資源簡介
Feature scaling for kernel Fisher discriminant analysis using leave-one-out cross validation. FS-KFDA is a package for implementing feature scaling for kernel fisher discriminant analysis.-Feature scaling for kernel Fisher discrim inant analysis using leave-one-out cross vali dation. FS-KFDA is a package for implementing f eature scaling for kernel fisher discriminant analysis.

代碼片段和文件信息
clear
load?blfsegment;
load?segmentindex;
trainnum?=?500;
starttime?=?cputime;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
for?k?=?1:10
????%?Tanslate?the?data
????index1????????????=??segmentindex(k1:trainnum);
????index2????????????=??segmentindex(k1+trainnum:end);
????trainsamples??????=??segmentsamples(index1:);
????testsamples???????=??segmentsamples(index2:);
????[trainsamplesAB]=??scaletrain(trainsamples);
????testsamples???????=??scaletest(testsamplesAB);
????trainlabels???????=??segmentlabels(index1);
????testlabels????????=??segmentlabels(index2);
?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%???calculate?the?correct?rate???????
??????
????options?=?optimset(‘GradObj‘‘on‘);
????options?=?optimset(options‘LargeScale‘‘off‘);
????options?=?optimset(options‘DerivativeCheck‘‘off‘);
????options?=?optimset(options‘Display‘‘iter‘);
????options?=?optimset(options‘MaxIter‘50);
????options?=?optimset(options‘TolFun‘1e-4);
????options?=?optimset(options‘TolX‘1e-4);
????options?=?optimset(options‘LineSearchType‘‘cubicpoly‘);
????
????D?=?size(trainsamples2);
????x?=?log(1/D)*ones(size(trainsamples2)1);
????[xfxc]?????=?fminunc(‘mkfdakernel‘[x;0*ones(11)]optionstrainsamplestrainlabels);
????[abloo(k)]?=?mkfdakernel(xtrainsamplestrainlabels);
????fprintf(‘%f\n‘loo(k));
????result???????=?mkfdapred(x?trainsamples?trainlabelstestsamples);
????testerror(k)?=?1?-?mean(result?==?testlabels);
?????
?????fprintf(‘meanerror?=?%f\n‘mean(testerror));
?????fprintf(‘Iteration?=?%d\n‘k);
?end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%???print?the?output
endtime???=?cputime;
meanerror?=?mean(testerror);
stderror??=?std(testerror);
fprintf(‘meanerror?=?%f\n‘meanerror);
fprintf(‘stderror??=?%f\n‘stderror);
fprintf(‘time??????=?%f\n‘endtime-starttime);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
?屬性????????????大小?????日期????時間???名稱
-----------?---------??----------?-----??----
?????文件???????2501??2006-10-10?20:18??fs-kfda?1.0\bench_segment.m
?????文件?????335224??2006-10-10?17:39??fs-kfda?1.0\blfsegment.mat
?????文件????????712??2006-10-10?20:26??fs-kfda?1.0\content.m
?????文件????????568??2006-01-16?11:05??fs-kfda?1.0\evalkernel.m
?????文件???????1749??2005-07-08?11:22??fs-kfda?1.0\kfdakernel.m
?????文件???????1186??2005-06-29?11:13??fs-kfda?1.0\kfdapred.m
?????文件???????2127??2005-04-08?20:18??fs-kfda?1.0\mkfdakernel.m
?????文件???????1525??2005-04-06?20:36??fs-kfda?1.0\mkfdapred.m
?????文件????????180??2005-01-21?20:18??fs-kfda?1.0\scaletest.m
?????文件????????231??2005-06-11?09:41??fs-kfda?1.0\scaletrain.m
?????文件?????380757??2006-10-10?19:22??fs-kfda?1.0\segmentindex.mat
?????文件????????380??2011-03-07?11:24??fs-kfda?1.0\注釋.txt
?????目錄??????????0??2006-10-10?20:15??fs-kfda?1.0
-----------?---------??----------?-----??----
???????????????727140????????????????????13
評論
共有 條評論