Merge pull request #258 from lvgig/257-bug-datetimeinfoextractor-prev…

…ents-pipeline-from-being-saved-using-joblib-or-pickle

🐛 set mappings_provided to list type so object is picklable

CHANGELOG.rst

@@ -15,7 +15,12 @@ Subsections for each version can be one of the following;
 - ``Security`` in case of vulnerabilities.
 
 Each individual change should have a link to the pull request after the description of the change.
+1.3.1 (unreleased)
+------------------
 
+Changed
+^^^^^^^
+- DatetimeInfoExtractor.mappings_provided changed from a dict.keys() object to list so transformer is serialisable. `#258 <https://github.com/lvgig/tubular/pull/258>`_
 
 1.3.0 (2024-06-13)
 ------------------

tests/base_tests.py

@@ -2,6 +2,7 @@
 import copy
 import re
 
+import joblib
 import numpy as np
 import pandas as pd
 import pytest
@@ -816,3 +817,9 @@ class OtherBaseBehaviourTests(
     Class to collect and hold tests for BaseTransformerBehaviour outside the three standard methods.
     Note this deliberately avoids starting with "Tests" so that the tests are not run on import.
     """
+
+    def test_is_serialisable(self, initialized_transformers, tmp_path):
+        path = tmp_path / "transformer.pkl"
+
+        # serialise without raising error
+        joblib.dump(initialized_transformers[self.transformer_name], path)

tests/dates/test_DateTimeInfoExtractor.py

@@ -1,3 +1,4 @@
+import joblib
 import numpy as np
 import pandas as pd
 import pytest
@@ -391,3 +392,12 @@ def test_correct_df_returned_datetime_input(self):
             expected,
             "incorrect dataframe returned",
         )
+
+    def test_is_serialisable(self, tmp_path):
+        transformer = DatetimeInfoExtractor(columns=["b"], include=["timeofyear"])
+
+        # pickle transformer
+        path = tmp_path / "transformer.pkl"
+
+        # serialise without raising error
+        joblib.dump(transformer, path)

tubular/_version.py

@@ -1 +1 @@
-__version__ = "1.3.0"
+__version__ = "1.3.1"

tubular/dates.py

@@ -966,7 +966,7 @@ def __init__(
 
         self.include = include
         self.datetime_mappings = datetime_mappings
-        self.mappings_provided = self.datetime_mappings.keys()
+        self.mappings_provided = list(self.datetime_mappings.keys())
 
         self._process_provided_mappings()