2 results for “svm”
Loss for support vector machines; zero once a point is correctly classified with margin.
Width of the separating margin an SVM maximizes between classes.