Describe the bug
A tiny bug/typo in the preprocessing roll method for the fill strategy leading to NameError when using roll function. The reason is that the variable name used is X_X_rolling instead of X_rolling in the row 324.
To Reproduce
import polars as pl
from functime.preprocessing import roll
from pathlib import Path
file_path = Path(__file__).parent.parent / "data" / "commodities.parquet"
df = pl.read_parquet(source=file_path)
transformer = roll(
window_sizes=[10, 30, 60],
stats=["sum", "mean", "std"],
freq="1mo",
fill_strategy="backward",
)
df.pipe(transformer).collect()
Describe the bug A tiny bug/typo in the preprocessing roll method for the fill strategy leading to NameError when using roll function. The reason is that the variable name used is X_X_rolling instead of X_rolling in the row 324.
To Reproduce