FIX missing_indices were calculated twice in OrdinalEncoder (scikit-learn#27017)

xuefeng-xu · thomasjpfan · TamaraAtanasoska · commit 6ef75233b7b8 · 2023-08-21T22:58:07.000+02:00
Co-authored-by: Thomas J. Fan &lt;thomasjpfan@gmail.com&gt;
diff --git a/doc/whats_new/v1.4.rst b/doc/whats_new/v1.4.rst
@@ -167,6 +167,10 @@ Changelog
 :mod:`sklearn.preprocessing`
 ............................
 
+- |Efficiency| :class:`preprocessing.OrdinalEncoder` avoids calculating
+  missing indices twice to improve efficiency.
+  :pr:`27017` by `Xuefeng Xu <xuefeng-xu>`.
+
 - |Fix| :class:`preprocessing.OneHotEncoder` shows a more informative error message
   when `sparse_output=True` and the output is configured to be pandas.
   :pr:`26931` by `Thomas Fan`_.
diff --git a/sklearn/preprocessing/_encoders.py b/sklearn/preprocessing/_encoders.py
@@ -1508,15 +1508,11 @@ def fit(self, X, y=None):
                 if infrequent is not None:
                     cardinalities[feature_idx] -= len(infrequent)
 
-        # stores the missing indices per category
-        self._missing_indices = {}
+        # missing values are not considered part of the cardinality
+        # when considering unknown categories or encoded_missing_value
         for cat_idx, categories_for_idx in enumerate(self.categories_):
-            for i, cat in enumerate(categories_for_idx):
+            for cat in categories_for_idx:
                 if is_scalar_nan(cat):
-                    self._missing_indices[cat_idx] = i
-
-                    # missing values are not considered part of the cardinality
-                    # when considering unknown categories or encoded_missing_value
                     cardinalities[cat_idx] -= 1
                     continue
 
