annotate toolboxes/FullBNT-1.0.7/nethelp3.3/gmmem.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
rev   line source
Daniel@0 1 <html>
Daniel@0 2 <head>
Daniel@0 3 <title>
Daniel@0 4 Netlab Reference Manual gmmem
Daniel@0 5 </title>
Daniel@0 6 </head>
Daniel@0 7 <body>
Daniel@0 8 <H1> gmmem
Daniel@0 9 </H1>
Daniel@0 10 <h2>
Daniel@0 11 Purpose
Daniel@0 12 </h2>
Daniel@0 13 EM algorithm for Gaussian mixture model.
Daniel@0 14
Daniel@0 15 <p><h2>
Daniel@0 16 Synopsis
Daniel@0 17 </h2>
Daniel@0 18 <PRE>
Daniel@0 19
Daniel@0 20 [mix, options, errlog] = gmmem(mix, x, options)
Daniel@0 21 </PRE>
Daniel@0 22
Daniel@0 23
Daniel@0 24 <p><h2>
Daniel@0 25 Description
Daniel@0 26 </h2>
Daniel@0 27 <CODE>[mix, options, errlog] = gmmem(mix, x, options)</CODE> uses the Expectation
Daniel@0 28 Maximization algorithm of Dempster et al. to estimate the parameters of
Daniel@0 29 a Gaussian mixture model defined by a data structure <CODE>mix</CODE>.
Daniel@0 30 The matrix <CODE>x</CODE> represents the data whose expectation
Daniel@0 31 is maximized, with each row corresponding to a vector.
Daniel@0 32
Daniel@0 33 The optional parameters have the following interpretations.
Daniel@0 34
Daniel@0 35 <p><CODE>options(1)</CODE> is set to 1 to display error values; also logs error
Daniel@0 36 values in the return argument <CODE>errlog</CODE>.
Daniel@0 37 If <CODE>options(1)</CODE> is set to 0,
Daniel@0 38 then only warning messages are displayed. If <CODE>options(1)</CODE> is -1,
Daniel@0 39 then nothing is displayed.
Daniel@0 40
Daniel@0 41 <p><CODE>options(3)</CODE> is a measure of the absolute precision required of the error
Daniel@0 42 function at the solution. If the change in log likelihood between two steps of
Daniel@0 43 the EM algorithm is less than this value, then the function terminates.
Daniel@0 44
Daniel@0 45 <p><CODE>options(5)</CODE> is set to 1 if a covariance matrix is reset to its
Daniel@0 46 original value when any of its singular values are too small (less
Daniel@0 47 than MIN_COVAR which has the value eps).
Daniel@0 48 With the default value of 0 no action is taken.
Daniel@0 49
Daniel@0 50 <p><CODE>options(14)</CODE> is the maximum number of iterations; default 100.
Daniel@0 51
Daniel@0 52 <p>The optional return value <CODE>options</CODE> contains the final error value
Daniel@0 53 (i.e. data log likelihood) in
Daniel@0 54 <CODE>options(8)</CODE>.
Daniel@0 55
Daniel@0 56 <p><h2>
Daniel@0 57 Examples
Daniel@0 58 </h2>
Daniel@0 59 The following code fragment sets up a Gaussian mixture model, initialises
Daniel@0 60 the parameters from the data, sets the options and trains the model.
Daniel@0 61 <PRE>
Daniel@0 62
Daniel@0 63 mix = gmm(inputdim, ncentres, 'full');
Daniel@0 64
Daniel@0 65 <p>options = foptions;
Daniel@0 66 options(14) = 5;
Daniel@0 67 mix = gmminit(mix, data, options);
Daniel@0 68
Daniel@0 69 <p>options(1) = 1; % Prints out error values.
Daniel@0 70 options(14) = 30; % Max. number of iterations.
Daniel@0 71
Daniel@0 72 <p>mix = gmmem(mix, data, options);
Daniel@0 73 </PRE>
Daniel@0 74
Daniel@0 75
Daniel@0 76 <p><h2>
Daniel@0 77 See Also
Daniel@0 78 </h2>
Daniel@0 79 <CODE><a href="gmm.htm">gmm</a></CODE>, <CODE><a href="gmminit.htm">gmminit</a></CODE><hr>
Daniel@0 80 <b>Pages:</b>
Daniel@0 81 <a href="index.htm">Index</a>
Daniel@0 82 <hr>
Daniel@0 83 <p>Copyright (c) Ian T Nabney (1996-9)
Daniel@0 84
Daniel@0 85
Daniel@0 86 </body>
Daniel@0 87 </html>