Problem with `.to_dict()` type inference

Describe the bug

I have a dataframe or pandas series and when I use the .to_dict() the inferred type by pandas-stubs is Dict[Hashable, Any]. In my case I know that that the index is an int but there is an error when I type the output as Dict[int, Any] saying that int is not a Hashable type.

Is there any solution for better inferring the type in this case?

To Reproduce

Please complete the following information:

1. Provide a minimal runnable pandas example that is not properly checked by the stubs.

import pandas as pd

id=[1,2,3]
value=["a", "b", "c"]

df = pd.DataFrame(zip(id, value), columns=["id", "value"])
df.set_index("id")["value"].to_dict()

def funct(df_dict: Dict[int, Any]) -> None:
     pass

1. Indicate which type checker you are using (mypy or pyright).3. Mypy

2. Show the error message received from that type checker while checking your example.4.

error: Argument 4 to "funct" has incompatible type "Dict[Hashable, Any]"; expected "Dict[int, Any]"

Additional context I am running it in python version 3.10

[twoertwein]

DataFrame/Series are not Generic with respect to the index type, so we cannot return dict[int, Any. I think we would need to relax the annotation to just dict to avoid any mypy/pyright errors when users implicitly “cast” it to dict[int, Any]. I’d be okay with that change (for input arguments, I would still require something like dict[HashableT, Any] but for return values, it might be worth relaxing it).

@Dr-Irv @bashtage

[twoertwein]

Mapping makes only the values covariant. I would prioritize getting the container type right (dict vs (Mutable)Mapping) as opposed to differentiating Any vs Hashable.