BUG: Grouper specified by key not regarded as in-axis

doc/source/whatsnew/v2.0.0.rst

@@ -932,6 +932,8 @@ Groupby/resample/rolling
 - Bug in :class:`.DataFrameGroupBy` and :class:`.SeriesGroupBy` with ``dropna=False`` would drop NA values when the grouper was categorical (:issue:`36327`)
 - Bug in :meth:`.SeriesGroupBy.nunique` would incorrectly raise when the grouper was an empty categorical and ``observed=True`` (:issue:`21334`)
 - Bug in :meth:`.SeriesGroupBy.nth` would raise when grouper contained NA values after subsetting from a :class:`DataFrameGroupBy` (:issue:`26454`)
+- Bug in :meth:`DataFrame.groupby` would not include a :class:`.Grouper` specified by ``key`` in the result when ``as_index=False`` (:issue:`50413`)
+-
 
 Reshaping
 ^^^^^^^^^

pandas/core/groupby/groupby.py

@@ -1014,10 +1014,10 @@ def _set_group_selection(self) -> None:
         """
         # This is a no-op for SeriesGroupBy
         grp = self.grouper
-        if not (
-            grp.groupings is not None
-            and self.obj.ndim > 1
-            and self._group_selection is None
+        if (
+            grp.groupings is None
+            or self.obj.ndim == 1
+            or self._group_selection is not None
         ):
             return
 

pandas/core/groupby/grouper.py

@@ -906,7 +906,7 @@ def is_in_obj(gpr) -> bool:
         elif isinstance(gpr, Grouper) and gpr.key is not None:
             # Add key to exclusions
             exclusions.add(gpr.key)
-            in_axis = False
+            in_axis = True
         else:
             in_axis = False
 

pandas/tests/groupby/test_grouping.py

@@ -7,6 +7,7 @@
 from pandas import (
     CategoricalIndex,
     DataFrame,
+    Grouper,
     Index,
     MultiIndex,
     Series,
@@ -168,7 +169,7 @@ def test_grouper_index_types(self, index):
     def test_grouper_multilevel_freq(self):
 
         # GH 7885
-        # with level and freq specified in a pd.Grouper
+        # with level and freq specified in a Grouper
         from datetime import (
             date,
             timedelta,
@@ -182,20 +183,20 @@ def test_grouper_multilevel_freq(self):
         # Check string level
         expected = (
             df.reset_index()
-            .groupby([pd.Grouper(key="foo", freq="W"), pd.Grouper(key="bar", freq="W")])
+            .groupby([Grouper(key="foo", freq="W"), Grouper(key="bar", freq="W")])
             .sum()
         )
         # reset index changes columns dtype to object
         expected.columns = Index([0], dtype="int64")
 
         result = df.groupby(
-            [pd.Grouper(level="foo", freq="W"), pd.Grouper(level="bar", freq="W")]
+            [Grouper(level="foo", freq="W"), Grouper(level="bar", freq="W")]
         ).sum()
         tm.assert_frame_equal(result, expected)
 
         # Check integer level
         result = df.groupby(
-            [pd.Grouper(level=0, freq="W"), pd.Grouper(level=1, freq="W")]
+            [Grouper(level=0, freq="W"), Grouper(level=1, freq="W")]
         ).sum()
         tm.assert_frame_equal(result, expected)
 
@@ -206,11 +207,11 @@ def test_grouper_creation_bug(self):
         g = df.groupby("A")
         expected = g.sum()
 
-        g = df.groupby(pd.Grouper(key="A"))
+        g = df.groupby(Grouper(key="A"))
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
-        g = df.groupby(pd.Grouper(key="A", axis=0))
+        g = df.groupby(Grouper(key="A", axis=0))
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
@@ -220,13 +221,13 @@ def test_grouper_creation_bug(self):
         tm.assert_frame_equal(result, expected)
 
         # GH14334
-        # pd.Grouper(key=...) may be passed in a list
+        # Grouper(key=...) may be passed in a list
         df = DataFrame(
             {"A": [0, 0, 0, 1, 1, 1], "B": [1, 1, 2, 2, 3, 3], "C": [1, 2, 3, 4, 5, 6]}
         )
         # Group by single column
         expected = df.groupby("A").sum()
-        g = df.groupby([pd.Grouper(key="A")])
+        g = df.groupby([Grouper(key="A")])
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
@@ -235,17 +236,17 @@ def test_grouper_creation_bug(self):
         expected = df.groupby(["A", "B"]).sum()
 
         # Group with two Grouper objects
-        g = df.groupby([pd.Grouper(key="A"), pd.Grouper(key="B")])
+        g = df.groupby([Grouper(key="A"), Grouper(key="B")])
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
         # Group with a string and a Grouper object
-        g = df.groupby(["A", pd.Grouper(key="B")])
+        g = df.groupby(["A", Grouper(key="B")])
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
         # Group with a Grouper object and a string
-        g = df.groupby([pd.Grouper(key="A"), "B"])
+        g = df.groupby([Grouper(key="A"), "B"])
         result = g.sum()
         tm.assert_frame_equal(result, expected)
 
@@ -257,15 +258,15 @@ def test_grouper_creation_bug(self):
                 names=["one", "two", "three"],
             ),
         )
-        result = s.groupby(pd.Grouper(level="three", freq="M")).sum()
+        result = s.groupby(Grouper(level="three", freq="M")).sum()
         expected = Series(
             [28],
             index=pd.DatetimeIndex([Timestamp("2013-01-31")], freq="M", name="three"),
         )
         tm.assert_series_equal(result, expected)
 
         # just specifying a level breaks
-        result = s.groupby(pd.Grouper(level="one")).sum()
+        result = s.groupby(Grouper(level="one")).sum()
         expected = s.groupby(level="one").sum()
         tm.assert_series_equal(result, expected)
 
