Join 36 million developers who use GitHub
issues to help identify, assign, and keep track of the features and
bug fixes your projects need.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and
privacy statement. We’ll occasionally send you account related emails.
Already on GitHub?
to your account
Calling dask.array.fix() on a Dask array created from a CuPy array fails as it depends on unimplemented __array_wrap__ from CuPy. There's an issue open to add that feature to CuPy cupy/cupy#589.
FWIW raised issue ( numpy/numpy#13100 ) to see if fix and a few other functions could be converted to ufuncs, which would be one way to solve this. Please feel free to join the discussion over there if it is of interest.
Thanks @jakirkham, I've subscribed for notifications, I'll join if I can help with something. :)
Just in case it is needed, here's a workaround.
np.where(a < 0, np.ceil(a), np.floor(a))
CuPy does implement where, which could be dispatched to via NEP 18. Also ceil and floor are both ufuncs, which already dispatch through the __array_ufunc__ protocol.
Closing this, since #4708 was merged already.