formatting

Author: SvenKlaassen

doubleml/did/tests/test_did_binary_tune_ml_models.py

@@ -39,7 +39,7 @@ def test_doubleml_did_binary_optuna_tune(sampler_name, optuna_sampler, score):
 
     g_value, t_value_pre, t_value_eval = _select_binary_periods(panel_data)
 
-    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1) # underfit
+    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1)  # underfit
     ml_m = DecisionTreeClassifier(random_state=654)
     dml_did_binary = DoubleMLDIDBinary(
         obj_dml_data=panel_data,

doubleml/did/tests/test_did_cs_binary_tune_ml_models.py

@@ -37,7 +37,7 @@ def test_doubleml_did_cs_binary_optuna_tune(sampler_name, optuna_sampler, score)
     print(df_panel.head())
     g_value, t_value_pre, t_value_eval = _select_binary_periods(panel_data)
 
-    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1) # underfit
+    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1)  # underfit
     ml_m = DecisionTreeClassifier(random_state=654)
 
     dml_did_cs_binary = DoubleMLDIDCSBinary(
@@ -68,4 +68,4 @@ def test_doubleml_did_cs_binary_optuna_tune(sampler_name, optuna_sampler, score)
         _assert_tree_params(tuned_params)
 
         # ensure tuning improved RMSE or LogLoss
-        assert tuned_score[learner_name] < untuned_score[learner_name]
+        assert tuned_score[learner_name] < untuned_score[learner_name]

doubleml/did/tests/test_did_cs_tune_ml_models.py

@@ -25,7 +25,7 @@ def test_doubleml_did_cs_optuna_tune(sampler_name, optuna_sampler, score):
         return_type="DoubleMLDIDData",
     )
 
-    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1) # underfit
+    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1)  # underfit
     if score == "observational":
         ml_m = DecisionTreeClassifier(random_state=654)
         dml_did_cs = dml.DoubleMLDIDCS(dml_data, ml_g, ml_m, score=score, n_folds=5)

doubleml/did/tests/test_did_tune_ml_models.py

@@ -22,7 +22,7 @@ def test_doubleml_did_optuna_tune(sampler_name, optuna_sampler, score):
     np.random.seed(3150)
     dml_data = make_did_SZ2020(n_obs=500, dgp_type=4, return_type="DoubleMLDIDData")
 
-    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1) # underfit
+    ml_g = DecisionTreeRegressor(random_state=321, max_depth=1)  # underfit
     if score == "observational":
         ml_m = DecisionTreeClassifier(random_state=654)
         dml_did = dml.DoubleMLDID(dml_data, ml_g, ml_m, score=score, n_folds=5)