def test_when_a_step_does_not_improve_best_score(self):
# type: () -> None
params = {} # type: Dict
valid_data = np.zeros((10, 10))
valid_sets = lgb.Dataset(valid_data)
with warnings.catch_warnings():
warnings.simplefilter("ignore", category=DeprecationWarning)
tuner = LightGBMTuner(params, None, valid_sets=valid_sets)
assert not tuner.higher_is_better()
with mock.patch("lightgbm.train"), mock.patch.object(
BaseTuner, "_get_booster_best_score", return_value=0.9
):
tuner.tune_feature_fraction()
assert "feature_fraction" in tuner.best_params
assert tuner.best_score == 0.9
# Assume that tuning `num_leaves` doesn't improve the `best_score`.
with mock.patch("lightgbm.train"), mock.patch.object(
BaseTuner, "_get_booster_best_score", return_value=1.1
):
tuner.tune_num_leaves()
# `num_leaves` should be same as default.
assert tuner.best_params["num_leaves"] == 31
assert tuner.best_score == 0.9
def test_when_a_step_does_not_improve_best_score(self):
# type: () -> None
params = {} # type: Dict
valid_data = np.zeros((10, 10))
valid_sets = lgb.Dataset(valid_data)
tuner = LightGBMTuner(params, None, valid_sets=valid_sets)
assert not tuner.higher_is_better()
with mock.patch("lightgbm.train"), mock.patch.object(
BaseTuner, "_get_booster_best_score", return_value=0.9
):
tuner.tune_feature_fraction()
assert "feature_fraction" in tuner.best_params
assert tuner.best_score == 0.9
# Assume that tuning `num_leaves` doesn't improve the `best_score`.
with mock.patch("lightgbm.train"), mock.patch.object(
BaseTuner, "_get_booster_best_score", return_value=1.1
):
tuner.tune_num_leaves()
def test_when_a_step_does_not_improve_best_score(self):
# type: () -> None
params = {} # type: Dict
valid_data = np.zeros((10, 10))
valid_sets = lgb.Dataset(valid_data)
tuner = LightGBMTuner(params, None, valid_sets=valid_sets)
assert not tuner.higher_is_better()
objective_class_name = 'optuna.integration.lightgbm_tuner.optimize.OptunaObjective'
with mock.patch(objective_class_name) as objective_mock,\
mock.patch('optuna.study.Study') as study_mock,\
mock.patch('optuna.trial.Trial') as trial_mock:
fake_objective = mock.MagicMock(spec=OptunaObjective)
fake_objective.report = []
fake_objective.best_booster = None
objective_mock.return_value = fake_objective
fake_study = mock.MagicMock(spec=Study)
fake_study._storage = mock.MagicMock()
fake_study.best_value = 0.9
study_mock.return_value = fake_study
fake_trial = mock.MagicMock(spec=Trial)
fake_trial.best_params = {
'feature_fraction': 0.2,
}
trial_mock.return_value = fake_trial
tuner.tune_feature_fraction()
fake_study.optimize.assert_called()
assert 'feature_fraction' in tuner.best_params
assert tuner.best_score == 0.9
with mock.patch(objective_class_name) as objective_mock,\
mock.patch('optuna.study.Study') as study_mock,\
mock.patch('optuna.trial.Trial') as trial_mock:
fake_objective = mock.MagicMock(spec=OptunaObjective)
fake_objective.report = []
fake_objective.best_booster = None
objective_mock.return_value = fake_objective
# Assume that tuning `num_leaves` doesn't improve the `best_score`.
fake_study = mock.MagicMock(spec=Study)
fake_study._storage = mock.MagicMock()
fake_study.best_value = 1.1
study_mock.return_value = fake_study
fake_trial = mock.MagicMock(spec=Trial)
fake_trial.best_params = {
'num_leaves': 128,
}
trial_mock.return_value = fake_trial
tuner.tune_num_leaves()
fake_study.optimize.assert_called()
# `num_leaves` should not be same as default.
assert tuner.best_params['num_leaves'] == 31
assert tuner.best_score == 0.9
