[python-package] add type hints for functions accepting dtypes #5773
Conversation
jameslamb
requested review from
StrikerRUS,
shiyu1994 and
jmoralez
as code owners
| @@ -240,7 +240,7 @@ def _is_numpy_column_array(data: Any) -> bool: | |||
| return len(shape) == 2 and shape[1] == 1 | |||
|
|
|||
|
|
|||
| def _cast_numpy_array_to_dtype(array: np.ndarray, dtype: np.dtype) -> np.ndarray: | |||
| def _cast_numpy_array_to_dtype(array: np.ndarray, dtype: "np.typing.DTypeLike") -> np.ndarray: | |||
There was a problem hiding this comment.
these annotations come from https://numpy.org/devdocs/reference/typing.html.
numpy.typing is new as of numpy 1.20 (January 2021).
Expressing these annotations as string literals means that they won't cause issues at runtime for users using lightgbm + older numpy versions. See https://mypy.readthedocs.io/en/stable/runtime_troubles.html#import-cycles.
| @@ -612,7 +616,7 @@ def _c_int_array(data): | |||
| return (ptr_data, type_data, data) # return `data` to avoid the temporary copy is freed | |||
|
|
|||
|
|
|||
| def _is_allowed_numpy_dtype(dtype) -> bool: | |||
| def _is_allowed_numpy_dtype(dtype: type) -> bool: | |||
There was a problem hiding this comment.
why is this one just type?
Setting it to np.typing.DTypeLike results in these mypy errors.
python-package/lightgbm/basic.py:618: error: Argument 1 to "issubclass" has incompatible type "Union[dtype[Any], None, Type[Any], _SupportsDType[dtype[Any]], str, Union[Tuple[Any, int], Tuple[Any, Union[SupportsIndex, Sequence[SupportsIndex]]], List[Any], _DTypeDict, Tuple[Any, Any]]]"; expected "type" [arg-type]
python-package/lightgbm/basic.py:619: error: Argument 1 to "issubclass" has incompatible type "Union[dtype[Any], None, Type[Any], _SupportsDType[dtype[Any]], str, Union[Tuple[Any, int], Tuple[Any, Union[SupportsIndex, Sequence[SupportsIndex]]], List[Any], _DTypeDict, Tuple[Any, Any]]]"; expected "type" [arg-type]
And that's because it's not being passed a dtype... it's being passed a type.
LightGBM/python-package/lightgbm/basic.py
Line 280 in 1585ee1
LightGBM/python-package/lightgbm/basic.py
Lines 624 to 628 in 1585ee1
import pandas as pd
s = pd.Series([1, 2, 3, 4, 5], dtype="Int64")
for column_name, pandas_dtype in s.to_frame().dtypes.items():
print(pandas_dtype)
print(type(pandas_dtype))
# Int64
# <class 'pandas.core.arrays.integer.Int64Dtype'>
|
This pull request has been automatically locked since there has not been any recent activity since it was closed. |
Contributes to #3756.
Contributes to #3867.
Fixes the following errors from
mypy: