[BUG] cudf.pandas exponentially weighted mean differ from pandas when there are leading NULLs #20554
Description
Describe the bug
See the repro
Steps/Code to reproduce bug
import pandas as pd
s = pd.Series([float("nan"), 1.0, float("nan"), 1.0, 1.0])
s.ewm(com=3, adjust=False, ignore_na=False, min_periods=0).mean()With cudf.pandas
Out[1]:
0 NaN
1 NaN
2 NaN
3 NaN
4 NaN
dtype: float64Without cudf.pandas
Out[1]:
0 NaN
1 1.0
2 1.0
3 1.0
4 1.0
dtype: float64
Expected behavior
Match pandas semantics
Metadata
Metadata
Assignees
Type
Projects
Status