doi
/
NHP-pRF
másolva ChrisKlink/NHP-pRF


			
			
				
					
						
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766
							<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
                "http://www.w3.org/TR/REC-html40/loose.dtd">
<html>
<head>
  <title>Description of auto_classify</title>
  <meta name="keywords" content="auto_classify">
  <meta name="description" content="AUTO_CLASSIFY M-file for auto_classify.fig">
  <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
  <meta name="generator" content="m2html &copy; 2005 Guillaume Flandin">
  <meta name="robots" content="index, follow">
  <link type="text/css" rel="stylesheet" href="../../m2html.css">
  <script type="text/javascript">
    if (top.frames.length == 0) { top.location = "../../index.html"; };
  </script>
</head>
<body>
<a name="_top"></a>
<!-- ../menu.html chronux_2_10 --><!-- menu.html wave_browser -->
<h1>auto_classify
</h1>

<h2><a name="_name"></a>PURPOSE <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
<div class="box"><strong>AUTO_CLASSIFY M-file for auto_classify.fig</strong></div>

<h2><a name="_synopsis"></a>SYNOPSIS <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
<div class="box"><strong>function varargout = auto_classify(varargin) </strong></div>

<h2><a name="_description"></a>DESCRIPTION <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
<div class="fragment"><pre class="comment"> AUTO_CLASSIFY M-file for auto_classify.fig
      AUTO_CLASSIFY, by itself, creates a new AUTO_CLASSIFY or raises the existing
      singleton*.

      H = AUTO_CLASSIFY returns the handle to a new AUTO_CLASSIFY or the handle to
      the existing singleton*.

      AUTO_CLASSIFY('CALLBACK',hObject,eventData,handles,...) calls the local
      function named CALLBACK in AUTO_CLASSIFY.M with the given input arguments.

      AUTO_CLASSIFY('Property','Value',...) creates a new AUTO_CLASSIFY or raises the
      existing singleton*.  Starting from the left, property value pairs are
      applied to the GUI before auto_classify_OpeningFunction gets called.  An
      unrecognized property name or invalid value makes property application
      stop.  All inputs are passed to auto_classify_OpeningFcn via varargin.

      *See GUI Options on GUIDE's Tools menu.  Choose &quot;GUI allows only one
      instance to run (singleton)&quot;.

 See also: GUIDE, GUIDATA, GUIHANDLES</pre></div>

<!-- crossreference -->
<h2><a name="_cross"></a>CROSS-REFERENCE INFORMATION <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
This function calls:
<ul style="list-style-image:url(../../matlabicon.gif)">
<li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006a/@videoReader/close.html" class="code" title="function vr = close(vr)">close</a>	VR=CLOSE(VR)</li><li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006a/@videoWriter/close.html" class="code" title="function vw = close(vw)">close</a>	VW=CLOSE(VW)</li><li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006b/@videoReader/close.html" class="code" title="function vr = close(vr)">close</a>	VR=CLOSE(VR)</li><li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006b/@videoWriter/close.html" class="code" title="function vw = close(vw)">close</a>	VW=CLOSE(VW)</li><li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2007a/@videoReader/close.html" class="code" title="function vr = close(vr)">close</a>	VR=CLOSE(VR)</li><li><a href="../../chronux_2_10/fly_track/videoIO/videoIO_2007a/@videoWriter/close.html" class="code" title="function vw = close(vw)">close</a>	VW=CLOSE(VW)</li></ul>
This function is called by:
<ul style="list-style-image:url(../../matlabicon.gif)">
<li><a href="classify_spectra.html" class="code" title="function varargout = classify_spectra(varargin)">classify_spectra</a>	CLASSIFY_SPECTRA M-file for classify_spectra.fig</li></ul>
<!-- crossreference -->

<h2><a name="_subfunctions"></a>SUBFUNCTIONS <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
<ul style="list-style-image:url(../../matlabicon.gif)">
<li><a href="#_sub1" class="code">function auto_classify_OpeningFcn(hObject, eventdata, handles, varargin)</a></li><li><a href="#_sub2" class="code">function varargout = auto_classify_OutputFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub3" class="code">function clusters = clusterdata(X,k,dmeasure)</a></li><li><a href="#_sub4" class="code">function AutoClassifyButton_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub5" class="code">function CancelButton_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub6" class="code">function AcceptButton_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub7" class="code">function CepstralEdit_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub8" class="code">function CepstralEdit_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub9" class="code">function MinClusters_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub10" class="code">function MinClusters_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub11" class="code">function MaxClusters_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub12" class="code">function MaxClusters_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub13" class="code">function TransformPopupMenu_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub14" class="code">function TransformPopupMenu_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub15" class="code">function KClasses_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub16" class="code">function KClasses_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub17" class="code">function CepstralPopupMenu_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub18" class="code">function CepstralPopupMenu_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub19" class="code">function RangeSpecify_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub20" class="code">function DistancePopupMenu_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub21" class="code">function DistancePopupMenu_CreateFcn(hObject, eventdata, handles)</a></li><li><a href="#_sub22" class="code">function