pandas-dev#50990 Increase to_dict('list') performance (pandas-dev#54824)

* autoformat

* autoformat

* whatsnew

* reformat

* rm values

* to_numpy()

* to_numpy()

* rm newline

* .items

* unit test

* comment

Author: taytzehao

doc/source/whatsnew/v2.2.0.rst

@@ -158,9 +158,9 @@ Deprecations
 
 Performance improvements
 ~~~~~~~~~~~~~~~~~~~~~~~~
-- Performance improvement in :meth:`DataFrame.sort_index` and :meth:`Series.sort_index` when indexed by a :class:`MultiIndex` (:issue:`54835`, :issue:`54883`)
+- Performance improvement in :func:`to_dict` on converting DataFrame to dictionary (:issue:`50990`)
+- Performance improvement in :meth:`DataFrame.sort_index` and :meth:`Series.sort_index` when indexed by a :class:`MultiIndex` (:issue:`54835`)
 - Performance improvement when indexing with more than 4 keys (:issue:`54550`)
--
 
 .. ---------------------------------------------------------------------------
 .. _whatsnew_220.bug_fixes:

pandas/core/methods/to_dict.py

@@ -106,13 +106,13 @@ def to_dict(
         return into_c((k, v.to_dict(into)) for k, v in df.items())
 
     elif orient == "list":
-        object_dtype_indices_as_set = set(box_native_indices)
+        object_dtype_indices_as_set: set[int] = set(box_native_indices)
         return into_c(
             (
                 k,
-                list(map(maybe_box_native, v.tolist()))
+                list(map(maybe_box_native, v.to_numpy().tolist()))
                 if i in object_dtype_indices_as_set
-                else v.tolist(),
+                else v.to_numpy().tolist(),
             )
             for i, (k, v) in enumerate(df.items())
         )

pandas/tests/frame/methods/test_to_dict.py

@@ -166,6 +166,21 @@ def test_to_dict_not_unique_warning(self):
         with tm.assert_produces_warning(UserWarning):
             df.to_dict()
 
+    @pytest.mark.filterwarnings("ignore::UserWarning")
+    @pytest.mark.parametrize(
+        "orient,expected",
+        [
+            ("list", {"A": [2, 5], "B": [3, 6]}),
+            ("dict", {"A": {0: 2, 1: 5}, "B": {0: 3, 1: 6}}),
+        ],
+    )
+    def test_to_dict_not_unique(self, orient, expected):
+        # GH#54824: This is to make sure that dataframes with non-unique column
+        # would have uniform behavior throughout different orients
+        df = DataFrame([[1, 2, 3], [4, 5, 6]], columns=["A", "A", "B"])
+        result = df.to_dict(orient)
+        assert result == expected
+
     # orient - orient argument to to_dict function
     # item_getter - function for extracting value from
     # the resulting dict using column name and index