Consider the following pl.DataFrame:

np_linspace_start = [0, 0, 0]
np_linspace_stop = [8, 6, 7]
np_linspace_num = [5, 4, 4]

df = pl.DataFrame(
    data={
        "np_linspace_start": np_linspace_start, 
        "np_linspace_stop": np_linspace_stop,
        "np_linspace_num": np_linspace_num
    }
)

shape: (3, 3)
┌───────────────────┬──────────────────┬─────────────────┐
│ np_linspace_start ┆ np_linspace_stop ┆ np_linspace_num │
│ ---               ┆ ---              ┆ ---             │
│ i64               ┆ i64              ┆ i64             │
╞═══════════════════╪══════════════════╪═════════════════╡
│ 0                 ┆ 8                ┆ 5               │
│ 0                 ┆ 6                ┆ 4               │
│ 0                 ┆ 7                ┆ 4               │
└───────────────────┴──────────────────┴─────────────────┘

How can I create a new column ls, that is the result of the np.linspace function? This column will hold an np.array.

I was looking for something along those lines:

df.with_columns(
    ls=np.linspace(
        start=pl.col("np_linspace_start"),
        stop=pl.col("np_linspace_stop"),
        num=pl.col("np_linspace_num")
    )
)

Is there a polars equivalent to np.linspace?