DiagnosticCheckbox_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub23" class="code">function pcaplot(matrix2classify,classification)</a></li><li><a href="#_sub24" class="code">function classifymatrix = madnormalize(classifymatrix, cols2normalize)</a></li><li><a href="#_sub25" class="code">function c=cluster_aylin(p1,p2);</a></li><li><a href="#_sub26" class="code">function [p1,p2]=hneighbors(e);</a></li><li><a href="#_sub27" class="code">function classification = cluster_hierarchical(matrix2classify)</a></li><li><a href="#_sub28" class="code">function HierarchicalRadioButton_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub29" class="code">function KmeansRadioButton_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub30" class="code">function SegLenWeight_Callback(hObject, eventdata, handles)</a></li><li><a href="#_sub31" class="code">function SegLenWeight_CreateFcn(hObject, eventdata, handles)</a></li></ul>

<h2><a name="_source"></a>SOURCE CODE <a href="#_top"><img alt="^" border="0" src="../../up.png"></a></h2>
<div class="fragment"><pre>0001 <a name="_sub0" href="#_subfunctions" class="code">function varargout = auto_classify(varargin)</a>
0002 <span class="comment">% AUTO_CLASSIFY M-file for auto_classify.fig</span>
0003 <span class="comment">%      AUTO_CLASSIFY, by itself, creates a new AUTO_CLASSIFY or raises the existing</span>
0004 <span class="comment">%      singleton*.</span>
0005 <span class="comment">%</span>
0006 <span class="comment">%      H = AUTO_CLASSIFY returns the handle to a new AUTO_CLASSIFY or the handle to</span>
0007 <span class="comment">%      the existing singleton*.</span>
0008 <span class="comment">%</span>
0009 <span class="comment">%      AUTO_CLASSIFY('CALLBACK',hObject,eventData,handles,...) calls the local</span>
0010 <span class="comment">%      function named CALLBACK in AUTO_CLASSIFY.M with the given input arguments.</span>
0011 <span class="comment">%</span>
0012 <span class="comment">%      AUTO_CLASSIFY('Property','Value',...) creates a new AUTO_CLASSIFY or raises the</span>
0013 <span class="comment">%      existing singleton*.  Starting from the left, property value pairs are</span>
0014 <span class="comment">%      applied to the GUI before auto_classify_OpeningFunction gets called.  An</span>
0015 <span class="comment">%      unrecognized property name or invalid value makes property application</span>
0016 <span class="comment">%      stop.  All inputs are passed to auto_classify_OpeningFcn via varargin.</span>
0017 <span class="comment">%</span>
0018 <span class="comment">%      *See GUI Options on GUIDE's Tools menu.  Choose &quot;GUI allows only one</span>
0019 <span class="comment">%      instance to run (singleton)&quot;.</span>
0020 <span class="comment">%</span>
0021 <span class="comment">% See also: GUIDE, GUIDATA, GUIHANDLES</span>
0022 
0023 <span class="comment">% Copyright 2002-2003 The MathWorks, Inc.</span>
0024 
0025 <span class="comment">% Edit the above text to modify the response to help auto_classify</span>
0026 
0027 <span class="comment">% Last Modified by GUIDE v2.5 21-Jun-2006 00:34:13</span>
0028 
0029 <span class="comment">% Begin initialization code - DO NOT EDIT</span>
0030 gui_Singleton = 1;
0031 gui_State = struct(<span class="string">'gui_Name'</span>,       mfilename, <span class="keyword">...</span>
0032                    <span class="string">'gui_Singleton'</span>,  gui_Singleton, <span class="keyword">...</span>
0033                    <span class="string">'gui_OpeningFcn'</span>, @<a href="#_sub1" class="code" title="subfunction auto_classify_OpeningFcn(hObject, eventdata, handles, varargin)">auto_classify_OpeningFcn</a>, <span class="keyword">...</span>
0034                    <span class="string">'gui_OutputFcn'</span>,  @<a href="#_sub2" class="code" title="subfunction varargout = auto_classify_OutputFcn(hObject, eventdata, handles)">auto_classify_OutputFcn</a>, <span class="keyword">...</span>
0035                    <span class="string">'gui_LayoutFcn'</span>,  [] , <span class="keyword">...</span>
0036                    <span class="string">'gui_Callback'</span>,   []);
0037 <span class="keyword">if</span> nargin &amp;&amp; ischar(varargin{1})
0038     gui_State.gui_Callback = str2func(varargin{1});
0039 <span class="keyword">end</span>
0040 
0041 <span class="keyword">if</span> nargout
0042     [varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:});
0043 <span class="keyword">else</span>
0044     gui_mainfcn(gui_State, varargin{:});
0045 <span class="keyword">end</span>
0046 <span class="comment">% End initialization code - DO NOT EDIT</span>
0047 
0048 
0049 <span class="comment">% --- Executes just before auto_classify is made visible.</span>
0050 <a name="_sub1" href="#_subfunctions" class="code">function auto_classify_OpeningFcn(hObject, eventdata, handles, varargin)</a>
0051 <span class="comment">% This function has no output args, see OutputFcn.</span>
0052 <span class="comment">% hObject    handle to figure</span>
0053 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0054 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0055 <span class="comment">% varargin   command line arguments to auto_classify (see VARARGIN)</span>
0056 
0057 <span class="comment">% input parameters</span>
0058 
0059 <span class="keyword">if</span> isempty(varargin)
0060     handles.matrix2classify = [];
0061     handles.ncepstral = 5;
0062 <span class="keyword">else</span>
0063     handles.matrix2classify = varargin{1};
0064     handles.ncepstral = varargin{2};
0065 <span class="keyword">end</span>
0066 
0067 handles.distancemeasure = <span class="string">'sqEuclidean'</span>;
0068 
0069 handles.cluster_method = <span class="string">'kmeans'</span>;
0070 
0071 set(handles.CepstralPopupMenu,<span class="string">'String'</span>,num2str([0:handles.ncepstral]'));
0072 set(handles.CepstralPopupMenu,<span class="string">'Value'</span>,handles.ncepstral); <span class="comment">%handles.ncepstral +1 );</span>
0073 <span class="comment">% set(handles.CepstralPopupMenu,'Value',handles.ncepstral-1);</span>
0074 
