%RBSVC Trainable automatic radial basis Support Vector Classifier
%
% [W,KERNEL,NU,C] = RBSVC(A)
% [W,KERNEL,NU,C] = A*RBSVC
%
% INPUT
% A Dataset
%
% OUTPUT
% W Mapping: Radial Basis Support Vector Classifier
% KERNEL Untrained mapping, representing the optimised kernel
% NU Resulting value for NU from NUSVC (W = NUSVC(A,KERNEL,C)
% C Resulting value for C (W = SVC(A,KERNEL,C)
%
% DESCRIPTION
% This routine computes a classifier by NUSVC using a radial basis kernel
% with an optimised standard deviation by REGOPTC. The resulting classifier
% W is identical to NUSVC(A,KERNEL,NU). As the kernel optimisation is based
% on internal cross-validation the dataset A should be sufficiently large.
% Moreover it is very time-consuming as the kernel optimisation needs
% about 100 calls to SVC.
%
% SEE ALSO (<a href="http://37steps.com/prtools">PRTools Guide</a>)
% MAPPINGS, DATASETS, PROXM, SVC, NUSVC, REGOPTC
% Copyright: R.P.W. Duin, r.p.w.duin@37steps.com
% Faculty EWI, Delft University of Technology
% P.O. Box 5031, 2600 GA Delft, The Netherlands
function [w,kernel,nu,c] = rbsvc(a,sig)
if nargin < 2 | isempty(sig)
sig = NaN;
end
if nargin < 1 | isempty(a)
w = prmapping(mfilename,{sig});
else
islabtype(a,'crisp');
isvaldfile(a,2,2); % at least 1 object per class, 2 classes
a = testdatasize(a,'objects');
c = getsize(a,3);
if (c > 2)
% Compute c classifiers: each class against all others.
w = mclassc(a,prmapping(mfilename,{sig}));
else
if isnan(sig) % optimise sigma
% find upper bound
d = sqrt(+distm(a));
sigmax = min(max(d)); % max: smallest furthest neighbor distance
% find lower bound
d = d + 1e100*eye(size(a,1));
sigmin = max(min(d)); % min: largest nearest neighbor distance
% call optimiser
defs = {1};
parmin_max = [sigmin,sigmax];
[w,kernel,nu,c] = regoptc(a,mfilename,{sig},defs,[1],parmin_max,testc([],'soft'));
else % kernel is given
kernel = proxm([],'r',sig);
[w,J,nu,c] = nusvc(a,kernel);
end
end
end
w = setname(w,'RB-SVM');
return