tests: 100% coverage for _nn_runner_base.py

tests/test_runners/test_nn_runner_base.py

@@ -728,3 +728,78 @@ def test_nn_runner_base_grid_search_score_intercept(self):
         score = runner._grid_search_score_intercept(y_true, y_pred)
 
         assert np.isnan(score)
+
+    def test_nn_runner_base_teardown_rename_with_existing_file(self):
+        """Test _NNRunnerBase _tear_down method when correct_filename already exists"""
+        with patch("os.makedirs"):
+            runner = _NNRunnerBase(
+                x_train=np.random.rand(100, 10),
+                y_train=np.random.randint(2, size=100),
+                x_test=np.random.rand(20, 10),
+                y_test=np.random.randint(2, size=20),
+                experiment_name="test_experiment",
+                seed=SEED,
+                iteration_list=[1, 2, 3],
+                grid_search_parameters={"param1": [0.1, 0.2], "param2": [1, 2]},
+                grid_search_scorer_method=skmt.accuracy_score,
+                output_directory="test_output",
+            )
+
+            runner.runner_name = MagicMock(return_value="TestRunner")
+            runner.best_params = {"param1": 0.1, "param2": 1}
+            runner.replay_mode = MagicMock(return_value=False)
+            runner._check_match = MagicMock(return_value=True)  # All files are correct
+
+            # Mock the necessary functions and data
+            filename_root = "test_output/test_experiment/testrunner__test_experiment"
+            path = "test_output/test_experiment"
+            filename_part = "testrunner__test_experiment"
+
+            # Prepare filenames with correct md5 hash
+            correct_md5 = "ABCDEF123456"
+            correct_filename_with_md5 = f"{filename_part}_df_{correct_md5}.p"
+            correct_filename = correct_filename_with_md5.replace(f"__{correct_md5}", "")
+            filenames = [correct_filename_with_md5]
+
+            # Define correct dataframe
+            correct_df = pd.DataFrame([{"param1": "0.1", "param2": "1"}])
+
+            # Helper function to return different data based on filename
+            def mock_pickle_load(file):
+                if correct_md5 in file.name:
+                    return correct_df
+                return pd.DataFrame()
+
+            # Mock open to set the filename attribute
+            mock_file_correct = mock_open(read_data=pk.dumps(correct_df)).return_value
+            mock_file_correct.name = os.path.join(path, correct_filename_with_md5)
+
+            # Side effect for open to return different mock files
+            def open_side_effect(file, mode="rb"):
+                if correct_md5 in file:
+                    return mock_file_correct
+                return mock_open().return_value
+
+            with (
+                patch("mlrose_ky.runners._runner_base._RunnerBase._get_pickle_filename_root", return_value=filename_root),
+                patch("os.path.isdir", return_value=True),
+                patch("os.listdir", return_value=filenames),
+                patch("os.rename") as mock_rename,
+                patch("os.path.exists", return_value=True),  # Mock os.path.exists to return True
+                patch("builtins.open", side_effect=open_side_effect),
+                patch("pickle.load", side_effect=mock_pickle_load),
+                patch.object(runner, "_check_match", return_value=True),
+            ):
+
+                runner._tear_down()
+
+                # Check that os.rename was called twice:
+                # First to backup existing file, then to rename the new correct file
+                correct_file_path = os.path.join(path, correct_filename)
+                correct_file_with_md5_path = os.path.join(path, correct_filename_with_md5)
+                backup_file_path = f"{correct_file_path}.bak"
+
+                expected_calls = [call(correct_file_path, backup_file_path), call(correct_file_with_md5_path, correct_file_path)]
+
+                mock_rename.assert_has_calls(expected_calls)
+                assert mock_rename.call_count == 2