svmclass2.m

来自「this a SVM toolbox,it is very useful for」· M 代码 · 共 33 行

% SVMCLASS2 classifies patterns using SVM decision function.
%  [pred_labels,dec_fun] = svmclass2(tst_data,trn_data,trn_labels,Alpha,bias,ker,arg)
%
% SVMCLASS2 classifies patterns using SVM decision function.
%  This function is faster C-source code version of svmclass.m.
%
%  To make executable file run 'mex svmclass2.c kernel.c'.
%
% Input:
%  tst_data [D x M ] matrix of M testing D-dimensional patterns.
%  trn_data [D x N ] matrix of N training D-dimensional patterns.
%  trn_labels [1 x N] pattern labels (1 for 1st class, 2 for 2nd class ).
%  Alpha [1 x N] Lagrange multipliers of the training patterns.
%  bias [real] bias of the decision function.
%  ker [string] kernel identifier: 'linear', 'poly', 'rbf'.
%  arg [real] kernel argument; for 'linear' is arg=[]; 
%    for 'poly' arg is degree of polynom; for 'rbf' arg is sigma.
%
% Outputs:
%  pred_labels [1xM] predicted labels of testing patterns (1 if fpred>=0 else 2).
%  dec_fun [1 x M] values of decision function for testing patterns.
%
% See also SVMCLASS, SVM.
%

% Statistical Pattern Recognition Toolbox, Vojtech Franc, Vaclav Hlavac
% (c) Czech Technical University Prague, http://cmp.felk.cvut.cz.
%
% Modifications
% 21-Oct-2001, V.Franc
% 30-September-2001, V.Franc, comments.
% 19-September-2001, V.Franc, created