www.gusucode.com > 支持向量机工具箱 - LIBSVM OSU_SVM LS_SVM源码程序 > 支持向量机工具箱 - LIBSVM OSU_SVM LS_SVM\stprtool\svm\svmclass2.m
% 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