# Cross validating one-against-all SVMs

I'm training a bunch of SVM models to do one-against-all multiclass classification (a test instance is classified as the class which produces the largest positive SVM response).

What's the best way to do cross-validation for selection of the regularization parameter?

Should I do cross-validation separately for each SVM model I train, potentially getting a different regularization parameter for each of the models? Or should I cross-validate as a group, where I try a particular regularization parameter across all models?

Also, what is a good metric for cross-validation? Accuracy? Precision? F-Measure?

• @Articuno: The e1071 package in R provides an interface to libsvm as well as functions for one-against-all multiclass SVMs based on libsvm. For example model <- svm(Species ~ ., data = iris, cost = 100, gamma = 1) – Zach Dec 5 '11 at 19:40