bug: Set additionalProperties to True for preprocessing schemas.

ludwig/schema/features/preprocessing/utils.py

@@ -53,7 +53,7 @@ def _jsonschema_type_mapping():
                 "type": "object",
                 "properties": props,
                 "title": "preprocessing_options",
-                "additionalProperties": False,
+                "additionalProperties": True,
             }
 
     try:

ludwig/schema/preprocessing.py

@@ -48,7 +48,7 @@ def get_preprocessing_jsonschema():
     return {
         "type": "object",
         "properties": props,
-        "additionalProperties": False,
+        "additionalProperties": True,
         "title": "global_preprocessing_options",
         "description": "Select the preprocessing type.",
     }

tests/ludwig/schema/test_validate_config_features.py

@@ -25,9 +25,10 @@ def test_incorrect_input_features_config():
         "output_features": [binary_feature()],
     }
 
-    # Not a preprocessing param for category feature
-    with pytest.raises(ValidationError):
-        validate_config(config)
+    # TODO(ksbrar): Circle back after discussing whether additional properties should be allowed long-term.
+    # # Not a preprocessing param for category feature
+    # with pytest.raises(ValidationError):
+    #     validate_config(config)
 
     config = {
         "input_features": [

tests/ludwig/schema/test_validate_config_misc.py

@@ -161,27 +161,28 @@ def test_config_bad_encoder_name():
         validate_config(config)
 
 
-def test_config_bad_preprocessing_param():
-    config = {
-        "input_features": [
-            sequence_feature(encoder={"type": "parallel_cnn", "reduce_output": "sum"}),
-            image_feature(
-                "/tmp/destination_folder",
-                preprocessing={
-                    "in_memory": True,
-                    "height": 12,
-                    "width": 12,
-                    "num_channels": 3,
-                    "tokenizer": "space",
-                },
-            ),
-        ],
-        "output_features": [category_feature(encoder={"vocab_size": 2}, reduce_input="sum")],
-        "combiner": {"type": "concat", "output_size": 14},
-    }
-
-    with pytest.raises(ValidationError, match=r"^Additional properties are not allowed .*"):
-        validate_config(config)
+# TODO(ksbrar): Circle back after discussing whether additional properties should be allowed long-term.
+# def test_config_bad_preprocessing_param():
+#     config = {
+#         "input_features": [
+#             sequence_feature(encoder={"type": "parallel_cnn", "reduce_output": "sum"}),
+#             image_feature(
+#                 "/tmp/destination_folder",
+#                 preprocessing={
+#                     "in_memory": True,
+#                     "height": 12,
+#                     "width": 12,
+#                     "num_channels": 3,
+#                     "tokenizer": "space",
+#                 },
+#             ),
+#         ],
+#         "output_features": [category_feature(encoder={"vocab_size": 2}, reduce_input="sum")],
+#         "combiner": {"type": "concat", "output_size": 14},
+#     }
+
+#     with pytest.raises(ValidationError, match=r"^Additional properties are not allowed .*"):
+#         validate_config(config)
 
 
 def test_config_fill_values():

tests/ludwig/schema/test_validate_config_preprocessing.py

@@ -1,4 +1,4 @@
-import pytest
+# import pytest
 
 from ludwig.schema import validate_config
 from tests.integration_tests.utils import binary_feature, category_feature
@@ -22,7 +22,8 @@ def test_config_preprocessing():
 
     validate_config(config)
 
-    config["preprocessing"]["fake_parameter"] = True
+    # TODO(ksbrar): Circle back after discussing whether additional properties should be allowed long-term.
+    # config["preprocessing"]["fake_parameter"] = True
 
-    with pytest.raises(Exception):
-        validate_config(config)
+    # with pytest.raises(Exception):
+    #     validate_config(config)