Python 如何指定列数据类型

我有以下代码:

import polars as pl
from typing import NamedTuple


class Event(NamedTuple):
    name: str
    description: str


def event_table(num) -> list[Event]:
    events = []
    for i in range(num):
        events.append(Event("name", "description"))
    return events


data = {"events": [1, 2]}
df = pl.DataFrame(data).select(events=pl.col("events").map_elements(event_table))

"""
shape: (2, 1)
┌───────────────────────────────────┐
│ events                            │
│ ---                               │
│ list[struct[2]]                   │
╞═══════════════════════════════════╡
│ [{"name","description"}]          │
│ [{"name","description"}, {"name"… │
└───────────────────────────────────┘
"""

但是如果第一个列表是空的，我得到的是list[list[str]]而不是我需要的list[struct[2]]:

data = {"events": [0, 1, 2]}
df = pl.DataFrame(data).select(events=pl.col("events").map_elements(event_table))
print(df)

"""
shape: (3, 1)
┌───────────────────────────────────┐
│ events                            │
│ ---                               │
│ list[list[str]]                   │
╞═══════════════════════════════════╡
│ []                                │
│ [["name", "description"]]         │
│ [["name", "description"], ["name… │
└───────────────────────────────────┘
"""

我试着使用map_elements函数的return_dtype，比如:

data = {"events": [0, 1, 2]}
df = pl.DataFrame(data).select(
    events=pl.col("events").map_elements(
        event_table,
        return_dtype=pl.List(pl.Struct({"name": pl.String, "description": pl.String})),
    )
)

但这失败了:

Traceback (most recent call last):
  File "script.py", line 18, in <module>
    df = pl.DataFrame(data).select(
         ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.11/site-packages/polars/dataframe/frame.py", line 8193, in select
    return self.lazy().select(*exprs, **named_exprs).collect(_eager=True)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.11/site-packages/polars/lazyframe/frame.py", line 1943, in collect
    return wrap_df(ldf.collect())
                   ^^^^^^^^^^^^^
polars.exceptions.SchemaError: expected output type 'List(Struct([Field { name: "name", dtype: String }, Field { name: "description", dtype: String }]))', got 'List(List(String))'; set `return_dtype` to the proper datatype

我怎么才能让它工作？如果第一个列表是空的，我需要这个列的类型是list[struct[2]]事件.