polars.cumfold

polars.cumfold(

acc: IntoExpr,

function: Callable[[Series, Series], Series],

exprs: Sequence[Expr | str] | Expr,

*,

include_init: bool = False,

) → Expr

Cumulatively accumulate over multiple columns horizontally/ row wise with a left fold.

Every cumulative result is added as a separate field in a Struct column.

Parameters:

acc

Accumulator Expression. This is the value that will be initialized when the fold starts. For a sum this could for instance be lit(0).

function

Function to apply over the accumulator and the value. Fn(acc, value) -> new_value

exprs

Expressions to aggregate over. May also be a wildcard expression.

include_init

Include the initial accumulator state as struct field.

Notes

If you simply want the first encountered expression as accumulator, consider using cumreduce.

Examples

>>> df = pl.DataFrame(
...     {
...         "a": [1, 2, 3],
...         "b": [3, 4, 5],
...         "c": [5, 6, 7],
...     }
... )
>>> df
shape: (3, 3)
┌─────┬─────┬─────┐
│ a   ┆ b   ┆ c   │
│ --- ┆ --- ┆ --- │
│ i64 ┆ i64 ┆ i64 │
╞═════╪═════╪═════╡
│ 1   ┆ 3   ┆ 5   │
│ 2   ┆ 4   ┆ 6   │
│ 3   ┆ 5   ┆ 7   │
└─────┴─────┴─────┘

>>> df.select(
...     pl.cumfold(
...         acc=pl.lit(1), function=lambda acc, x: acc + x, exprs=pl.col("*")
...     ).alias("cumfold"),
... )
shape: (3, 1)
┌───────────┐
│ cumfold   │
│ ---       │
│ struct[3] │
╞═══════════╡
│ {2,5,10}  │
│ {3,7,13}  │
│ {4,9,16}  │
└───────────┘