You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If that is the case, should we make this explicit in the two doc pages linked above?
I came across this issue recently when I was writing some custom serialization/deserialization functions that take into account the shape and type of ndarrays from the type annotations, and found the typing convention isn't clearly documented. Having a convention for annotation ndarray could help make the type annotations more interoperable.
The text was updated successfully, but these errors were encountered:
Issue with current documentation:
Currently it isn't obvious how one should type
ndarray
: should one type them likendarray[shape, dtype]
orndarray[dtype, shape]
?In the few places in the docs where I think this is relevant, no information is given as to how one should type
ndarray
:Both pages link to numpy.typing.NDArray, it appears that one should follow the convention of
ndarray[shape, dtype]
Idea or request for content:
If that is the case, should we make this explicit in the two doc pages linked above?
I came across this issue recently when I was writing some custom serialization/deserialization functions that take into account the shape and type of
ndarray
s from the type annotations, and found the typing convention isn't clearly documented. Having a convention for annotationndarray
could help make the type annotations more interoperable.The text was updated successfully, but these errors were encountered: