def test_feature_importance(X_y_regression):
X, y = X_y_regression
sk_clf = SKElasticNetRegressor(alpha=0.5,
l1_ratio=0.5,
random_state=0,
normalize=False,
max_iter=1000)
sk_clf.fit(X, y)
clf = ElasticNetRegressor()
clf.fit(X, y)
np.testing.assert_almost_equal(sk_clf.coef_, clf.feature_importance, decimal=5)
def test_fit_predict(X_y_regression):
X, y = X_y_regression
sk_clf = SKElasticNetRegressor(alpha=0.5,
l1_ratio=0.5,
random_state=0,
normalize=False,
max_iter=1000)
sk_clf.fit(X, y)
y_pred_sk = sk_clf.predict(X)
clf = ElasticNetRegressor()
fitted = clf.fit(X, y)
assert isinstance(fitted, ElasticNetRegressor)
y_pred = clf.predict(X)
np.testing.assert_almost_equal(y_pred, y_pred_sk, decimal=5)
def test_en_parameters():
clf = ElasticNetRegressor(alpha=0.75, l1_ratio=0.5, random_state=2)
expected_parameters = {
"alpha": 0.75,
"l1_ratio": 0.5,
'max_iter': 1000,
'normalize': False
}
assert clf.parameters == expected_parameters