REGR: SeriesGroupBy.apply sets name attribute if result is DataFrame (#49908)

Author: MarcoGorelli

doc/source/whatsnew/v1.5.3.rst

@@ -16,6 +16,7 @@ Fixed regressions
 - Fixed performance regression in :meth:`Series.isin` when ``values`` is empty (:issue:`49839`)
 - Fixed regression in :meth:`DataFrameGroupBy.transform` when used with ``as_index=False`` (:issue:`49834`)
 - Enforced reversion of ``color`` as an alias for ``c`` and ``size`` as an alias for ``s`` in function :meth:`DataFrame.plot.scatter` (:issue:`49732`)
+- Fixed regression in :meth:`SeriesGroupBy.apply` setting a ``name`` attribute on the result if the result was a :class:`DataFrame` (:issue:`49907`)
 -
 
 .. ---------------------------------------------------------------------------

pandas/core/groupby/generic.py

@@ -401,7 +401,8 @@ def _wrap_applied_output(
                 not_indexed_same=not_indexed_same,
                 is_transform=is_transform,
             )
-            result.name = self.obj.name
+            if isinstance(result, Series):
+                result.name = self.obj.name
             return result
         else:
             # GH #6265 #24880

pandas/tests/groupby/test_apply.py

@@ -334,6 +334,7 @@ def f(piece):
     result = grouped.apply(f)
 
     assert isinstance(result, DataFrame)
+    assert not hasattr(result, "name")  # GH49907
     tm.assert_index_equal(result.index, ts.index)