0075 <span class="comment">% Choose default command line output for auto_classify</span>
0076 <span class="comment">% handles.output = hObject;</span>
0077 
0078 <span class="comment">% Update handles structure</span>
0079 guidata(hObject, handles);
0080 
0081 <span class="comment">% UIWAIT makes auto_classify wait for user response (see UIRESUME)</span>
0082 <span class="comment">% uiwait(handles.figure1);</span>
0083 
0084 uiwait(handles.figure1);
0085     
0086 <span class="comment">% --- Outputs from this function are returned to the command line.</span>
0087 <a name="_sub2" href="#_subfunctions" class="code">function varargout = auto_classify_OutputFcn(hObject, eventdata, handles) </a>
0088 <span class="comment">% varargout  cell array for returning output args (see VARARGOUT);</span>
0089 <span class="comment">% hObject    handle to figure</span>
0090 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0091 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0092 
0093 <span class="comment">% Get default command line output from handles structure</span>
0094 varargout{1} = handles.output;
0095 
0096 <a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006a/@videoReader/close.html" class="code" title="function vr = close(vr)">close</a>;
0097 
0098 <a name="_sub3" href="#_subfunctions" class="code">function clusters = clusterdata(X,k,dmeasure)</a>
0099 <span class="comment">% A wrapper function for clustering</span>
0100 
0101 <span class="keyword">try</span>
0102     clusters = kmeans(X,k,<span class="string">'distance'</span>,dmeasure,<span class="string">'EmptyAction'</span>,<span class="string">'singleton'</span>,<span class="string">'replicates'</span>,30);
0103 <span class="keyword">catch</span>
0104     clusters = k;
0105 <span class="keyword">end</span>
0106 
0107 <span class="comment">% --- Executes on button press in AutoClassifyButton.</span>
0108 <a name="_sub4" href="#_subfunctions" class="code">function AutoClassifyButton_Callback(hObject, eventdata, handles)</a>
0109 <span class="comment">% hObject    handle to AutoClassifyButton (see GCBO)</span>
0110 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0111 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0112 
0113 set(handles.AcceptButton,<span class="string">'Enable'</span>,<span class="string">'off'</span>);
0114 guidata(gcbo,handles);
0115 
0116 cncepstral = get(handles.CepstralPopupMenu,<span class="string">'Value'</span>);
0117 
0118 matrix2classify = handles.matrix2classify;
0119 matrix2classify = <a href="#_sub24" class="code" title="subfunction classifymatrix = madnormalize(classifymatrix, cols2normalize)">madnormalize</a>(matrix2classify,1); <span class="comment">% normalize the columns</span>
0120 matrix2classify = matrix2classify(:,1:cncepstral+1); <span class="comment">% include the cepstral coefficients that you want</span>
0121 
0122 <span class="keyword">if</span> length(matrix2classify(1,:)) &gt; 3
0123     matrix2classify = matrix2classify(:,[1,3:cncepstral+1]); <span class="comment">% exclude the first cepestral coefficient</span>
0124 <span class="keyword">end</span>
0125 
0126 <span class="keyword">try</span>
0127     segweight = str2num(get(handles.SegLenWeight,<span class="string">'String'</span>));
0128 <span class="keyword">catch</span>
0129     segweight = 1;
0130     set(handles.SegLenWeight,<span class="string">'String'</span>,<span class="string">'1'</span>);
0131 <span class="keyword">end</span>
0132 
0133 <span class="comment">% Weight the segment lengths by square root of the weight</span>
0134 matrix2classify(:,1) = matrix2classify(:,1)*sqrt(segweight);
0135  
0136 <span class="comment">% Other weightings can be added here but the cepestral coefficients have a</span>
0137 <span class="comment">% natural exponential decline which acts a weighting</span>
0138 
0139 <span class="comment">%Handle transformations to segment length</span>
0140 
0141 contents = get(handles.TransformPopupMenu,<span class="string">'String'</span>);
0142 value = get(handles.TransformPopupMenu,<span class="string">'Value'</span>);
0143 
0144 <span class="keyword">switch</span> contents{value}
0145     <span class="keyword">case</span> <span class="string">'None'</span>
0146         matrix2classify = matrix2classify;
0147     <span class="keyword">case</span> <span class="string">'Exclude'</span>
0148         matrix2classify = matrix2classify(:,2:length(matrix2classify(1,:)));
0149     <span class="keyword">case</span> <span class="string">'Log'</span>
0150         matrix2classify(:,1) = log(matrix2classify(:,1));
0151     <span class="keyword">case</span> <span class="string">'Exp'</span>
0152         matrix2classify(:,1) = exp(matrix2classify(:,1));
0153 <span class="keyword">end</span>
0154 
0155 <span class="keyword">if</span> length(matrix2classify(1,:)) == 0 <span class="comment">% matrix to classify has no information</span>
0156     <span class="keyword">return</span>
0157 <span class="keyword">end</span>
0158 
0159 rangestate = get(handles.RangeSpecify,<span class="string">'Value'</span>);
0160 diagnosticstate = get(handles.DiagnosticCheckbox,<span class="string">'Value'</span>);
0161 
0162 <span class="keyword">if</span> strcmp(handles.cluster_method,<span class="string">'hierarchical'</span>)
0163     handles.classification = <a href="#_sub27" class="code" title="subfunction classification = cluster_hierarchical(matrix2classify)">cluster_hierarchical</a>(matrix2classify);
0164     nclusters = length(unique(handles.classification)) - 1;
0165     set(handles.KClasses,<span class="string">'String'</span>,nclusters);
0166     
0167     <span class="keyword">if</span> diagnosticstate
0168         <a href="#_sub23" class="code" title="subfunction pcaplot(matrix2classify,classification)">pcaplot</a>(matrix2classify,handles.classification);
0169     <span class="keyword">end</span>
0170     
0171 <span class="keyword">else</span> <span class="keyword">if</span> strcmp(handles.cluster_method, <span class="string">'kmeans'</span>)
0172 
0173 <span class="keyword">if</span> not(rangestate)
0174 
0175     kclassstr = get(handles.KClasses,<span class="string">'String'</span>);
