Skip to content

API: reimplement FixedWindowIndexer.get_window_bounds #37035

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.

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
merged 6 commits into from
Oct 10, 2020
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
1 change: 1 addition & 0 deletions doc/source/whatsnew/v1.2.0.rst
Original file line number Diff line number Diff line change
Expand Up @@ -444,6 +444,7 @@ Groupby/resample/rolling
- Bug in :meth:`Rolling.count` returned ``np.nan`` with :class:`pandas.api.indexers.FixedForwardWindowIndexer` as window, ``min_periods=0`` and only missing values in window (:issue:`35579`)
- Bug where :class:`pandas.core.window.Rolling` produces incorrect window sizes when using a ``PeriodIndex`` (:issue:`34225`)
- Bug in :meth:`RollingGroupby.count` where a ``ValueError`` was raised when specifying the ``closed`` parameter (:issue:`35869`)
- Bug in :meth:`DataFrame.groupby.rolling` returning wrong values with partial centered window (:issue:`36040`).

Reshaping
^^^^^^^^^
Expand Down
30 changes: 8 additions & 22 deletions pandas/core/window/indexers.py
Original file line number Diff line number Diff line change
Expand Up @@ -78,30 +78,16 @@ def get_window_bounds(
closed: Optional[str] = None,
) -> Tuple[np.ndarray, np.ndarray]:

start_s = np.zeros(self.window_size, dtype="int64")
start_e = (
np.arange(self.window_size, num_values, dtype="int64")
- self.window_size
+ 1
)
start = np.concatenate([start_s, start_e])[:num_values]

end_s = np.arange(self.window_size, dtype="int64") + 1
end_e = start_e + self.window_size
end = np.concatenate([end_s, end_e])[:num_values]

if center and self.window_size > 2:
offset = min((self.window_size - 1) // 2, num_values - 1)
start_s_buffer = np.roll(start, -offset)[: num_values - offset]
end_s_buffer = np.roll(end, -offset)[: num_values - offset]
if center:
offset = (self.window_size - 1) // 2
else:
offset = 0

start_e_buffer = np.arange(
start[-1] + 1, start[-1] + 1 + offset, dtype="int64"
)
end_e_buffer = np.array([end[-1]] * offset, dtype="int64")
end = np.arange(1 + offset, num_values + 1 + offset, dtype="int64")
start = end - self.window_size

start = np.concatenate([start_s_buffer, start_e_buffer])
end = np.concatenate([end_s_buffer, end_e_buffer])
end = np.clip(end, 0, num_values)
start = np.clip(start, 0, num_values)

return start, end

Expand Down
27 changes: 27 additions & 0 deletions pandas/tests/window/test_grouper.py
Original file line number Diff line number Diff line change
Expand Up @@ -297,6 +297,33 @@ def test_groupby_rolling_center_center(self):
)
tm.assert_frame_equal(result, expected)

@pytest.mark.parametrize("min_periods", [5, 4, 3])
def test_groupby_rolling_center_min_periods(self, min_periods):
# GH 36040
df = pd.DataFrame({"group": ["A"] * 10 + ["B"] * 10, "data": range(20)})

window_size = 5
result = (
df.groupby("group")
.rolling(window_size, center=True, min_periods=min_periods)
.mean()
)
result = result.reset_index()[["group", "data"]]

grp_A_mean = [1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 7.5, 8.0]
grp_B_mean = [x + 10.0 for x in grp_A_mean]

num_nans = max(0, min_periods - 3) # For window_size of 5
nans = [np.nan] * num_nans
grp_A_expected = nans + grp_A_mean[num_nans : 10 - num_nans] + nans
grp_B_expected = nans + grp_B_mean[num_nans : 10 - num_nans] + nans

expected = pd.DataFrame(
{"group": ["A"] * 10 + ["B"] * 10, "data": grp_A_expected + grp_B_expected}
)

tm.assert_frame_equal(result, expected)

def test_groupby_subselect_rolling(self):
# GH 35486
df = DataFrame(
Expand Down