Evgeni Burovski

Thanks! Fix in https://github.com/scipy/scipy/pull/23350

> docs pages still need doing This PR or a follow-up?

The CI failure is curious. LAPACK 3.12 documentation for `?gesdd` mentions that a `nan` entry in the input matrix results in `info = -4` (https://www.netlib.org/lapack/explore-html/df/d22/group__gesdd_ga8941e5ff50de36580dae8940015e9cb0.html), and we test for it....

How does this compare to the current scipy version, with a decorator and `_delegators.py`? I guess I'm missing some context---starting with what are the problems this aims to solve?

Thanks, it's very helpful! So this is meant to allow registering new implementations of (parts of) SciPy API instead of hardcoding `cupyx.scipy` and `jax.scipy`, as it currently is, correct? This...

I'm still not entirely clear on who does what, am probably just being dense. There seems to be at least four moving parts (scipy, spatch, a backend implementation, and a...

> That is unless anyone has concerns I'd like to take a closer look at this PR, and did not have a chance manage to do it yet.

One more possible GSoC topic? How does the difficulty compare to a GSoC project scope? On Tue, Feb 28, 2017 at 10:30 PM, Robert Kern wrote: > ODRPACK was indeed...

Any chance you'd be interested in writing a project description / mentoring / co-mentoring it if there's a student? Fits pretty well, I think, for a student that's comfortable implementing...

Is cython not the least magical? I'd suggest the default to be a thin cython wrapper, with either no functionality at all (or maybe handling memory allocations for out and...