0176 
0177     <span class="keyword">if</span> not(isempty(kclassstr))    
0178         <span class="keyword">try</span>
0179             nclasses = str2num(kclassstr);
0180         <span class="keyword">catch</span>
0181             <span class="keyword">return</span>;
0182         <span class="keyword">end</span>
0183     
0184         classification =  <a href="#_sub3" class="code" title="subfunction clusters = clusterdata(X,k,dmeasure)">clusterdata</a>(matrix2classify,nclasses,handles.distancemeasure);<span class="comment">% this can be changed to a different clustering algorithm</span>
0185     
0186         <span class="comment">% clusterdata shows how a clustering algorithm can be hooked in where the number of</span>
0187         <span class="comment">% classes need to be assigned.</span>
0188         
0189         <span class="keyword">if</span> length(classification) == 1
0190             msgbox(<span class="string">'Error in clustering. Try changing the number of target clusters or rerunning.'</span>);
0191         <span class="keyword">else</span>
0192             <span class="keyword">if</span> diagnosticstate
0193                 figure();
0194                 [silh h] = silhouette(matrix2classify,classification);
0195                 <a href="#_sub23" class="code" title="subfunction pcaplot(matrix2classify,classification)">pcaplot</a>(matrix2classify,classification);
0196             <span class="keyword">end</span>
0197         handles.classification = classification;
0198         <span class="keyword">end</span>
0199     <span class="keyword">else</span>
0200         <span class="keyword">return</span>;
0201     <span class="keyword">end</span>
0202 <span class="keyword">else</span>
0203     <span class="keyword">try</span>
0204         minclust = str2num(get(handles.MinClusters,<span class="string">'String'</span>));
0205     <span class="keyword">catch</span>
0206         <span class="keyword">return</span>;
0207     <span class="keyword">end</span>
0208     <span class="keyword">try</span>
0209         maxclust = str2num(get(handles.MaxClusters,<span class="string">'String'</span>));
0210     <span class="keyword">catch</span>
0211         <span class="keyword">return</span>;
0212     <span class="keyword">end</span> 
0213     
0214     resultsclust = cell((1+ maxclust)-minclust,1);
0215     
0216     hw = waitbar(0,<span class="string">'Clustering data set '</span>);
0217     <span class="keyword">for</span> i = 1:(1+maxclust)-minclust
0218         resultsclust{i} = <a href="#_sub3" class="code" title="subfunction clusters = clusterdata(X,k,dmeasure)">clusterdata</a>(matrix2classify,minclust + (i-1),handles.distancemeasure);
0219         <span class="keyword">if</span> length(resultsclust{i}) == 1
0220             msgbox([<span class="string">'Clustering failed at '</span> num2str(resultsclust{i}) <span class="string">' clusters. Try changing the range of clusters or rerunning.'</span>]);
0221             <span class="keyword">return</span>;
0222         <span class="keyword">end</span>
0223         waitbar(i/(1+maxclust-minclust));
0224     <span class="keyword">end</span>
0225     <a href="../../chronux_2_10/fly_track/videoIO/videoIO_2006a/@videoReader/close.html" class="code" title="function vr = close(vr)">close</a>(hw);
0226     
0227     silhmean = zeros((1+ maxclust)-minclust,1);
0228     
0229     <span class="keyword">for</span> i = 1:1+(maxclust)-minclust
0230         
0231         <span class="keyword">if</span> diagnosticstate
0232             figure();
0233             [silh h] = silhouette(matrix2classify,resultsclust{i});
0234             <a href="#_sub23" class="code" title="subfunction pcaplot(matrix2classify,classification)">pcaplot</a>(matrix2classify,resultsclust{i});
0235         <span class="keyword">else</span>
0236             silh = silhouette(matrix2classify,resultsclust{i});
0237         <span class="keyword">end</span>
0238             
0239         silhmean(i) = mean(silh);
0240     <span class="keyword">end</span>
0241     
0242     [x I] = max(silhmean);
0243     handles.classification = resultsclust{I(1)};
0244     
0245     set(handles.KClasses,<span class="string">'String'</span>,num2str(minclust + (I(1)-1)));
0246 <span class="keyword">end</span>
0247 <span class="keyword">end</span>
0248     <span class="keyword">end</span>
0249 
0250 
0251 set(handles.AcceptButton,<span class="string">'Enable'</span>,<span class="string">'on'</span>);
0252 
0253 guidata(gcbo,handles);
0254 
0255 <span class="comment">% --- Executes on button press in CancelButton.</span>
0256 <a name="_sub5" href="#_subfunctions" class="code">function CancelButton_Callback(hObject, eventdata, handles)</a>
0257 <span class="comment">% hObject    handle to CancelButton (see GCBO)</span>
0258 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0259 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0260 
0261 handles.output = 0;
0262 guidata(hObject,handles);
0263 uiresume(handles.figure1);
0264 
0265 <span class="comment">% --- Executes on button press in AcceptButton.</span>
0266 <a name="_sub6" href="#_subfunctions" class="code">function AcceptButton_Callback(hObject, eventdata, handles)</a>
0267 <span class="comment">% hObject    handle to AcceptButton (see GCBO)</span>
0268 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0269 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0270 
0271 handles.output = handles.classification;
0272 guidata(hObject,handles);
0273 uiresume(handles.figure1);
0274 
0275 
0276 <a name="_sub7" href="#_subfunctions" class="code">function CepstralEdit_Callback(hObject, eventdata, handles)</a>
0277 <span class="comment">% hObject    handle to CepstralEdit (see GCBO)</span>
0278 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0279 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0280 
0281 <span class="comment">% Hints: get(hObject,'String') returns contents of CepstralEdit as text</span>
0282 <span class="comment">%        str2double(get(hObject,'String')) returns contents of CepstralEdit as a double</span>
0283 
0284 
