Skip to content
New issue

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? Sign in to your account

DOC: 1.12.0 release notes #19628

Merged
merged 8 commits into from Dec 10, 2023
Merged

Conversation

tylerjereddy
Copy link
Contributor

@tylerjereddy tylerjereddy commented Dec 3, 2023

  • draft release notes for SciPy 1.12.0, including transcription of the wiki version of the release notes

TODO:

  • manually go through the 95 merged PRs with 1.12.0 milestone and enchancement label, to fill in several parts of the release notes that were not filled in on the wiki (this is normal--stats folks usually do a good job of filling in the wiki though)
  • fix formatting/build errors in the new release notes; nested bullet lists, indentation, the usual...
  • .mailmap fixups to get a cleaner duplication/alias-free author list
  • deal with the 20 open PRs with 1.12.0 milestone, update the release notes again if/as needed (many of these will probably be milestone bumps on stagnant or difficult PRs)
  • (after merging this): target the wheel builds back to released NumPy (not 2.x), a faster-than-normal 1.13.0 is planned for NumPy 2.0 support a few months after this 1.12.0 release
  • decide how much I/we care about having the best possible NumPy main support before we branch--I've put substantial effort into dealing with BUG: wheel runs have a *lot* of test fails at the moment. #19605, but strictly speaking, I don't think we need to do that, and things could still shift a bit anyway
  • decide on which new features/improvements qualify as the release highlights
  • add some sensible wording for "in progress, but cool/please test me" type features like array API support in some modules
  • progressively turn the CI back on here

[skip cirrus] [skip circle] [skip ci]

* draft release notes for SciPy `1.12.0`, including
transcription of the wiki version of the release notes

[skip cirrus] [skip circle] [skip ci]
@tylerjereddy tylerjereddy added Documentation Issues related to the SciPy documentation. Also check https://github.com/scipy/scipy.org needs-work Items that are pending response from the author labels Dec 3, 2023
@tylerjereddy tylerjereddy added this to the 1.12.0 milestone Dec 3, 2023
@lucascolley
Copy link
Member

Some mailmap fixups at tylerjereddy#75

@mdhaber
Copy link
Contributor

mdhaber commented Dec 4, 2023

@tylerjereddy I hadn't taken a pass on the release notes for a few months. Can I still do that or do you want changes to be made here?

@steppi
Copy link
Contributor

steppi commented Dec 4, 2023

Sorry, it looks like I didn't assign @rlucas7's first stirling2 PR, #18103, the 1.12 milestone. Only his more recent one, #19287. I just added it on the older PR now. Under scipy.special improvements, the note should be something like "Added scipy.special.stirling2 for computation of Stirling numbers of the second kind." I just checked and found the original stirling2 issue, #17890, hadn't received the 1.12 milestone either

@tylerjereddy
Copy link
Contributor Author

I hadn't taken a pass on the release notes for a few months. Can I still do that or do you want changes to be made here?

You can edit the wiki if you want, I'll check it again a few times anyway

@rlucas7
Copy link
Member

rlucas7 commented Dec 4, 2023

Sorry, it looks like I didn't assign @rlucas7's first stirling2 PR, #18103, the 1.12 milestone. Only his more recent one, #19287. I just added it on the older PR now. Under scipy.special improvements, the note should be something like "Added scipy.special.stirling2 for computation of Stirling numbers of the second kind." I just checked and found the original stirling2 issue, #17890, hadn't received the 1.12 milestone either

Thanks for catching this @steppi and thanks for adding the entry. I added 1 sentence to mention that both exact and approximate methods are supported.

[skip ci] [skip circle]
@ev-br
Copy link
Member

ev-br commented Dec 4, 2023

For scipy.interpolate:

  • New class NdBSpline represents tensor-product splines in N dimensions. This class only known how to evaluate a tensor product given coefficients and knot vectors. This way it generalizes BSpline for 1D data to N-D, and parallels NdPPoly (which represents N-D tensor product polynomials). Evaluations exploit the localized nature of b-splines.

  • NeastNDInterpolator.__call__ accepts **query_options, which are passed through to the KDTree.query call to find nearest neighbors. This allows, for instance, to limit the neighbor search distance and parallelize the query using the workers keyword.

  • BarycentricInterpolator now allows computing the derivatives.

