svdd_fmin.m

模式识别工具箱,希望对大家有用!

%function [sigma,alf,R2,I] = svdd_fmin(x2,labx,fracsv,fracerr);
%
% See for more information  svdd.
% This algorithm is optimized to find the optimal sigma for one given
% fracsv, using the function fminbnd from the optimization toolbox. (So
% the optimization routine fminbnd should be available.)
%
% See also svdd
% 19-4-2000 Dxd

% Copyright: D.M.J. Tax, R.P.W. Duin, [e-mail]
% Faculty of Applied Physics, Delft University of Technology
% P.O. Box 5046, 2600 GA Delft, The Netherlands


function [sigma,alf,R2,I,Dx] = svdd_fmin(x2,labx,fracsv,fracerr)

if (max(size(fracsv))>1)
  error('gsvdd_fmin: please supply one fracsv');
end
maxsigma = sqrt(max(max(x2)));

options = optimset('Display','off','TolX',0.01);
sigma = fminbnd('f_svs',0,maxsigma,options,x2,labx,fracerr,fracsv);
[svdd_error,alf,R2,Dx,I] = f_svs(sigma,x2,labx,fracerr);

return