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 = SVC(class_weight={1: 0.1})
clf.fit(X, y, queue=queue)
assert_array_almost_equal(clf.predict(X, queue=queue), [2] * 6)
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 = SVC(kernel='linear')
clf.fit(X, y, sample_weight=[1] * 6, queue=queue)
assert_array_almost_equal(clf.intercept_, [0.0])
def test_decision_function_shape():
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 = SVC(kernel='linear',
decision_function_shape='ovo').fit(X_train, y_train)
dec = clf.decision_function(X_train)
assert dec.shape == (len(X_train), 10)
def test_decision_function(queue):
X = np.array([[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]], dtype=np.float32)
Y = np.array([1, 1, 1, 2, 2, 2], dtype=np.float32)
clf = SVC(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_decision_function():
X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]]
Y = [1, 1, 1, 2, 2, 2]
clf = SVC(kernel='rbf', gamma=1, decision_function_shape='ovo')
clf.fit(X, Y)
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))
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 = SVC(kernel='linear').fit(X, y, queue=queue)
assert_array_equal(clf.dual_coef_, [[-0.25, .25]])
assert_array_equal(clf.support_, [1, 3])
assert_array_equal(clf.support_vectors_, (X[1], X[3]))
assert_array_equal(clf.intercept_, [0.])
assert_array_equal(clf.predict(X), y)
def test_svc_sigmoid(queue, dtype):
X_train = np.array([[-1, 2], [0, 0], [2, -1],
[+1, +1], [+1, +2], [+2, +1]], dtype=dtype)
X_test = np.array([[0, 2], [0.5, 0.5],
[0.3, 0.1], [2, 0], [-1, -1]], dtype=dtype)
y_train = np.array([1, 1, 1, 2, 2, 2], dtype=dtype)
svc = SVC(kernel='sigmoid').fit(X_train, y_train, queue=queue)
assert_array_equal(svc.dual_coef_, [[-1, -1, -1, 1, 1, 1]])
assert_array_equal(svc.support_, [0, 1, 2, 3, 4, 5])
assert_array_equal(svc.predict(X_test, queue=queue), [2, 2, 1, 2, 1])
def test_pickle(queue):
iris = datasets.load_iris()
clf = SVC(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_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 = SVC(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)
with pytest.raises(ValueError, match="must be either 'ovr' or 'ovo'"):
SVC(decision_function_shape='bad').fit(X_train, y_train, queue=queue)
def _test_binary_dataset(queue, kernel):
X, y = make_classification(n_samples=80, n_features=20, n_classes=2, random_state=0)
sparse_X = sp.csr_matrix(X)
dataset = sparse_X, y, sparse_X
clf = SVC(kernel=kernel)
check_svm_model_equal(queue, clf, *dataset)
def _test_simple_dataset(queue, kernel):
X = np.array([[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]])
sparse_X = sp.lil_matrix(X)
Y = [1, 1, 1, 2, 2, 2]
X2 = np.array([[-1, -1], [2, 2], [3, 2]])
sparse_X2 = sp.dok_matrix(X2)
dataset = sparse_X, Y, sparse_X2
clf = SVC(kernel=kernel, gamma=1)
check_svm_model_equal(queue, clf, *dataset)
def _test_iris(queue, kernel):
iris = datasets.load_iris()
rng = np.random.RandomState(0)
perm = rng.permutation(iris.target.size)
iris.data = iris.data[perm]
iris.target = iris.target[perm]
sparse_iris_data = sp.csr_matrix(iris.data)
dataset = sparse_iris_data, iris.target, sparse_iris_data
clf = SVC(kernel=kernel)
check_svm_model_equal(queue, clf, *dataset, decimal=2)
def test_estimator():
def dummy(*args, **kwargs):
pass
md = sklearn.utils.estimator_checks
saved = _replace_and_save(md, [
'check_sample_weights_invariance', # Max absolute difference: 0.0008
'check_estimators_fit_returns_self', # ValueError: empty metadata
'check_classifiers_train', # assert y_pred.shape == (n_samples,)
'check_estimators_unfitted', # Call 'fit' with appropriate arguments
], dummy)
check_estimator(SVC())
_restore_from_saved(md, saved)
def test_sparse_realdata(queue):
data = np.array([0.03771744, 0.1003567, 0.01174647, 0.027069])
indices = np.array([6, 5, 35, 31])
indptr = np.array(
[0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 4, 4, 4])
X = sp.csr_matrix((data, indices, indptr))
y = np.array(
[1., 0., 2., 2., 1., 1., 1., 2., 2., 0., 1., 2., 2.,
0., 2., 0., 3., 0., 3., 0., 1., 1., 3., 2., 3., 2.,
0., 3., 1., 0., 2., 1., 2., 0., 1., 0., 2., 3., 1.,
3., 0., 1., 0., 0., 2., 0., 1., 2., 2., 2., 3., 2.,
0., 3., 2., 1., 2., 3., 2., 2., 0., 1., 0., 1., 2.,
3., 0., 0., 2., 2., 1., 3., 1., 1., 0., 1., 2., 1.,
1., 3.])
clf = SVC(kernel='linear').fit(X.toarray(), y, queue=queue)
sp_clf = SVC(kernel='linear').fit(X, y, queue=queue)
assert_array_equal(clf.support_vectors_, sp_clf.support_vectors_.toarray())
assert_array_equal(clf.dual_coef_, sp_clf.dual_coef_.toarray())
def test_estimator():
def dummy(*args, **kwargs):
pass
md = sklearn.utils.estimator_checks
saved = _replace_and_save(
md,
[
'check_sample_weights_invariance', # Max absolute difference: 0.0008
'check_estimators_fit_returns_self', # ValueError: empty metadata
'check_estimators_pickle', # NotImplementedError
'check_classifiers_predictions', # Cannot cast ufunc 'multiply'
'check_classifiers_train', # assert y_pred.shape == (n_samples,)
'check_classifiers_regression_target', # Did not raise ValueError
'check_supervised_y_2d', # expected 1 DataConversionWarning
'check_estimators_unfitted', # Call 'fit' with appropriate arguments
],
dummy)
check_estimator(SVC())
_restore_from_saved(md, saved)
def test_iris(queue):
iris = datasets.load_iris()
clf = SVC(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_))