SVR
Kernel epsilon-support-vector regression using the same SMO solver and kernels as SVC.
Algorithm
Fits the epsilon-insensitive dual with a kernel of choice; only samples outside the epsilon tube become support vectors.
Constructor
Skigen::SVR<Scalar> model(Scalar C = 1.0, Kernel = RBF, Scalar epsilon = 0.1, ...);
Parameters
| Parameter | Default | Description |
|---|---|---|
C | 1.0 | Penalty strength. |
kernel | RBF | Kernel policy. |
epsilon | 0.1 | Insensitive-tube width. |
gamma | scale | Kernel coefficient. |
Methods
| Method | Description |
|---|---|
fit(X, y) | Solve the epsilon-SVR dual. |
predict(X) | Real-valued predictions. |
score(X, y) | R². |
Fitted Attributes
| Accessor | Description |
|---|---|
n_support() | Number of support vectors. |
Example
Skigen::SVR<double> svr;
svr.fit(X, y);
auto preds = svr.predict(X_test);
Verified against scikit-learn
This estimator is checked by the parity suite. See the generator tests/parity/generate_svm_reference.py and the reference fixtures in tests/parity/data/svr/, exercised by tests/parity/parity_svm.cpp.
API Reference
For full signatures see the SVR API Reference.