SVC
#include <Skigen/SVM>
template <typename Scalar = double>
class Skigen::SVC(C=1.0, kernel=Kernel::RBF, degree=3, gamma=0, coef0=0, probability=false, tol=1e-3, max_passes=50, random_state=std::nullopt)
C-Support Vector Classification with kernels.
Mirrors sklearn.svm.SVC for the binary case (multiclass via one-vs-one as in libsvm).
When probability=true, Platt scaling is fitted via internal 5-fold cross-validation on the decision function scores, providing calibrated posterior probabilities through predict_proba().
Attributes:
-
C : Scalar
-
kernel : Kernel
-
gamma : Scalar
-
classes : const Eigen::VectorXi
-
n_classes : int
-
support : const std::vector< Eigen::Index > &
-
n_support : int
Methods
SKIGEN_PARAMS()
fit(X, y)
Fit from a sparse design matrix (densifies internally).
predict(X)
predict(X)
Predict from a sparse design matrix (densifies internally).
decision_function(X)
Raw decision function — sum_s alpha_s y_s K(x, sv_s) + b.
predict_proba(X)
Calibrated posterior probabilities via Platt scaling.
Requires probability=true at construction time. Returns a (n_samples, 2) matrix with columns [P(y=class_0), P(y=class_1)].