Evgeni Burovski

Ticked off Riccati equation solvers in the OP, based on https://github.com/scipy/scipy/pull/20335 having removed the xfails. We might want to revisit the tolerance bumps at some point after openblas 3.26 becoming...

> Ah okay, so does all of _lfilter.c.in need to stay? Indeed. Thank you Jake for chasing down where this dead code is despite the typo in the original PR...

This is an interesting fallout :-). Basically, this page should be static. And well discoverable, esp after interp2d starts raising hard errors. Whether it's back to ReST or link to...

cc @melissawm for vis

While we're at it, it'd be great to also address https://github.com/scipy/scipy/issues/20620. Can do in a follow-up though.

One other potentially tangentially related thing while we're at the subject of supporting other arrays types. There is an ongoing effort to provide a clone of scipy.signal API in cupy,...

https://github.com/data-apis/array-api/pull/862 has landed, so this issue can be closed --- unless there's something else to track to someone that I'm not seeing.

From experience adding array api support to `scipy.ndimage` and `scipy.signal`: - ignoring `None`s is a big +1. As mentioned, scipy has a helper anyway, and it does not seem risky...

Tried it locally and it does work as expected. Would be great to be able to use this downstream. So, for what it's worth, LGTM.

https://data-apis.org/array-api-compat/supported-array-libraries.html#pytorch now mentions that > Slices do not support negative steps. And from previous interactions with pytorch devs, my impression was that the assumption that slice steps are non-negative is...