def test_fit_dummy(make_whas500):
whas500 = make_whas500(with_mean=False,
with_std=False,
to_numeric=True)
base_estimators = [
('dummy_0', DummySurvivalRegressor(strategy="mean")),
('dummy_1', DummySurvivalRegressor(strategy="median")),
('dummy_2',
DummySurvivalRegressor(strategy="quantile", quantile=0.1)),
('dummy_3',
DummySurvivalRegressor(strategy="quantile", quantile=0.9)),
('dummy_4',
DummySurvivalRegressor(strategy="quantile", quantile=0.89)),
('dummy_5',
DummySurvivalRegressor(strategy="quantile", quantile=0.91)),
]
meta = EnsembleSelectionRegressor(base_estimators,
n_estimators=1,
min_score=5,
cv=5,
scorer=_score_rmse)
with pytest.raises(
ValueError,
match="no base estimator exceeds min_score, try decreasing it"
):
meta.fit(whas500.x, whas500.y)
def test_invalid_scorer(make_whas500):
whas500 = make_whas500(with_mean=False, with_std=False, to_numeric=True)
base_estimators = [
('dummy_0', DummySurvivalRegressor(strategy="mean")),
('dummy_1', DummySurvivalRegressor(strategy="median")),
]
def _score(est, X_test, y_test, **predict_params):
return 'invalid'
meta = EnsembleSelectionRegressor(base_estimators, n_estimators=1, min_score=5, cv=5,
scorer=_score)
with pytest.raises(ValueError,
match=r"scoring must return a number, got invalid "
r"\(<class 'str'>\) instead\."):
meta.fit(whas500.x, whas500.y)
def test_invalid_scorer(self):
base_estimators = [
('dummy_0', DummySurvivalRegressor(strategy="mean")),
('dummy_1', DummySurvivalRegressor(strategy="median")),
]
def _score(est, X_test, y_test, **predict_params):
return 'invalid'
meta = EnsembleSelectionRegressor(base_estimators, n_estimators=1, min_score=5, cv=5,
scorer=_score)
self.assertRaisesRegex(ValueError,
r"scoring must return a number, got invalid \(<class 'str'>\) instead.",
meta.fit, self.x, self.y)
def test_fit_dummy(self):
base_estimators = [
('dummy_0', DummySurvivalRegressor(strategy="mean")),
('dummy_1', DummySurvivalRegressor(strategy="median")),
('dummy_2', DummySurvivalRegressor(strategy="quantile", quantile=0.1)),
('dummy_3', DummySurvivalRegressor(strategy="quantile", quantile=0.9)),
('dummy_4', DummySurvivalRegressor(strategy="quantile", quantile=0.89)),
('dummy_5', DummySurvivalRegressor(strategy="quantile", quantile=0.91)),
]
meta = EnsembleSelectionRegressor(base_estimators, n_estimators=1, min_score=5, cv=5,
scorer=_score_rmse)
self.assertRaisesRegex(ValueError, "no base estimator exceeds min_score, try decreasing it",
meta.fit, self.x, self.y)
def _create_regression_ensemble():
aft_grid = ParameterGrid({"alpha": 2. ** numpy.arange(-2, 12, 2)})
svm_grid = ParameterGrid({"alpha": 2. ** numpy.arange(-12, 0, 2)})
base_estimators = []
for i, params in enumerate(aft_grid):
model = IPCRidge(max_iter=1000, **params)
base_estimators.append(("aft_%d" % i, model))
for i, params in enumerate(svm_grid):
model = FastSurvivalSVM(rank_ratio=0, fit_intercept=True, max_iter=100,
random_state=1, **params)
base_estimators.append(("svm_%d" % i, model))
cv = KFold(n_splits=4, shuffle=True, random_state=0)
meta = EnsembleSelectionRegressor(base_estimators, n_estimators=0.4,
scorer=_score_rmse,
cv=cv, n_jobs=1)
return meta