@@ -282,18 +283,14 @@ def test_grouper_column_and_index(self):
             {"A": np.arange(6), "B": ["one", "one", "two", "two", "one", "one"]},
             index=idx,
         )
-        result = df_multi.groupby(["B", pd.Grouper(level="inner")]).mean(
-            numeric_only=True
-        )
+        result = df_multi.groupby(["B", Grouper(level="inner")]).mean(numeric_only=True)
         expected = (
             df_multi.reset_index().groupby(["B", "inner"]).mean(numeric_only=True)
         )
         tm.assert_frame_equal(result, expected)
 
         # Test the reverse grouping order
-        result = df_multi.groupby([pd.Grouper(level="inner"), "B"]).mean(
-            numeric_only=True
-        )
+        result = df_multi.groupby([Grouper(level="inner"), "B"]).mean(numeric_only=True)
         expected = (
             df_multi.reset_index().groupby(["inner", "B"]).mean(numeric_only=True)
         )
@@ -302,7 +299,7 @@ def test_grouper_column_and_index(self):
         # Grouping a single-index frame by a column and the index should
         # be equivalent to resetting the index and grouping by two columns
         df_single = df_multi.reset_index("outer")
-        result = df_single.groupby(["B", pd.Grouper(level="inner")]).mean(
+        result = df_single.groupby(["B", Grouper(level="inner")]).mean(
             numeric_only=True
         )
         expected = (
@@ -311,7 +308,7 @@ def test_grouper_column_and_index(self):
         tm.assert_frame_equal(result, expected)
 
         # Test the reverse grouping order
-        result = df_single.groupby([pd.Grouper(level="inner"), "B"]).mean(
+        result = df_single.groupby([Grouper(level="inner"), "B"]).mean(
             numeric_only=True
         )
         expected = (
@@ -368,7 +365,7 @@ def test_grouper_getting_correct_binner(self):
             ),
         )
         result = df.groupby(
-            [pd.Grouper(level="one"), pd.Grouper(level="two", freq="M")]
+            [Grouper(level="one"), Grouper(level="two", freq="M")]
         ).sum()
         expected = DataFrame(
             {"A": [31, 28, 21, 31, 28, 21]},
@@ -646,7 +643,7 @@ def test_list_grouper_with_nat(self):
         # GH 14715
         df = DataFrame({"date": date_range("1/1/2011", periods=365, freq="D")})
         df.iloc[-1] = pd.NaT
-        grouper = pd.Grouper(key="date", freq="AS")
+        grouper = Grouper(key="date", freq="AS")
 
         # Grouper in a list grouping
         result = df.groupby([grouper])
@@ -847,7 +844,7 @@ def test_groupby_with_empty(self):
         index = pd.DatetimeIndex(())
         data = ()
         series = Series(data, index, dtype=object)
-        grouper = pd.Grouper(freq="D")
+        grouper = Grouper(freq="D")
         grouped = series.groupby(grouper)
         assert next(iter(grouped), None) is None
 
@@ -982,7 +979,7 @@ def test_groupby_with_small_elem(self):
             {"event": ["start", "start"], "change": [1234, 5678]},
             index=pd.DatetimeIndex(["2014-09-10", "2013-10-10"]),
         )
-        grouped = df.groupby([pd.Grouper(freq="M"), "event"])
+        grouped = df.groupby([Grouper(freq="M"), "event"])
         assert len(grouped.groups) == 2
         assert grouped.ngroups == 2
         assert (Timestamp("2014-09-30"), "start") in grouped.groups
@@ -997,7 +994,7 @@ def test_groupby_with_small_elem(self):
             {"event": ["start", "start", "start"], "change": [1234, 5678, 9123]},
             index=pd.DatetimeIndex(["2014-09-10", "2013-10-10", "2014-09-15"]),
         )
-        grouped = df.groupby([pd.Grouper(freq="M"), "event"])
+        grouped = df.groupby([Grouper(freq="M"), "event"])
         assert len(grouped.groups) == 2
         assert grouped.ngroups == 2
         assert (Timestamp("2014-09-30"), "start") in grouped.groups
@@ -1013,7 +1010,7 @@ def test_groupby_with_small_elem(self):
             {"event": ["start", "start", "start"], "change": [1234, 5678, 9123]},
             index=pd.DatetimeIndex(["2014-09-10", "2013-10-10", "2014-08-05"]),
         )
-        grouped = df.groupby([pd.Grouper(freq="M"), "event"])
+        grouped = df.groupby([Grouper(freq="M"), "event"])
         assert len(grouped.groups) == 3
         assert grouped.ngroups == 3
         assert (Timestamp("2014-09-30"), "start") in grouped.groups
@@ -1036,3 +1033,17 @@ def test_grouping_string_repr(self):
         result = gr.grouper.groupings[0].__repr__()
         expected = "Grouping(('A', 'a'))"
         assert result == expected
+
+
+def test_grouping_by_key_is_in_axis():
+    # GH#50413 - Groupers specified by key are in-axis
+    df = DataFrame({"a": [1, 1, 2], "b": [1, 1, 2], "c": [3, 4, 5]}).set_index("a")
+    gb = df.groupby([Grouper(level="a"), Grouper(key="b")], as_index=False)
+    assert not gb.grouper.groupings[0].in_axis
+    assert gb.grouper.groupings[1].in_axis
+
+    # Currently only in-axis groupings are including in the result when as_index=False;
+    # This is likely to change in the future.
+    result = gb.sum()
+    expected = DataFrame({"b": [1, 2], "c": [7, 5]})
+    tm.assert_frame_equal(result, expected)