Mercurial > hg > camir-aes2014

annotate 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
rev   line source
wolffd@0 1 <html>
wolffd@0 2 <head>
wolffd@0 3 <title>
wolffd@0 4 Netlab Reference Manual kmeans
wolffd@0 5 </title>
wolffd@0 6 </head>
wolffd@0 7 <body>
wolffd@0 8 <H1> kmeans
wolffd@0 9 </H1>
wolffd@0 10 <h2>
wolffd@0 11 Purpose
wolffd@0 12 </h2>
wolffd@0 13 Trains a k means cluster model.
wolffd@0 14
wolffd@0 15 <p><h2>
wolffd@0 16 Synopsis
wolffd@0 17 </h2>
wolffd@0 18 <PRE>
wolffd@0 19 centres = kmeans(centres, data, options)
wolffd@0 20 [centres, options] = kmeans(centres, data, options)
wolffd@0 21 [centres, options, post, errlog] = kmeans(centres, data, options)
wolffd@0 22 </PRE>
wolffd@0 23
wolffd@0 24
wolffd@0 25 <p><h2>
wolffd@0 26 Description
wolffd@0 27 </h2>
wolffd@0 28
wolffd@0 29 <CODE>centres = kmeans(centres, data, options)</CODE>
wolffd@0 30 uses the batch K-means algorithm to set the centres of a cluster model.
wolffd@0 31 The matrix <CODE>data</CODE> represents the data
wolffd@0 32 which is being clustered, with each row corresponding to a vector.
wolffd@0 33 The sum of squares error function is used. The point at which
wolffd@0 34 a local minimum is achieved is returned as <CODE>centres</CODE>. The
wolffd@0 35 error value at that point is returned in <CODE>options(8)</CODE>.
wolffd@0 36
wolffd@0 37 <p><CODE>[centres, options, post, errlog] = kmeans(centres, data, options)</CODE>
wolffd@0 38 also returns the cluster number (in a one-of-N encoding) for each data
wolffd@0 39 point in <CODE>post</CODE> and a log of the error values after each cycle in
wolffd@0 40 <CODE>errlog</CODE>.
wolffd@0 41
wolffd@0 42 The optional parameters have the following interpretations.
wolffd@0 43
wolffd@0 44 <p><CODE>options(1)</CODE> is set to 1 to display error values; also logs error
wolffd@0 45 values in the return argument <CODE>errlog</CODE>.
wolffd@0 46 If <CODE>options(1)</CODE> is set to 0,
wolffd@0 47 then only warning messages are displayed. If <CODE>options(1)</CODE> is -1,
wolffd@0 48 then nothing is displayed.
wolffd@0 49
wolffd@0 50 <p><CODE>options(2)</CODE> is a measure of the absolute precision required for the value
wolffd@0 51 of <CODE>centres</CODE> at the solution. If the absolute difference between
wolffd@0 52 the values of <CODE>centres</CODE> between two successive steps is less than
wolffd@0 53 <CODE>options(2)</CODE>, then this condition is satisfied.
wolffd@0 54
wolffd@0 55 <p><CODE>options(3)</CODE> is a measure of the precision required of the error
wolffd@0 56 function at the solution. If the absolute difference between the
wolffd@0 57 error functions between two successive steps is less than
wolffd@0 58 <CODE>options(3)</CODE>, then this condition is satisfied.
wolffd@0 59 Both this and the previous condition must be
wolffd@0 60 satisfied for termination.
wolffd@0 61
wolffd@0 62 <p><CODE>options(14)</CODE> is the maximum number of iterations; default 100.
wolffd@0 63
wolffd@0 64 <p><h2>
wolffd@0 65 Example
wolffd@0 66 </h2>
wolffd@0 67 <CODE>kmeans</CODE> can be used to initialise the centres of a Gaussian
wolffd@0 68 mixture model that is then trained with the EM algorithm.
wolffd@0 69 <PRE>
wolffd@0 70
wolffd@0 71 [priors, centres, var] = gmmunpak(p, md);
wolffd@0 72 centres = kmeans(centres, data, options);
wolffd@0 73 p = gmmpak(priors, centres, var);
wolffd@0 74 p = gmmem(p, md, data, options);
wolffd@0 75 </PRE>
wolffd@0 76
wolffd@0 77
wolffd@0 78 <p><h2>
wolffd@0 79 See Also
wolffd@0 80 </h2>
wolffd@0 81 <CODE><a href="gmminit.htm">gmminit</a></CODE>, <CODE><a href="gmmem.htm">gmmem</a></CODE><hr>
wolffd@0 82 <b>Pages:</b>
wolffd@0 83 <a href="index.htm">Index</a>
wolffd@0 84 <hr>
wolffd@0 85 <p>Copyright (c) Ian T Nabney (1996-9)
wolffd@0 86
wolffd@0 87
wolffd@0 88 </body>
wolffd@0 89 </html>