What would be the correct type hint for a function that accepts an one dimensional array-like object? More specifically, my function uses np.percentile
and I would like to 'match' np.percentile
's flexibility in terms of the kind of array it accepts (List, pandas Series, numpy array, etc.). Below illustrates what I'm looking for:
def foo(arr: array-like) -> float:
p = np.percentile(arr, 50)
return p
CodePudding user response:
Use numpy.typing.ArrayLike
:
from numpy.typing import ArrayLike
def foo(arr: ArrayLike) -> float:
p = np.percentile(arr, 50)
return p