def apply(self):
kernel = ["linear", "poly", "rbf", "sigmoid"][self.kernel_type]
common_args = dict(kernel=kernel,
degree=self.degree,
gamma=self.gamma,
coef0=self.coef0,
tol=self.tol,
preprocessors=self.preprocessors)
if self.svrtype == OWSVMRegression.Epsilon_SVR:
learner = svm.SVRLearner(C=self.epsilon_C,
epsilon=self.epsilon,
**common_args)
else:
learner = svm.NuSVRLearner(C=self.nu_C, nu=self.nu, **common_args)
learner.name = self.learner_name
predictor = None
sv = None
if self.data is not None:
predictor = learner(self.data)
predictor.name = self.learner_name
sv = self.data[predictor.skl_model.support_]
self.send("Learner", learner)
self.send("Predictor", predictor)
self.send("Support vectors", sv)
def test_NuSVR(self):
nrows = 500
ncols = 5
x = np.sort(10 * np.random.rand(nrows, ncols))
y = np.sum(np.sin(x), axis=1).reshape(nrows, 1)
x1, x2 = np.split(x, 2)
y1, y2 = np.split(y, 2)
t = data.Table(x1, y1)
learn = svm.NuSVRLearner(kernel='rbf', C=1e3, gamma=0.1)
clf = learn(t)
z = clf(x2)
self.assertTrue((abs(z.reshape(-1, 1) - y2) < 4.0).all())