Fix type hints errors in gymnasium/spaces #327
Conversation
vcharraut
changed the title
gymnasium/spaces/utils.py
Outdated
| return unflatten_x.reshape(-1, *unflatten_space.shape) | ||
| elif isinstance(unflatten_space, Discrete): | ||
| return np.asarray(np.nonzero(unflatten_x))[-1, :] | ||
|
|
There was a problem hiding this comment.
Can you be more specific ?
There was a problem hiding this comment.
else:
raise TypeError("Expected the ... ")
gymnasium/spaces/discrete.py
Outdated
| @@ -55,7 +56,7 @@ def is_np_flattenable(self): | |||
| """Checks whether this space can be flattened to a :class:`spaces.Box`.""" | |||
| return True | |||
|
|
|||
| def sample(self, mask: MaskNDArray | None = None) -> int: | |||
| def sample(self, mask: MaskNDArray | None = None) -> np.int64 | NDArray[np.int64]: | |||
There was a problem hiding this comment.
I still think this should just be np.int64
gymnasium/spaces/multi_binary.py
Outdated
| @@ -58,7 +58,7 @@ def is_np_flattenable(self): | |||
| """Checks whether this space can be flattened to a :class:`spaces.Box`.""" | |||
| return True | |||
|
|
|||
| def sample(self, mask: MaskNDArray | None = None) -> npt.NDArray[np.int8]: | |||
| def sample(self, mask: MaskNDArray | None = None) -> NDArray[np.bool_ | np.int8]: | |||
There was a problem hiding this comment.
What case causes np.bool_ as sample return type as it should be the same type as class MultiBinary(Space[NDArray[np.int8]]):
There was a problem hiding this comment.
Well, this comes from np_random.integers that has a wild possiblity of type return, like int, bool and ndarray.
I don't know how to deal with it.
Since np_random.integers accepts several type in args, I guess there would be a way to force the return type of the function based on the type of the args.
I have actually the exact same proble in the graph.py and utils.py files: there are some functions that can accept multiples spaces types but when used with specific spaces, the type raised errors (like flatten_space in utils.py.
That is already the same kind of problem I encountered with this part of code #327 (comment) and I ended up splitting the function into 2 functions to handle each type cases
There was a problem hiding this comment.
For example, even if self.n has a static type np.int64, this line is raises an error
tmp: np.int64 = self.np_random.integers(self.n)
Expression of type "ndarray[Any, dtype[int64]]" cannot be assigned to declared type "int64"
"ndarray[Any, dtype[int64]]" is incompatible with "int64"
There was a problem hiding this comment.
Can we cast the data in some way to fix this, otherwise, raise an issue in numpy.
There was a problem hiding this comment.
I will try to look for solutions and do another PR
I think it is better to merge everything for now, I reverted the things I didn't solve. |
Description
Removed
reportInvalidTypeVarUseandreportGeneralTypeIssueswith pyright and fixed the new raised issues.Files that are covered without errors:
Type of change
Please delete options that are not relevant.
Checklist:
pre-commitchecks withpre-commit run --all-files(seeCONTRIBUTING.mdinstructions to set it up)