0285 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0286 <a name="_sub8" href="#_subfunctions" class="code">function CepstralEdit_CreateFcn(hObject, eventdata, handles)</a>
0287 <span class="comment">% hObject    handle to CepstralEdit (see GCBO)</span>
0288 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0289 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0290 
0291 <span class="comment">% Hint: edit controls usually have a white background on Windows.</span>
0292 <span class="comment">%       See ISPC and COMPUTER.</span>
0293 <span class="keyword">if</span> ispc
0294     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0295 <span class="keyword">else</span>
0296     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0297 <span class="keyword">end</span>
0298 
0299 
0300 
0301 <a name="_sub9" href="#_subfunctions" class="code">function MinClusters_Callback(hObject, eventdata, handles)</a>
0302 <span class="comment">% hObject    handle to MinClusters (see GCBO)</span>
0303 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0304 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0305 
0306 <span class="comment">% Hints: get(hObject,'String') returns contents of MinClusters as text</span>
0307 <span class="comment">%        str2double(get(hObject,'String')) returns contents of MinClusters as a double</span>
0308 
0309 
0310 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0311 <a name="_sub10" href="#_subfunctions" class="code">function MinClusters_CreateFcn(hObject, eventdata, handles)</a>
0312 <span class="comment">% hObject    handle to MinClusters (see GCBO)</span>
0313 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0314 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0315 
0316 <span class="comment">% Hint: edit controls usually have a white background on Windows.</span>
0317 <span class="comment">%       See ISPC and COMPUTER.</span>
0318 <span class="keyword">if</span> ispc
0319     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0320 <span class="keyword">else</span>
0321     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0322 <span class="keyword">end</span>
0323 
0324 
0325 
0326 <a name="_sub11" href="#_subfunctions" class="code">function MaxClusters_Callback(hObject, eventdata, handles)</a>
0327 <span class="comment">% hObject    handle to MaxClusters (see GCBO)</span>
0328 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0329 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0330 
0331 <span class="comment">% Hints: get(hObject,'String') returns contents of MaxClusters as text</span>
0332 <span class="comment">%        str2double(get(hObject,'String')) returns contents of MaxClusters as a double</span>
0333 
0334 
0335 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0336 <a name="_sub12" href="#_subfunctions" class="code">function MaxClusters_CreateFcn(hObject, eventdata, handles)</a>
0337 <span class="comment">% hObject    handle to MaxClusters (see GCBO)</span>
0338 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0339 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0340 
0341 <span class="comment">% Hint: edit controls usually have a white background on Windows.</span>
0342 <span class="comment">%       See ISPC and COMPUTER.</span>
0343 <span class="keyword">if</span> ispc
0344     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0345 <span class="keyword">else</span>
0346     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0347 <span class="keyword">end</span>
0348 
0349 
0350 <span class="comment">% --- Executes on selection change in TransformPopupMenu.</span>
0351 <a name="_sub13" href="#_subfunctions" class="code">function TransformPopupMenu_Callback(hObject, eventdata, handles)</a>
0352 <span class="comment">% hObject    handle to TransformPopupMenu (see GCBO)</span>
0353 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0354 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0355 
0356 <span class="comment">% Hints: contents = get(hObject,'String') returns TransformPopupMenu contents as cell array</span>
0357 <span class="comment">%        contents{get(hObject,'Value')} returns selected item from TransformPopupMenu</span>
0358 
0359 <span class="comment">% value = get(hObject,'Value');</span>
0360 <span class="comment">%</span>
0361 <span class="comment">% if value == 2</span>
0362 <span class="comment">%     handles.matrix2classify(:,1) = log(handles.matrix2classify(:,1));</span>
0363 <span class="comment">% else</span>
0364 <span class="comment">%     handles.matrix2classify(:,1) = exp(handles.matrix2classify(:,1));</span>
0365 <span class="comment">% end</span>
0366 
0367 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0368 <a name="_sub14" href="#_subfunctions" class="code">function TransformPopupMenu_CreateFcn(hObject, eventdata, handles)</a>
0369 <span class="comment">% hObject    handle to TransformPopupMenu (see GCBO)</span>
0370 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0371 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0372 
0373 <span class="comment">% Hint: popupmenu controls usually have a white background on Windows.</span>
0374 <span class="comment">%       See ISPC and COMPUTER.</span>
0375 <span class="keyword">if</span> ispc
0376     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0377 <span class="keyword">else</span>
0378     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0379 <span class="keyword">end</span>
0380 
0381 <a name="_sub15" href="#_subfunctions" class="code">function KClasses_Callback(hObject, eventdata, handles)</a>
0382 <span class="comment">% hObject    handle to KClasses (see GCBO)</span>
0383 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0384 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0385 
0386 <span class="comment">% Hints: get(hObject,'String') returns contents of KClasses as text</span>
0387 <span class="comment">%        str2double(get(hObject,'String')) returns contents of KClasses as a double</span>
0388 
0389 
0390 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0391 <a name="_sub16" href="#_subfunctions" class="code">function KClasses_CreateFcn(hObject, eventdata, handles)</a>
