Shrink numeric columns to the minimal required datatype

Description

Shrink to the dtype needed to fit the extrema of this Series. This can be used to reduce memory pressure.

Usage

<Expr>$shrink_dtype()

Value

A polars expression

Examples

library("polars")

df <- pl$DataFrame(a = c(-112, 2, 112))$cast(pl$Int64)
df$with_columns(
  shrunk = pl$col("a")$shrink_dtype()
)

#> shape: (3, 2)
#> ┌──────┬────────┐
#> │ a    ┆ shrunk │
#> │ ---  ┆ ---    │
#> │ i64  ┆ i8     │
#> ╞══════╪════════╡
#> │ -112 ┆ -112   │
#> │ 2    ┆ 2      │
#> │ 112  ┆ 112    │
#> └──────┴────────┘