Ralf Gommers
Ralf Gommers
Thanks @neutrinoceros! @rdbisme, I suggest saying yes to @neutrinoceros, I'm familiar with his work in AstroPy and h5py and trust he'll use a commit bit responsibly.
Thanks @rdbisme. It seems like I have the permissions needed to invite people, so I took the liberty of sending @neutrinoceros an invite with maintainer permissions just now.
Agreed, -1 on pre-commit
Zero context given, just a push with that notice after recent activity: Let's see what happens on https://github.com/gauge-sh/tach/issues/791 first.
A note for those who, like me, weren't familiar with this environment variable. It is set at build time and is the documented way to retrieve Python's site-packages location: https://docs.conda.io/projects/conda-build/en/stable/user-guide/environment-variables.html#environment-variables-set-during-the-build-process....
@clin1234 I think this is covered by gh-1367 and hence this issue can be closed as a duplicate. As gh-1367 mentions, you need the fork of `cffi` linked from that...
> I saved the first 10,000 values from each CSV into a `.npz` file with fields `"samples"` and `"pdfs`. My unit tests use that `.npz` file. Let me know if...
A new PyWavelets release with Python 3.12 support is up on PyPI.
> Instead of viewing this discussion as introducing a breaking change, I argue this is a NumPy bug (value-dependent casting) that should have been fixed by NumPy 2.0. The second...
@ngoldbaum I don't understand all of that summary. There seems to be a conflict between: > That way people who are working with complex numbers already don't have to do...