0392 <span class="comment">% hObject    handle to KClasses (see GCBO)</span>
0393 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0394 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0395 
0396 <span class="comment">% Hint: edit controls usually have a white background on Windows.</span>
0397 <span class="comment">%       See ISPC and COMPUTER.</span>
0398 <span class="keyword">if</span> ispc
0399     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0400 <span class="keyword">else</span>
0401     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0402 <span class="keyword">end</span>
0403 
0404 
0405 <span class="comment">% --- Executes on selection change in CepstralPopupMenu.</span>
0406 <a name="_sub17" href="#_subfunctions" class="code">function CepstralPopupMenu_Callback(hObject, eventdata, handles)</a>
0407 <span class="comment">% hObject    handle to CepstralPopupMenu (see GCBO)</span>
0408 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0409 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0410 
0411 <span class="comment">% Hints: contents = get(hObject,'String') returns CepstralPopupMenu contents as cell array</span>
0412 <span class="comment">%        contents{get(hObject,'Value')} returns selected item from CepstralPopupMenu</span>
0413 
0414 
0415 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0416 <a name="_sub18" href="#_subfunctions" class="code">function CepstralPopupMenu_CreateFcn(hObject, eventdata, handles)</a>
0417 <span class="comment">% hObject    handle to CepstralPopupMenu (see GCBO)</span>
0418 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0419 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0420 
0421 <span class="comment">% Hint: popupmenu controls usually have a white background on Windows.</span>
0422 <span class="comment">%       See ISPC and COMPUTER.</span>
0423 <span class="keyword">if</span> ispc
0424     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0425 <span class="keyword">else</span>
0426     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0427 <span class="keyword">end</span>
0428 
0429 
0430 
0431 <span class="comment">% --- Executes on button press in RangeSpecify.</span>
0432 <a name="_sub19" href="#_subfunctions" class="code">function RangeSpecify_Callback(hObject, eventdata, handles)</a>
0433 <span class="comment">% hObject    handle to RangeSpecify (see GCBO)</span>
0434 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0435 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0436 
0437 <span class="comment">% Hint: get(hObject,'Value') returns toggle state of RangeSpecify</span>
0438 
0439 state = get(hObject,<span class="string">'Value'</span>);
0440 
0441 <span class="keyword">if</span> state
0442     set(handles.KClasses,<span class="string">'Enable'</span>,<span class="string">'off'</span>);
0443     set(handles.text3,<span class="string">'Visible'</span>,<span class="string">'on'</span>);
0444     set(handles.MinClusters,<span class="string">'Visible'</span>,<span class="string">'on'</span>);
0445     set(handles.text4,<span class="string">'Visible'</span>,<span class="string">'on'</span>);
0446     set(handles.MaxClusters,<span class="string">'Visible'</span>,<span class="string">'on'</span>);
0447 <span class="keyword">else</span>
0448     set(handles.KClasses,<span class="string">'Enable'</span>,<span class="string">'on'</span>);
0449     set(handles.text3,<span class="string">'Visible'</span>,<span class="string">'off'</span>);
0450     set(handles.MinClusters,<span class="string">'Visible'</span>,<span class="string">'off'</span>);
0451     set(handles.text4,<span class="string">'Visible'</span>,<span class="string">'off'</span>);
0452     set(handles.MaxClusters,<span class="string">'Visible'</span>,<span class="string">'off'</span>);
0453 <span class="keyword">end</span>
0454 
0455 
0456 <span class="comment">% --- Executes on selection change in DistancePopupMenu.</span>
0457 <a name="_sub20" href="#_subfunctions" class="code">function DistancePopupMenu_Callback(hObject, eventdata, handles)</a>
0458 <span class="comment">% hObject    handle to DistancePopupMenu (see GCBO)</span>
0459 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0460 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0461 
0462 <span class="comment">% Hints: contents = get(hObject,'String') returns DistancePopupMenu contents as cell array</span>
0463 <span class="comment">%        contents{get(hObject,'Value')} returns selected item from DistancePopupMenu</span>
0464 
0465 contents = get(hObject,<span class="string">'String'</span>);
0466 
0467 dstring = contents{get(hObject,<span class="string">'Value'</span>)};
0468 
0469 <span class="keyword">switch</span> dstring
0470     <span class="keyword">case</span> <span class="string">'Squared Euclidean'</span>
0471         handles.distancemeasure = <span class="string">'sqEuclidean'</span>
0472     <span class="keyword">case</span> <span class="string">'City Block (L1)'</span>
0473         handles.distancemeasure = <span class="string">'cityblock'</span>
0474     <span class="keyword">case</span> <span class="string">'Cosine'</span>
0475         handles.distancemeasure = <span class="string">'cosine'</span>
0476     <span class="keyword">case</span> <span class="string">'Correlation'</span>
0477         handles.distancemeasure = <span class="string">'correlation'</span>
0478 <span class="keyword">end</span>
0479 
0480 guidata(gcbo,handles);
0481 
0482 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0483 <a name="_sub21" href="#_subfunctions" class="code">function DistancePopupMenu_CreateFcn(hObject, eventdata, handles)</a>
0484 <span class="comment">% hObject    handle to DistancePopupMenu (see GCBO)</span>
0485 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0486 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0487 
0488 <span class="comment">% Hint: popupmenu controls usually have a white background on Windows.</span>
0489 <span class="comment">%       See ISPC and COMPUTER.</span>
