" 11 -- L2-regularized L2-loss support vector regression (primal)\n"
" 12 -- L2-regularized L2-loss support vector regression (dual)\n"
" 13 -- L2-regularized L1-loss support vector regression (dual)\n"
+ " for outlier detection\n"
+ " 21 -- one-class support vector machine (dual)\n"
"-c cost : set the parameter C (default 1)\n"
"-p epsilon : set the epsilon in loss function of SVR (default 0.1)\n"
"-n nu : set the parameter nu of one-class SVM (default 0.5)\n"