svmconfmat libsvm Toolbox svmgridlinear
libsvm Toolbox >> libsvm Toolbox > svmgrid

svmgrid

parameter selection tool for C-SVM classification using the RBF (radial basis function) kernel

Calling Sequence

[best_rate,best_c,best_g] = svmgrid(label,instance)
[best_rate,best_c,best_g] = svmgrid(label,instance,log2c,log2g)
[best_rate,best_c,best_g] = svmgrid(label,instance,log2c,log2g,v)
[best_rate,best_c,best_g] = svmgrid(label,instance,log2c,log2g,v,option_string)

Parameters

log2c :

[begin,end,step]

log2g :

[begin,end,step]

v :

fold

option_string:

additional parameters for svmtrain

best_rate :

cross validation accuracy for the best parameter combination

best_c :

best parameter c

best_g :

best parameter gamma

Description

svmgrid is a parameter selection tool for C-SVM classification using the RBF (radial basis function) kernel. 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.

Examples

[label,instance]=libsvmread("demos/heart_scale");
[best_rate,best_c,best_g] = svmgrid(label,instance)
best_g  =    0.0004883
best_c  =    512.
best_rate  =    84.444444

Authors

svmconfmat libsvm Toolbox svmgridlinear