Skip to main content

OneClassSVM

#include <Skigen/SVM>

template <typename Scalar = double>
class Skigen::OneClassSVM(kernel=Kernel::RBF, degree=3, gamma=0, coef0=0, nu=0.5)

One-class SVM — placeholder.

Mirrors sklearn.svm.OneClassSVM constructor signature so user code can be written against the API surface, but fit() throws: the one-class formulation requires the same nu-SVM SMO variant that backs NuSVC/NuSVR, which is not implemented.



Methods

SKIGEN_PARAMS()


predict()