Skip to content

BUG: .transform(...) with "first" and "last" fail when axis=1 #46074

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jreback merged 2 commits into from
Feb 26, 2022
+15 −41
Merged

BUG: .transform(...) with "first" and "last" fail when axis=1 #46074

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
eb362b1
CLN: Always use transform_fast
rhshadrach Feb 14, 2022
07429bc
whatsnew
rhshadrach Feb 19, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
CLN: Always use transform_fast
  • Loading branch information
@rhshadrach
rhshadrach committed Feb 19, 2022
commit eb362b19638382e979376bb34db01928f0a7dc8f
9 changes: 0 additions & 9 deletions pandas/core/groupby/generic.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -472,9 +472,6 @@ def _transform_general(self, func: Callable, *args, **kwargs) -> Series:
result.name = self.obj.name
return result

def _can_use_transform_fast(self, func: str, result) -> bool:
return True

def filter(self, func, dropna: bool = True, *args, **kwargs):
"""
Return a copy of a Series excluding elements from groups that
Expand Down Expand Up @@ -1184,12 +1181,6 @@ def transform(self, func, *args, engine=None, engine_kwargs=None, **kwargs):
func, *args, engine=engine, engine_kwargs=engine_kwargs, **kwargs
)

def _can_use_transform_fast(self, func: str, result) -> bool:
return func == "size" or (
isinstance(result, DataFrame)
and result.columns.equals(self._obj_with_exclusions.columns)
)

def _define_paths(self, func, *args, **kwargs):
if isinstance(func, str):
fast_path = lambda group: getattr(group, func)(*args, **kwargs)
Expand Down
6 changes: 1 addition & 5 deletions pandas/core/groupby/groupby.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1650,11 +1650,7 @@ def _transform(self, func, *args, engine=None, engine_kwargs=None, **kwargs):
with com.temp_setattr(self, "observed", True):
result = getattr(self, func)(*args, **kwargs)

if self._can_use_transform_fast(func, result):
return self._wrap_transform_fast_result(result)

# only reached for DataFrameGroupBy
return self._transform_general(func, *args, **kwargs)
return self._wrap_transform_fast_result(result)

@final
def _wrap_transform_fast_result(self, result: NDFrameT) -> NDFrameT:
Expand Down
40 changes: 13 additions & 27 deletions pandas/tests/groupby/transform/test_transform.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,10 +21,7 @@
)
import pandas._testing as tm
from pandas.core.groupby.base import maybe_normalize_deprecated_kernels
from pandas.core.groupby.generic import (
DataFrameGroupBy,
SeriesGroupBy,
)
from pandas.core.groupby.generic import DataFrameGroupBy


def assert_fp_equal(a, b):
Expand Down Expand Up @@ -195,10 +192,8 @@ def test_transform_axis_1_reducer(request, reduction_func):
# GH#45715
if reduction_func in (
"corrwith",
"first",
"idxmax",
"idxmin",
"last",
"ngroup",
"nth",
):
Expand Down Expand Up @@ -418,45 +413,36 @@ def test_transform_select_columns(df):
tm.assert_frame_equal(result, expected)


@pytest.mark.parametrize("duplicates", [True, False])
def test_transform_exclude_nuisance(df, duplicates):
Copy link
Member Author

@rhshadrach rhshadrach Feb 19, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In order to still hit the warning this was testing for, I needed to switch from a duplicated float column to a duplicated string column, as we no longer warn in the case of the duplicated float column. But using a string column would always fail with SeriesGroupBy, which is why duplicates=False was removed.

def test_transform_exclude_nuisance(df):
# case that goes through _transform_item_by_item

if duplicates:
# make sure we work with duplicate columns GH#41427
df.columns = ["A", "C", "C", "D"]
df.columns = ["A", "B", "B", "D"]

# this also tests orderings in transform between
# series/frame to make sure it's consistent
expected = {}
grouped = df.groupby("A")

gbc = grouped["C"]
warn = FutureWarning if duplicates else None
with tm.assert_produces_warning(warn, match="Dropping invalid columns"):
expected["C"] = gbc.transform(np.mean)
if duplicates:
# squeeze 1-column DataFrame down to Series
expected["C"] = expected["C"]["C"]
gbc = grouped["B"]
with tm.assert_produces_warning(FutureWarning, match="Dropping invalid columns"):
expected["B"] = gbc.transform(lambda x: np.mean(x))
# squeeze 1-column DataFrame down to Series
expected["B"] = expected["B"]["B"]

assert isinstance(gbc.obj, DataFrame)
assert isinstance(gbc, DataFrameGroupBy)
else:
assert isinstance(gbc, SeriesGroupBy)
assert isinstance(gbc.obj, Series)
assert isinstance(gbc.obj, DataFrame)
assert isinstance(gbc, DataFrameGroupBy)

expected["D"] = grouped["D"].transform(np.mean)
expected = DataFrame(expected)
with tm.assert_produces_warning(FutureWarning, match="Dropping invalid columns"):
result = df.groupby("A").transform(np.mean)
result = df.groupby("A").transform(lambda x: np.mean(x))

tm.assert_frame_equal(result, expected)


def test_transform_function_aliases(df):
with tm.assert_produces_warning(FutureWarning, match="Dropping invalid columns"):
result = df.groupby("A").transform("mean")
expected = df.groupby("A").transform(np.mean)
result = df.groupby("A").transform("mean")
expected = df.groupby("A").transform(np.mean)
tm.assert_frame_equal(result, expected)

result = df.groupby("A")["C"].transform("mean")
Expand Down