0490 <span class="keyword">if</span> ispc
0491     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0492 <span class="keyword">else</span>
0493     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0494 <span class="keyword">end</span>
0495 
0496 handles.distancemeasure = <span class="string">'sqEuclidean'</span>;
0497 
0498 guidata(gcbo,handles);
0499 
0500 <span class="comment">% --- Executes on button press in DiagnosticCheckbox.</span>
0501 <a name="_sub22" href="#_subfunctions" class="code">function DiagnosticCheckbox_Callback(hObject, eventdata, handles)</a>
0502 <span class="comment">% hObject    handle to DiagnosticCheckbox (see GCBO)</span>
0503 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0504 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0505 
0506 <span class="comment">% Hint: get(hObject,'Value') returns toggle state of DiagnosticCheckbox</span>
0507 
0508 <a name="_sub23" href="#_subfunctions" class="code">function pcaplot(matrix2classify,classification)</a>
0509 <span class="comment">% Plots the first two principal components uses different shapes selected</span>
0510 <span class="comment">% randomly and different colors selected randomly to show classes.</span>
0511 
0512 nclusters = length(unique(classification));
0513 
0514 <span class="keyword">if</span> not(isempty(find(classification == 0)))
0515     nclusters = nclusters -1;
0516 <span class="keyword">end</span>
0517 
0518 colors = rand(nclusters,3);
0519 mark = mod(1:nclusters,13)+1;
0520 
0521 [coef, score] = princomp(matrix2classify);
0522 
0523 figure();
0524 
0525 <span class="keyword">for</span> i = 1:length(classification)
0526     h = line(score(i,1),score(i,2));
0527     markers = set(h,<span class="string">'Marker'</span>);
0528     <span class="keyword">if</span> classification(i) == 0 <span class="comment">% For segments which are not classified</span>
0529         set(h,<span class="string">'Marker'</span>,markers{4});   
0530         set(h,<span class="string">'Color'</span>,[0 0 0]);
0531     <span class="keyword">else</span>
0532         set(h,<span class="string">'Marker'</span>,markers{mark(classification(i))});   
0533         set(h,<span class="string">'Color'</span>,colors(classification(i),:));
0534     <span class="keyword">end</span>
0535 <span class="keyword">end</span>
0536 
0537 xlabel(<span class="string">'PCA 1'</span>);
0538 ylabel(<span class="string">'PCA 2'</span>);
0539 
0540 <a name="_sub24" href="#_subfunctions" class="code">function classifymatrix = madnormalize(classifymatrix, cols2normalize)</a>
0541   <span class="keyword">for</span> i = cols2normalize
0542     classifymatrix(:,i) = (classifymatrix(:,i) - median(classifymatrix(:,i))) / mad(classifymatrix(:,i));
0543   <span class="keyword">end</span>
0544   
0545   
0546 <span class="comment">% This code is modified from http://phys.columbia.edu/~aylin/clustering</span>
0547 
0548 <a name="_sub25" href="#_subfunctions" class="code">function c=cluster_aylin(p1,p2);</a>
0549 <span class="comment">%enter the p1,p2 found from hneighbors.m and get the struct array c.</span>
0550 <span class="comment">%c(i).c will contain the indices of the objects that is in cluster i.</span>
0551 <span class="comment">%c(1).c will be the cluster that has the maximum number of objects.</span>
0552 nn=size(p1,2);
0553 p=std(p1);g=p(p2);p=p';
0554 z=zeros(nn,1);zz=zeros(nn,1);
0555 n=zeros(1,nn);
0556 <span class="keyword">for</span> i=1:nn
0557     n(i)=max(max(p1(:,p2(:,i))));
0558 <span class="keyword">end</span>
0559 gg=[1:nn; max(g); n]';n=find(n&lt;1);
0560 [q1,q1]=sort(gg(n,2)');
0561 g=gg(n(q1),1);
0562 t=cputime;
0563 j=1;m=1
0564     <span class="keyword">for</span> i=1:length(n)
0565         ii=g(i);b=p2(:,ii);a=b;aa=a(find(gg(a,2)&lt;1));
0566         <span class="keyword">if</span> length([0; unique(z(aa))])&lt;=2;a1=1;a2=0;
0567             <span class="keyword">while</span> a1~=a2;a1=length(aa);
0568                 a=unique(p2(:,a));a=a(find(gg(a,2)&lt;1));
0569                 a=unique(a(find(gg(a,2)&lt;=mean(gg(aa,2))+std(gg(aa,2)))));
0570                 a=a(find(ismember(a,aa)==0));
0571                 <span class="keyword">if</span> ~isempty(a);aa=[aa;a];
0572                     jj=aa(find(z(aa)));
0573                     <span class="keyword">if</span> ~isempty(jj);;
0574                          u=unique(z(jj));
0575                          <span class="keyword">if</span> length(u)==1;
0576                             zz(aa(find(~z(aa))))=m;z(aa)=u;m=m+1;
0577                          <span class="keyword">end</span>
0578                          <span class="keyword">break</span>;
0579                     <span class="keyword">end</span>
0580                 <span class="keyword">end</span>;a2=length(aa);
0581             <span class="keyword">end</span>;a=aa;
0582             jj=a(find(z(a)));
0583             <span class="keyword">if</span> isempty(jj);
0584                 z(a)=j;j=j+1;
0585                 zz(a)=m;zz(b)=m-.1;zz(ii)=m-.2;m=m+1;
0586             <span class="keyword">end</span>
0587         <span class="keyword">end</span>
0588     <span class="keyword">end</span>
0589     u=unique(z)
0590     u=u(find(u));v=length(u);vv=floor(1/v);<span class="keyword">if</span> vv;vv=<span class="string">' is'</span>;<span class="keyword">else</span> vv=<span class="string">'s      are'</span>;<span class="keyword">end</span>
0591     fprintf([int2str(v) <span class="string">' cluster'</span> vv <span class="string">' found in '</span> num2str(cputime-t) <span class="string">'sec\n\n'</span>])
0592     q0=[];<span class="keyword">for</span> i=1:v;qq=find(z==u(i));q0=[q0 length(qq)];<span class="keyword">end</span>;
0593     [q1,q2]=sort(q0);<span class="comment">%q2=q2(find(100*q1/nn&gt;1));v=length(q2);</span>
0594     c=[];<span class="keyword">for</span> i=1:v;qq=find(z==u(q2(i)));[j,j]=sort(zz(qq));c(v-i+1).c=qq(j);<span class="keyword">end</span>;
0595     <span class="keyword">if</span> isempty(c);fprintf(<span class="string">'\tNo cluster was found. \n \tscale the data (step size must be 1)\n'</span>);<span class="keyword">end</span> 
