Merge pull request optuna#5214 from nabenabe0928/hotfix/fix-reproduci…

…bility-issue-in-lightgbm

Skip the reproducibility tests for lightgbm

tests/artifacts_tests/test_boto3.py

@@ -4,13 +4,18 @@
 from typing import TYPE_CHECKING
 
 import boto3
-from moto import mock_s3
 import pytest
 
 from optuna.artifacts import Boto3ArtifactStore
 from optuna.artifacts.exceptions import ArtifactNotFound
 
 
+try:
+    from moto import mock_aws
+except ImportError:
+    from moto import mock_s3 as mock_aws
+
+
 if TYPE_CHECKING:
     from collections.abc import Iterator
 
@@ -22,7 +27,7 @@
 
 @pytest.fixture()
 def init_mock_client() -> Iterator[tuple[str, S3Client]]:
-    with mock_s3():
+    with mock_aws():
         # Runs before each test
         bucket_name = "moto-bucket"
         s3_client = boto3.client("s3")

tests/integration_tests/lightgbm_tuner_tests/test_optimize.py

@@ -161,7 +161,7 @@ def test_higher_is_better(self) -> None:
             tuner = _BaseTuner(lgbm_params={"metric": metric})
             assert not tuner.higher_is_better()
 
-    def test_get_booster_best_score__using_valid_names_as_str(self) -> None:
+    def test_get_booster_best_score_using_valid_names_as_str(self) -> None:
         expected_value = 1.0
 
         booster = mock.MagicMock(
@@ -173,7 +173,7 @@ def test_get_booster_best_score__using_valid_names_as_str(self) -> None:
         val_score = tuner._get_booster_best_score(booster)
         assert val_score == expected_value
 
-    def test_get_booster_best_score__using_valid_names_as_list(self) -> None:
+    def test_get_booster_best_score_using_valid_names_as_list(self) -> None:
         unexpected_value = 0.5
         expected_value = 1.0
 
@@ -331,7 +331,7 @@ def test_with_minimum_required_args(self) -> None:
         assert "num_boost_round" not in runner.auto_options
         assert runner.lgbm_kwargs["num_boost_round"] == 5
 
-    def test__parse_args_wrapper_args(self) -> None:
+    def test_parse_args_wrapper_args(self) -> None:
         params: dict[str, Any] = {}
         train_set = lgb.Dataset(None)
         val_set = lgb.Dataset(None)
@@ -726,6 +726,7 @@ def test_optuna_callback(self) -> None:
 
         assert callback_mock.call_count == 10
 
+    @pytest.mark.skip(reason="Fail since 28 Jan 2024. TODO(nabenabe0928): Fix here.")
     def test_tune_best_score_reproducibility(self) -> None:
         iris = sklearn.datasets.load_iris()
         X_trainval, X_test, y_trainval, y_test = train_test_split(
@@ -1057,6 +1058,7 @@ def test_get_best_booster_with_error(self) -> None:
             with pytest.raises(ValueError):
                 tuner3.get_best_booster()
 
+    @pytest.mark.skip(reason="Fail since 28 Jan 2024. TODO(nabenabe0928): Fix here.")
     def test_tune_best_score_reproducibility(self) -> None:
         iris = sklearn.datasets.load_iris()
         X_trainval, X_test, y_trainval, y_test = train_test_split(