fix

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

Author: xadupre

skl2onnx/operator_converters/ordinal_encoder.py

@@ -43,16 +43,6 @@ def convert_sklearn_ordinal_encoder(
     )
 
     ordinal_categories = ordinal_op.categories_
-    if (
-        hasattr(ordinal_op, "infrequent_categories_")
-        and ordinal_op.infrequent_categories_
-    ):
-        # exclude infrequent categories if present
-        ordinal_categories = [
-            categories
-            for categories in ordinal_categories
-            if categories not in ordinal_op.infrequent_categories_
-        ]
 
     for categories in ordinal_categories:
         if len(categories) == 0:

tests/test_sklearn_ordinal_encoder.py

@@ -254,12 +254,7 @@ def test_model_ordinal_encoder_encoded_missing_value(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data,
-            },
-        )
+        got = sess.run(None, {"input": data})
 
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
@@ -292,12 +287,7 @@ def test_model_ordinal_encoder_encoded_missing_value_no_nan(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data,
-            },
-        )
+        got = sess.run(None, {"input": data})
 
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
@@ -395,9 +385,7 @@ def test_model_ordinal_encoder_max_categories(self):
         data = np.array(
             [["a"], ["b"], ["c"], ["d"], ["a"], ["b"], ["c"], ["e"]], dtype=np.object_
         )
-
         expected = model.fit_transform(data)
-
         model_onnx = convert_sklearn(
             model,
             "scikit-learn ordinal encoder",
@@ -415,12 +403,7 @@ def test_model_ordinal_encoder_max_categories(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data,
-            },
-        )
+        got = sess.run(None, {"input": data})
 
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
@@ -455,12 +438,7 @@ def test_model_ordinal_encoder_min_frequency(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data,
-            },
-        )
+        got = sess.run(None, {"input": data})
 
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
@@ -495,12 +473,7 @@ def test_model_ordinal_encoder_unknown_value_nan(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data_with_missing_value,
-            },
-        )
+        got = sess.run(None, {"input": data_with_missing_value})
 
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
@@ -559,21 +532,11 @@ def test_model_ordinal_encoder_min_frequency_multi_column(self):
         sess = InferenceSession(
             model_onnx.SerializeToString(), providers=["CPUExecutionProvider"]
         )
-        got = sess.run(
-            None,
-            {
-                "input": data,
-            },
-        )
+        got = sess.run(None, {"input": data})
         assert_almost_equal(expected.reshape(-1), got[0].reshape(-1))
 
         # Test with test data
-        got_test = sess.run(
-            None,
-            {
-                "input": test_data,
-            },
-        )
+        got_test = sess.run(None, {"input": test_data})
         assert_almost_equal(expected_test.reshape(-1), got_test[0].reshape(-1))