0596     
0597     
0598     <a name="_sub26" href="#_subfunctions" class="code">function [p1,p2]=hneighbors(e);</a>
0599     <span class="comment">% this function finds the neighbors of each object in 'e' within a unit hypercube</span>
0600     <span class="comment">% and returns the sorted object distances to q.q1 and their identities to q.q2 , 'e' is a</span>
0601     <span class="comment">% matrix where i th row and j th column represents the j th component of the i th object.</span>
0602     s=<span class="string">'find('</span>;i=<span class="string">'abs(e(:,%d)-e(i,%d))&lt;1&amp;'</span>;
0603     <span class="comment">% assuming that the data is given scaled and the characteristic step size is 1, variable s</span>
0604     <span class="comment">% keeps a script to find the objects that lie within a unit hypercube around the i th object.</span>
0605     <span class="keyword">for</span> j=1:size(e,2);
0606         s=[s sprintf(i,j,j)];
0607     <span class="keyword">end</span>;s([end end+1])=<span class="string">');'</span>;
0608      <span class="comment">% runs the script s for each of the objects and stores the sorted distances</span>
0609      <span class="comment">% from the i th object in q(i).q1 and their indentities in q(i).q2</span>
0610     nn=size(e,1);m=ceil(nn^(1/4));p1=ones(m,nn);p2=kron(ones(m,1),1:nn);
0611     <span class="keyword">for</span> i=1:nn;
0612         j=eval(s);
0613         [q,qq]=sort(sqrt(sum((e(j,:)-kron(ones(length(j),1),e(i,:)))'.^2)));q=q(find(q&lt;1));mm=length(q);
0614         qq=j(qq(1:mm))';mn=min([m mm]);
0615         p1(1:mn,i)=q(1:mn)';
0616         p2(1:mn,i)=qq(1:mn)';
0617     <span class="keyword">end</span>
0618     
0619     
0620 <a name="_sub27" href="#_subfunctions" class="code">function classification = cluster_hierarchical(matrix2classify)</a>
0621     [p1,p2] = <a href="#_sub26" class="code" title="subfunction [p1,p2]=hneighbors(e);">hneighbors</a>(matrix2classify);
0622     c = <a href="#_sub25" class="code" title="subfunction c=cluster_aylin(p1,p2);">cluster_aylin</a>(p1,p2);
0623     
0624     nclusters = size(c,2); 
0625     
0626     total_classified = 0
0627     
0628     nsegments = size(matrix2classify,1);
0629             
0630     classification = zeros(nsegments,1);
0631     
0632     <span class="keyword">for</span> i = 1:nclusters
0633         <span class="keyword">for</span> j = 1:length(c(i).c)
0634             classification(c(i).c(j)) = i;
0635         <span class="keyword">end</span>
0636     <span class="keyword">end</span>
0637 ;
0638 
0639 
0640 <span class="comment">% --- Executes on button press in HierarchicalRadioButton.</span>
0641 <a name="_sub28" href="#_subfunctions" class="code">function HierarchicalRadioButton_Callback(hObject, eventdata, handles)</a>
0642 <span class="comment">% hObject    handle to HierarchicalRadioButton (see GCBO)</span>
0643 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0644 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0645 
0646 <span class="comment">% Hint: get(hObject,'Value') returns toggle state of HierarchicalRadioButton</span>
0647 
0648 selected = get(hObject,<span class="string">'Value'</span>);
0649 
0650 <span class="keyword">if</span> selected
0651     handles.cluster_method = <span class="string">'hierarchical'</span>;
0652 <span class="keyword">end</span>
0653 
0654 guidata(gcbo,handles);
0655 
0656 
0657 <span class="comment">% --- Executes on button press in KmeansRadioButton.</span>
0658 <a name="_sub29" href="#_subfunctions" class="code">function KmeansRadioButton_Callback(hObject, eventdata, handles)</a>
0659 <span class="comment">% hObject    handle to KmeansRadioButton (see GCBO)</span>
0660 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0661 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0662 
0663 <span class="comment">% Hint: get(hObject,'Value') returns toggle state of KmeansRadioButton</span>
0664 
0665 selected = get(hObject,<span class="string">'Value'</span>);
0666 
0667 <span class="keyword">if</span> selected
0668     handles.cluster_method = <span class="string">'kmeans'</span>;
0669 <span class="keyword">end</span>
0670 
0671 guidata(gcbo,handles);
0672 
0673 
0674 
0675 
0676 <a name="_sub30" href="#_subfunctions" class="code">function SegLenWeight_Callback(hObject, eventdata, handles)</a>
0677 <span class="comment">% hObject    handle to SegLenWeight (see GCBO)</span>
0678 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0679 <span class="comment">% handles    structure with handles and user data (see GUIDATA)</span>
0680 
0681 <span class="comment">% Hints: get(hObject,'String') returns contents of SegLenWeight as text</span>
0682 <span class="comment">%        str2double(get(hObject,'String')) returns contents of SegLenWeight as a double</span>
0683 
0684 
0685 <span class="comment">% --- Executes during object creation, after setting all properties.</span>
0686 <a name="_sub31" href="#_subfunctions" class="code">function SegLenWeight_CreateFcn(hObject, eventdata, handles)</a>
0687 <span class="comment">% hObject    handle to SegLenWeight (see GCBO)</span>
0688 <span class="comment">% eventdata  reserved - to be defined in a future version of MATLAB</span>
0689 <span class="comment">% handles    empty - handles not created until after all CreateFcns called</span>
0690 
0691 <span class="comment">% Hint: edit controls usually have a white background on Windows.</span>
0692 <span class="comment">%       See ISPC and COMPUTER.</span>
0693 <span class="keyword">if</span> ispc
0694     set(hObject,<span class="string">'BackgroundColor'</span>,<span class="string">'white'</span>);
0695 <span class="keyword">else</span>
0696     set(hObject,<span class="string">'BackgroundColor'</span>,get(0,<span class="string">'defaultUicontrolBackgroundColor'</span>));
0697 <span class="keyword">end</span>
0698 
0699</pre></div>
<hr><address>Generated on Fri 12-Aug-2011 11:36:15 by <strong><a href="http://www.artefact.tk/software/matlab/m2html/" target="_parent">m2html</a></strong> &copy; 2005</address>
</body>
</html>