Ralf Gommers

follow-up actions for array API support in `cluster`

13

These are follow-up items from gh-18668: - [x] a better name for `as_xparray` (see [this comment](https://github.com/scipy/scipy/pull/18668#issuecomment-1631049650)) - [ ] a TODO comment in `hierarchy.py` for rewriting a fancy indexing expression...

Tracker: array types (CuPy, PyTorch & co) and array API standard support

10

See gh-18286 for the proposed design changes, and gh-18668 for the main PR that laid the foundations (it supported all of `scipy.cluster`, added a CI job that used both `pytorch-cpu`...

[BUG] new warning `__sync_fetch_and_sub_4` overflows destination

5

### Describe the bug This warning is new in Cython 3 I believe: ``` [213/220] Compiling C object scipy/interpolate/_ppoly.cpython-310-x86_64-linux-gnu.so.p/meson-generated__ppoly.c.o In function '__Pyx_XCLEAR_MEMVIEW', inlined from '__pyx_pf_5scipy_11interpolate_6_ppoly_20evaluate_nd' at scipy/interpolate/_ppoly.cpython-310-x86_64-linux-gnu.so.p/_ppoly.c:22469:5, inlined from '__pyx_fuse_0__pyx_pw_5scipy_11interpolate_6_ppoly_21evaluate_nd'...

[BUG]: latest pybind11 (2.11.1) version not supporting numpy 2.0

4

### Required prerequisites - [X] Make sure you've read the [documentation](https://pybind11.readthedocs.io). Your issue may be addressed there. - [X] Search the [issue tracker](https://github.com/pybind/pybind11/issues) and [Discussions](https:/pybind/pybind11/discussions) to verify that this hasn't...

HiGHS added a runtime dependency on `libatomic` on 32-bit platforms

4

We observed this in SciPy, after updating to a recent version of HiGHS, and then receiving reports of broken installs on 32-bit armv7 and 32-bit PowerPC. There is fairly heavy...

Only the default (partial pivoting) algorithm that is implemented in all libraries and for all devices is added here. gh-627 has details on the no-pivoting case, but it's not universally...

Need for dealing with strides when writing performant code against multiple array libraries

8

Early on when putting the array API standard together, we made a decision that strides must not be exposed in the Python API, because multiple array libraries don't support that...

RFC: static vs. dynamic shapes and JAX's `.at` for simulating in-place ops

7

_This is a continuation of a discussion that started a few weeks ago in gh-597 (Cc @soraros). It is closely related to gh-84 (boolean indexing) and gh-24 (mutability and copy/views)._...

Adding html docs for the project

10

I'd like to propose and work on adding html documentation for the project. This would make it easier for users - both direct C/Fortran API users and downstream users who...

Support for "virtual packages" / dependencies on interfaces?

1

Hi, thanks for all your work on PURL, it looks excellent. I have a question/proposal, but let me first give some context on why I'm asking. Right now I am...