fix unit test for the latest version of scikit-learn

Signed-off-by: Xavier Dupre <[email protected]>

tests/test_sklearn_array_feature_extractor.py

@@ -28,10 +28,10 @@ class TestSklearnArrayFeatureExtractor(unittest.TestCase):
     def test_array_feature_extractor(self):
         data_to_cluster = pd.DataFrame(
             [[1, 2, 3.5, 4.5], [1, 2, 1.7, 4.0], [2, 4, 2.4, 4.3], [2, 4, 2.5, 4.0]],
-            columns=[1, 2, 3, 4],
+            columns=["x1", "x2", "x3", "x4"],
         )
-        cat_attributes_clustering = [1, 2]
-        num_attributes_clustering = [3, 4]  # this is of length 12 in reality
+        cat_attributes_clustering = ["x1", "x2"]
+        num_attributes_clustering = ["x3", "x4"]  # this is of length 12 in reality
         gmm = GaussianMixture(n_components=2, random_state=1)
         ohe_cat = [
             OneHotEncoder(categories="auto", sparse_output=False, drop=None)
@@ -44,11 +44,7 @@ def test_array_feature_extractor(self):
             ],
             remainder="passthrough",
         )
-        onehotencoding_pipeline = Pipeline(
-            [
-                ("columnTransformer", ct_cat),
-            ]
-        )
+        onehotencoding_pipeline = Pipeline([("columnTransformer", ct_cat)])
         clustering_pipeline = Pipeline(
             [("onehotencoder_and_scaler", onehotencoding_pipeline), ("clustering", gmm)]
         )

tests/test_sklearn_calibrated_classifier_cv_converter.py

@@ -191,7 +191,7 @@ def test_model_calibrated_classifier_cv_logistic_regression(self):
         X, y = data.data, data.target
         y[y > 1] = 1
         model = CalibratedClassifierCV(
-            base_estimator=LogisticRegression(), method="sigmoid"
+            estimator=LogisticRegression(), method="sigmoid"
         ).fit(X, y)
         model_onnx = convert_sklearn(
             model,
@@ -215,7 +215,7 @@ def test_model_calibrated_classifier_cv_rf(self):
         X, y = data.data, data.target
         y[y > 1] = 1
         model = CalibratedClassifierCV(
-            base_estimator=RandomForestClassifier(n_estimators=2), method="sigmoid"
+            estimator=RandomForestClassifier(n_estimators=2), method="sigmoid"
         ).fit(X, y)
         model_onnx = convert_sklearn(
             model,
@@ -239,7 +239,7 @@ def test_model_calibrated_classifier_cv_gbt(self):
         X, y = data.data, data.target
         y[y > 1] = 1
         model = CalibratedClassifierCV(
-            base_estimator=GradientBoostingClassifier(n_estimators=2), method="sigmoid"
+            estimator=GradientBoostingClassifier(n_estimators=2), method="sigmoid"
         ).fit(X, y)
         model_onnx = convert_sklearn(
             model,
@@ -264,7 +264,7 @@ def test_model_calibrated_classifier_cv_hgbt(self):
         X, y = data.data, data.target
         y[y > 1] = 1
         model = CalibratedClassifierCV(
-            base_estimator=HistGradientBoostingClassifier(max_iter=4), method="sigmoid"
+            estimator=HistGradientBoostingClassifier(max_iter=4), method="sigmoid"
         ).fit(X, y)
         model_onnx = convert_sklearn(
             model,
@@ -288,7 +288,7 @@ def test_model_calibrated_classifier_cv_tree(self):
         X, y = data.data, data.target
         y[y > 1] = 1
         model = CalibratedClassifierCV(
-            base_estimator=DecisionTreeClassifier(), method="sigmoid"
+            estimator=DecisionTreeClassifier(), method="sigmoid"
         ).fit(X, y)
         model_onnx = convert_sklearn(
             model,
@@ -311,7 +311,7 @@ def test_model_calibrated_classifier_cv_tree(self):
     def test_model_calibrated_classifier_cv_svc(self):
         data = load_iris()
         X, y = data.data, data.target
-        model = CalibratedClassifierCV(base_estimator=SVC(), method="sigmoid").fit(X, y)
+        model = CalibratedClassifierCV(estimator=SVC(), method="sigmoid").fit(X, y)
         model_onnx = convert_sklearn(
             model,
             "unused",
@@ -333,9 +333,9 @@ def test_model_calibrated_classifier_cv_svc(self):
     def test_model_calibrated_classifier_cv_linearsvc(self):
         data = load_iris()
         X, y = data.data, data.target
-        model = CalibratedClassifierCV(
-            base_estimator=LinearSVC(), method="sigmoid"
-        ).fit(X, y)
+        model = CalibratedClassifierCV(estimator=LinearSVC(), method="sigmoid").fit(
+            X, y
+        )
         model_onnx = convert_sklearn(
             model,
             "unused",
@@ -359,9 +359,9 @@ def test_model_calibrated_classifier_cv_linearsvc2(self):
         X, y = data.data, data.target
         y[y == 2] = 0
         self.assertEqual(len(set(y)), 2)
-        model = CalibratedClassifierCV(
-            base_estimator=LinearSVC(), method="sigmoid"
-        ).fit(X, y)
+        model = CalibratedClassifierCV(estimator=LinearSVC(), method="sigmoid").fit(
+            X, y
+        )
         model_onnx = convert_sklearn(
             model,
             "unused",
@@ -391,7 +391,7 @@ def test_model_calibrated_classifier_cv_svc2_binary(self):
             model_sub.fit(X, y)
             with self.subTest(model=model_sub):
                 model = CalibratedClassifierCV(
-                    base_estimator=model_sub, cv=2, method="sigmoid"
+                    estimator=model_sub, cv=2, method="sigmoid"
                 ).fit(X, y)
                 model_onnx = convert_sklearn(
                     model,