def test_sample_weight(queue):
X = np.array([[-2, 0], [-1, -1], [0, -2], [0, 2], [1, 1], [2, 2]])
y = np.array([1, 1, 1, 2, 2, 2])
clf = NuSVC(kernel='linear')
clf.fit(X, y, sample_weight=[1] * 6, queue=queue)
assert_array_almost_equal(clf.intercept_, [0.0])
def test_class_weight(queue):
X = np.array([[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]])
y = np.array([1, 1, 1, 2, 2, 2])
clf = NuSVC(class_weight={1: 0.1})
clf.fit(X, y, queue=queue)
assert_array_almost_equal(clf.predict(X, queue=queue), [2] * 6)
def test_decision_function_shape(queue):
X, y = make_blobs(n_samples=80, centers=5, random_state=0)
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)
# check shape of ovo_decition_function=True
clf = NuSVC(kernel='linear',
decision_function_shape='ovo').fit(X_train, y_train, queue=queue)
dec = clf.decision_function(X_train, queue=queue)
assert dec.shape == (len(X_train), 10)
def test_decision_function(queue):
X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]]
Y = [1, 1, 1, 2, 2, 2]
clf = NuSVC(kernel='rbf', gamma=1, decision_function_shape='ovo')
clf.fit(X, Y, queue=queue)
rbfs = rbf_kernel(X, clf.support_vectors_, gamma=clf.gamma)
dec = np.dot(rbfs, clf.dual_coef_.T) + clf.intercept_
assert_array_almost_equal(dec.ravel(), clf.decision_function(X, queue=queue))
def _test_libsvm_parameters(queue, array_constr, dtype):
X = array_constr([[-2, -1], [-1, -1], [-1, -2],
[1, 1], [1, 2], [2, 1]], dtype=dtype)
y = array_constr([1, 1, 1, 2, 2, 2], dtype=dtype)
clf = NuSVC(kernel='linear').fit(X, y, queue=queue)
assert_array_almost_equal(
clf.dual_coef_, [[-0.04761905, -0.0952381, 0.0952381, 0.04761905]])
assert_array_equal(clf.support_, [0, 1, 3, 4])
assert_array_equal(clf.support_vectors_, X[clf.support_])
assert_array_equal(clf.intercept_, [0.])
assert_array_equal(clf.predict(X, queue=queue), y)
def test_pickle(queue):
iris = datasets.load_iris()
clf = NuSVC(kernel='linear').fit(iris.data, iris.target, queue=queue)
expected = clf.decision_function(iris.data, queue=queue)
import pickle
dump = pickle.dumps(clf)
clf2 = pickle.loads(dump)
assert type(clf2) == clf.__class__
result = clf2.decision_function(iris.data, queue=queue)
assert_array_equal(expected, result)
def _test_cancer_poly_compare_with_sklearn(queue, params):
cancer = datasets.load_breast_cancer()
clf = NuSVC(kernel='poly', **params)
clf.fit(cancer.data, cancer.target, queue=queue)
result = clf.score(cancer.data, cancer.target, queue=queue)
clf = SklearnNuSVC(kernel='poly', **params)
clf.fit(cancer.data, cancer.target)
expected = clf.score(cancer.data, cancer.target)
assert result > 0.5
assert abs(result - expected) < 1e-4
def _test_cancer_linear_compare_with_sklearn(queue, nu):
cancer = datasets.load_breast_cancer()
clf = NuSVC(kernel='linear', nu=nu)
clf.fit(cancer.data, cancer.target, queue=queue)
result = clf.score(cancer.data, cancer.target, queue=queue)
clf = SklearnNuSVC(kernel='linear', nu=nu)
clf.fit(cancer.data, cancer.target)
expected = clf.score(cancer.data, cancer.target)
assert result > 0.5
assert abs(result - expected) < 1e-3
def _test_cancer_rbf_compare_with_sklearn(queue, nu, gamma):
cancer = datasets.load_breast_cancer()
clf = NuSVC(kernel='rbf', gamma=gamma, nu=nu)
clf.fit(cancer.data, cancer.target, queue=queue)
result = clf.score(cancer.data, cancer.target, queue=queue)
clf = SklearnNuSVC(kernel='rbf', gamma=gamma, nu=nu)
clf.fit(cancer.data, cancer.target)
expected = clf.score(cancer.data, cancer.target)
assert result > 0.4
assert abs(result - expected) < 1e-4
def test_iris(queue):
iris = datasets.load_iris()
clf = NuSVC(kernel='linear').fit(iris.data, iris.target, queue=queue)
assert clf.score(iris.data, iris.target, queue=queue) > 0.9
assert_array_equal(clf.classes_, np.sort(clf.classes_))