@j-bowhay
Copy link
Member

j-bowhay commented Dec 4, 2023

Probably worth mentioning #18279 in the deprecation section. Error messages have been made clearer for things that don't exist in the public namespace and warnings sharpened for private attributes that are not supposed to be imported at all

* add `scipy.interpolate` release note entries/suggestions
from Evgeni

* also deal with wavelets deprecation, mentioned by Evgeni

* mention sharpened/clarified deprecation warnings, as suggested
by Jake

* updates from recent wiki relnotes changes

* updates from scanning through merged enhancement PRs with
appropriate milestone (incomplete)

[skip ci] [skip circle]
doc/source/release/1.12.0-notes.rst Outdated Show resolved Hide resolved
doc/source/release/1.12.0-notes.rst Outdated Show resolved Hide resolved
tylerjereddy and others added 2 commits December 5, 2023 15:36
[skip ci] [ci skip] [skip circle]

Co-authored-by: Lucas Colley <lucas.colley8@gmail.com>
Co-authored-by: Matt Haberland <mhaberla@calpoly.edu>
* more updates from manual inspection of merged
enhancement PRs

[skip ci] [ci skip] [skip circle]
@lucascolley
Copy link
Member

Feel free to modify the fft and cluster array API parts to match what Matt has written for special if anything looks like an improvement.

[ci skip] [skip ci] [skip circle]

Co-authored-by: Lucas Colley <lucas.colley8@gmail.com>
Co-authored-by: Matt Haberland <mhaberla@calpoly.edu>
@tylerjereddy
Copy link
Contributor Author

Obviously I'm a bit behind schedule, but I don't think we're too bad off, and folks have been helping a lot. So yeah, probably a little bit of a branching delay because end of year stuff made me a little busier, but not too worried I don't think.

* update release notes based on manual
inspection of merged enchancement PRs
with appropriate milestone

[skip ci] [ci skip] [skip circle]
.mailmap Outdated Show resolved Hide resolved
@tylerjereddy tylerjereddy force-pushed the treddy_112_relnotes branch 5 times, most recently from 04f4ca9 to fb527b6 Compare December 8, 2023 23:02
@tylerjereddy
Copy link
Contributor Author

I checked off the manual comb-through of the enhancement PRs and re-checking of the wiki. I believe there have been some merges since I started doing my pass, but anyway it took me a few days because of various interruptions.

I'll think a bit more about the highlights next, then fixing up the formatting/doc build.

@tylerjereddy tylerjereddy force-pushed the treddy_112_relnotes branch 3 times, most recently from 7502014 to 5f0ad4a Compare December 10, 2023 02:12
@tylerjereddy
Copy link
Contributor Author

Doc build is passing locally--let's see if CI agrees (minus cirrus for now).

Copy link
Contributor

@dschmitz89 dschmitz89 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A few minor things, mostly typos.

- `scipy.stats.jf_skew_t` implements the Jones and Faddy skew-t distribution.
- `scipy.stats.anderson_ksamp` now supports a permutation version of the test
using the ``method`` parameter.
- The ``fit`` methods of `scipy.stats.halcauchy`, `scipy.stats.halflogistic`, and
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
- The ``fit`` methods of `scipy.stats.halcauchy`, `scipy.stats.halflogistic`, and
- The ``fit`` methods of `scipy.stats.halfcauchy`, `scipy.stats.halflogistic`, and

using the ``method`` parameter.
- The ``fit`` methods of `scipy.stats.halcauchy`, `scipy.stats.halflogistic`, and
`scipy.stats.halfnorm` are faster and more accurate.
- `scipy.stats.beta` ``entropy`` accuracy has been improved for extreme values of
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
- `scipy.stats.beta` ``entropy`` accuracy has been improved for extreme values of
- ``scipy.stats.beta.entropy`` accuracy has been improved for extreme values of

- A source of intermediate overflow in `scipy.norminvgauss.pdf` has been eliminated.
- Added support for masked arrays to ``stats.circmean``, ``stats.circvar``,
``stats.circstd``, and ``stats.entropy``.
- ``gausshyper`` ``cdf`` and ``rvs`` methods have improved performance.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
- ``gausshyper`` ``cdf`` and ``rvs`` methods have improved performance.

The PR #18799 this refers to is difficult to decipher: in the end it was basically a simplifcation of a formula with no notable impact for users.

* apply inline suggestions from code review
and additional manual adjustments to the
SciPy `1.12.0` release notes

* draft in the release highlights section

* fixes needed for successful doc build
@tylerjereddy tylerjereddy changed the title WIP, DOC: 1.12.0 release notes DOC: 1.12.0 release notes Dec 10, 2023
@tylerjereddy
Copy link
Contributor Author

I still see blemishes in the rendered version of the release notes for 1.12.0, but CI is green (including ARM architectures and doc build) and I'm behind schedule on this release cycle so I'll proceed with (squash) merge and the release notes can continue to be improved during the release process (this happens pretty much every time to be honest).

@tylerjereddy tylerjereddy merged commit 20d131e into scipy:main Dec 10, 2023
28 checks passed
@tylerjereddy tylerjereddy deleted the treddy_112_relnotes branch December 10, 2023 17:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Documentation Issues related to the SciPy documentation. Also check https://github.com/scipy/scipy.org needs-work Items that are pending response from the author
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

9 participants