def test_pickle():
diabetes = datasets.load_diabetes()
clf = SVR(kernel='rbf', C=10.)
clf.fit(diabetes.data, diabetes.target)
expected = clf.predict(diabetes.data)
import pickle
dump = pickle.dumps(clf)
clf2 = pickle.loads(dump)
assert type(clf2) == clf.__class__
result = clf2.predict(diabetes.data)
assert_array_equal(expected, result)
def _test_diabetes_compare_with_sklearn(kernel):
diabetes = datasets.load_diabetes()
clf_onedal = SVR(kernel=kernel, C=10.)
clf_onedal.fit(diabetes.data, diabetes.target)
result = clf_onedal.score(diabetes.data, diabetes.target)
clf_sklearn = SklearnSVR(kernel=kernel, C=10.)
clf_sklearn.fit(diabetes.data, diabetes.target)
expected = clf_sklearn.score(diabetes.data, diabetes.target)
assert result > expected - 1e-5
assert_allclose(clf_sklearn.intercept_, clf_onedal.intercept_, atol=1e-4)
assert_allclose(clf_sklearn.support_vectors_.shape,
clf_sklearn.support_vectors_.shape)
assert_allclose(clf_sklearn.dual_coef_, clf_onedal.dual_coef_, atol=1e-2)
def _test_boston_poly_compare_with_sklearn(params):
diabetes = datasets.load_boston()
clf = SVR(kernel='poly', **params)
clf.fit(diabetes.data, diabetes.target)
result = clf.score(diabetes.data, diabetes.target)
clf = SklearnSVR(kernel='poly', **params)
clf.fit(diabetes.data, diabetes.target)
expected = clf.score(diabetes.data, diabetes.target)
assert result > 0.5
assert result > expected - 1e-5
def _test_boston_linear_compare_with_sklearn(C):
diabetes = datasets.load_boston()
clf = SVR(kernel='linear', C=C)
clf.fit(diabetes.data, diabetes.target)
result = clf.score(diabetes.data, diabetes.target)
clf = SklearnSVR(kernel='linear', C=C)
clf.fit(diabetes.data, diabetes.target)
expected = clf.score(diabetes.data, diabetes.target)
assert result > 0.5
assert result > expected - 1e-3
def _test_boston_rbf_compare_with_sklearn(C, gamma):
diabetes = datasets.load_boston()
clf = SVR(kernel='rbf', gamma=gamma, C=C)
clf.fit(diabetes.data, diabetes.target)
result = clf.score(diabetes.data, diabetes.target)
clf = SklearnSVR(kernel='rbf', gamma=gamma, C=C)
clf.fit(diabetes.data, diabetes.target)
expected = clf.score(diabetes.data, diabetes.target)
assert result > 0.4
assert result > expected - 1e-5
def _test_diabetes(queue, kernel):
diabetes = datasets.load_diabetes()
sparse_diabetes_data = sp.csr_matrix(diabetes.data)
dataset = sparse_diabetes_data, diabetes.target, sparse_diabetes_data
clf = SVR(kernel=kernel, C=0.1)
check_svm_model_equal(queue, clf, *dataset)
def test_run_to_run_fit():
diabetes = datasets.load_diabetes()
clf_first = SVR(kernel='linear', C=10.)
clf_first.fit(diabetes.data, diabetes.target)
for _ in range(10):
clf = SVR(kernel='linear', C=10.)
clf.fit(diabetes.data, diabetes.target)
assert_allclose(clf_first.intercept_, clf.intercept_)
assert_allclose(clf_first.support_vectors_, clf.support_vectors_)
assert_allclose(clf_first.dual_coef_, clf.dual_coef_)
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: 0.0002
'check_estimators_fit_returns_self', # ???
'check_regressors_train', # Cannot get data type from empty metadata
'check_estimators_unfitted', # expected NotFittedError from sklearn
], dummy)
check_estimator(SVR())
_restore_from_saved(md, saved)
def test_input_format_for_diabetes():
diabetes = datasets.load_diabetes()
c_contiguous_numpy = np.asanyarray(diabetes.data, dtype='float', order='C')
assert c_contiguous_numpy.flags.c_contiguous
assert not c_contiguous_numpy.flags.f_contiguous
assert not c_contiguous_numpy.flags.fnc
clf = SVR(kernel='linear', C=10.)
clf.fit(c_contiguous_numpy, diabetes.target)
dual_c_contiguous_numpy = clf.dual_coef_
res_c_contiguous_numpy = clf.predict(c_contiguous_numpy)
f_contiguous_numpy = np.asanyarray(diabetes.data, dtype='float', order='F')
assert not f_contiguous_numpy.flags.c_contiguous
assert f_contiguous_numpy.flags.f_contiguous
assert f_contiguous_numpy.flags.fnc
clf = SVR(kernel='linear', C=10.)
clf.fit(f_contiguous_numpy, diabetes.target)
dual_f_contiguous_numpy = clf.dual_coef_
res_f_contiguous_numpy = clf.predict(f_contiguous_numpy)
assert_allclose(dual_c_contiguous_numpy, dual_f_contiguous_numpy)
assert_allclose(res_c_contiguous_numpy, res_f_contiguous_numpy)
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: 0.0002
'check_estimators_fit_returns_self', # ???
'check_regressors_train', # Cannot get data type from empty metadata
'check_supervised_y_2d', # need warning, why?
'check_regressors_int', # very bad accuracy
'check_estimators_unfitted', # expected NotFittedError from sklearn
'check_fit_idempotent', # again run fit - error. need to fix
'check_estimators_pickle', # NotImplementedError
],
dummy)
check_estimator(SVR())
_restore_from_saved(md, saved)
def test_predict():
iris = datasets.load_iris()
X = iris.data
y = iris.target
reg = SVR(kernel='linear', C=0.1).fit(X, y)
linear = np.dot(X, reg.support_vectors_.T)
dec = np.dot(linear, reg.dual_coef_.T) + reg.intercept_
assert_array_almost_equal(dec.ravel(), reg.predict(X).ravel())
reg = SVR(kernel='rbf', gamma=1).fit(X, y)
rbfs = rbf_kernel(X, reg.support_vectors_, gamma=reg.gamma)
dec = np.dot(rbfs, reg.dual_coef_.T) + reg.intercept_
assert_array_almost_equal(dec.ravel(), reg.predict(X).ravel())
def test_diabetes_simple():
diabetes = datasets.load_diabetes()
clf = SVR(kernel='linear', C=10.)
clf.fit(diabetes.data, diabetes.target)
assert clf.score(diabetes.data, diabetes.target) > 0.02
def test_sided_sample_weight():
clf = SVR(C=1e-2, kernel='linear')
X = [[-2, 0], [-1, -1], [0, -2], [0, 2], [1, 1], [2, 0]]
Y = [1, 1, 1, 2, 2, 2]
sample_weight = [10., .1, .1, .1, .1, 10]
clf.fit(X, Y, sample_weight=sample_weight)
y_pred = clf.predict([[-1., 1.]])
assert y_pred < 1.5
sample_weight = [1., .1, 10., 10., .1, .1]
clf.fit(X, Y, sample_weight=sample_weight)
y_pred = clf.predict([[-1., 1.]])
assert y_pred > 1.5
sample_weight = [1] * 6
clf.fit(X, Y, sample_weight=sample_weight)
y_pred = clf.predict([[-1., 1.]])
assert y_pred == pytest.approx(1.5)
