def test_set_params():
meta = Stacking(
LogisticRegression(),
[('tree', DecisionTreeClassifier(max_depth=1, random_state=0)),
('svm', SVC(probability=True, random_state=0))],
probabilities=True)
assert 2 == len(meta)
meta.set_params(tree__min_samples_split=7, svm__C=0.05)
assert 7 == meta.get_params()["tree__min_samples_split"]
assert 0.05 == meta.get_params()["svm__C"]
assert isinstance(meta.get_params()["meta_estimator"],
LogisticRegression)
assert meta.get_params()["probabilities"]
meta.set_params(meta_estimator=DecisionTreeClassifier(),
probabilities=False)
assert isinstance(meta.get_params()["meta_estimator"],
DecisionTreeClassifier)
assert not meta.get_params()["probabilities"]
p = meta.get_params(deep=False)
assert set(p.keys()) == {
"meta_estimator", "base_estimators", "probabilities"
}
def test_set_params(self):
meta = Stacking(_PredictDummy(),
[('coxph', CoxPHSurvivalAnalysis()),
('svm', FastSurvivalSVM(random_state=0))],
probabilities=False)
meta.set_params(coxph__alpha=1.0, svm__alpha=0.4132)
self.assertEqual(1.0, meta.get_params()["coxph__alpha"])
self.assertEqual(0.4132, meta.get_params()["svm__alpha"])
def test_set_params(self):
meta = Stacking(LogisticRegression(), [('tree', DecisionTreeClassifier(max_depth=1, random_state=0)),
('svm', SVC(probability=True, random_state=0))],
probabilities=True)
self.assertEqual(2, len(meta))
meta.set_params(tree__min_samples_split=7, svm__C=0.05)
self.assertEqual(7, meta.get_params()["tree__min_samples_split"])
self.assertEqual(0.05, meta.get_params()["svm__C"])
self.assertIsInstance(meta.get_params()["meta_estimator"], LogisticRegression)
self.assertTrue(meta.get_params()["probabilities"])
meta.set_params(meta_estimator=DecisionTreeClassifier(), probabilities=False)
self.assertIsInstance(meta.get_params()["meta_estimator"], DecisionTreeClassifier)
self.assertFalse(meta.get_params()["probabilities"])
