parameter selection tool for linear classification
[best_rate,best_c,best_g] = svmgridlinear(label,instance) [best_rate,best_c,best_g] = svmgridlinear(label,instance,log2c) [best_rate,best_c,best_g] = svmgridlinear(label,instance,log2c,s_seq) [best_rate,best_c,best_g] = svmgridlinear(label,instance,log2c,s_seq,v) [best_rate,best_c,best_g] = svmgridlinear(label,instance,log2c,s_seq,v,option_string)
[begin,end,step]
linear kernels (0 - 7), e.g. [0 1 4]
fold
additional parameters for svmtrain
svmgrid is a parameter selection tool for linear classification. It uses cross validation (CV) technique to estimate the accuracy of each parameter combination in the specified range and helps you to decide the best parameters for your problem.
[label,instance]=libsvmread("demos/heart_scale"); best_rate = svmgridlinear(label,instance);