODE>.
|
|
wolffd@0
|
30 The matrix <CODE>x</CODE> represents the data whose expectation
|
|
wolffd@0
|
31 is maximized, with each row corresponding to a vector.
|
|
wolffd@0
|
32
|
|
wolffd@0
|
33 The optional parameters have the following interpretations.
|
|
wolffd@0
|
34
|
|
wolffd@0
|
35 <p><CODE>options(1)</CODE> is set to 1 to display error values; also logs error
|
|
wolffd@0
|
36 values in the return argument <CODE>errlog</CODE>.
|
|
wolffd@0
|
37 If <CODE>options(1)</CODE> is set to 0,
|
|
wolffd@0
|
38 then only warning messages are displayed. If <CODE>options(1)</CODE> is -1,
|
|
wolffd@0
|
39 then nothing is displayed.
|
|
wolffd@0
|
40
|
|
wolffd@0
|
41 <p><CODE>options(3)</CODE> is a measure of the absolute precision required of the error
|
|
wolffd@0
|
42 function at the solution. If the change in log likelihood between two steps of
|
|
wolffd@0
|
43 the EM algorithm is less than this value, then the function terminates.
|
|
wolffd@0
|
44
|
|
wolffd@0
|
45 <p><CODE>options(5)</CODE> is set to 1 if a covariance matrix is reset to its
|
|
wolffd@0
|
46 original value when any of its singular values are too small (less
|
|
wolffd@0
|
47 than MIN_COVAR which has the value eps).
|
|
wolffd@0
|
48 With the default value of 0 no action is taken.
|
|
wolffd@0
|
49
|
|
wolffd@0
|
50 <p><CODE>options(14)</CODE> is the maximum number of iterations; default 100.
|
|
wolffd@0
|
51
|
|
wolffd@0
|
52 <p>The optional return value <CODE>options</CODE> contains the final error value
|
|
wolffd@0
|
53 (i.e. data log likelihood) in
|
|
wolffd@0
|
54 <CODE>options(8)</CODE>.
|
|
wolffd@0
|
55
|
|
wolffd@0
|
56 <p><h2>
|
|
wolffd@0
|
57 Examples
|
|
wolffd@0
|
58 </h2>
|
|
wolffd@0
|
59 The following code fragment sets up a Gaussian mixture model, initialises
|
|
wolffd@0
|
60 the parameters from the data, sets the options and trains the model.
|
|
wolffd@0
|
61 <PRE>
|
|
wolffd@0
|
62
|
|
wolffd@0
|
63 mix = gmm(inputdim, ncentres, 'full');
|
|
wolffd@0
|
64
|
|
wolffd@0
|
65 <p>options = foptions;
|
|
wolffd@0
|
66 options(14) = 5;
|
|
wolffd@0
|
67 mix = gmminit(mix, data, options);
|
|
wolffd@0
|
68
|
|
wolffd@0
|
69 <p>options(1) = 1; % Prints out error values.
|
|
wolffd@0
|
70 options(14) = 30; % Max. number of iterations.
|
|
wolffd@0
|
71
|
|
wolffd@0
|
72 <p>mix = gmmem(mix, data, options);
|
|
wolffd@0
|
73 </PRE>
|
|
wolffd@0
|
74
|
|
wolffd@0
|
75
|
|
wolffd@0
|
76 <p><h2>
|
|
wolffd@0
|
77 See Also
|
|
wolffd@0
|
78 </h2>
|
|
wolffd@0
|
79 <CODE><a href="gmm.htm">gmm</a></CODE>, <CODE><a href="gmminit.htm">gmminit</a></CODE><hr>
|
|
wolffd@0
|
80 <b>Pages:</b>
|
|
wolffd@0
|
81 <a href="index.htm">Index</a>
|
|
wolffd@0
|
82 <hr>
|
|
wolffd@0
|
83 <p>Copyright (c) Ian T Nabney (1996-9)
|
|
wolffd@0
|
84
|
|
wolffd@0
|
85
|
|
wolffd@0
|
86 </body>
|
|
wolffd@0
|
87 </html> |
