Skip to content

fix: Keep null values of the latest row in deduplicate_keep_latest #7050

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

Open
ShryukGrandhi wants to merge 1 commit into
base: master
Choose a base branch
from
Open

fix: Keep null values of the latest row in deduplicate_keep_latest #7050

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions sdk/python/feast/aggregation/tiling/tile_subtraction.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -190,6 +190,6 @@ def deduplicate_keep_latest(

return (
df.sort_values(by=timestamp_col, ascending=False)
.groupby(entity_keys, as_index=False)
.first()
.drop_duplicates(subset=entity_keys, keep="first")
.reset_index(drop=True)
)
27 changes: 27 additions & 0 deletions sdk/python/tests/unit/test_aggregation_ops.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -288,3 +288,30 @@ def windowed(frame):
assert len(expected) == 8
assert result["event_timestamp"].tolist() == expected["event_timestamp"].tolist()
assert result[feature_name].tolist() == expected[feature_name].tolist()


def test_deduplicate_keep_latest_keeps_null_values_of_latest_row():
"""A null in the latest row is kept rather than filled from an older row."""
from feast.aggregation.tiling.tile_subtraction import deduplicate_keep_latest

start = pd.Timestamp("2024-01-01")
df = pd.DataFrame(
{
"driver_id": [1, 1, 2],
"event_timestamp": [
start,
start + timedelta(minutes=5),
start,
],
# std over a window with a single event is NaN
"std_trips_1200s": [2.5, float("nan"), 1.0],
}
)

result = deduplicate_keep_latest(df, ["driver_id"], "event_timestamp")
result = result.set_index("driver_id")

assert len(result) == 2
assert result.loc[1, "event_timestamp"] == start + timedelta(minutes=5)
assert pd.isna(result.loc[1, "std_trips_1200s"])
assert result.loc[2, "std_trips_1200s"] == 1.0