Releases: dswah/pyGAM
Releases · dswah/pyGAM
v0.12.0
What's Changed
Bug Fixes
- [BUG] fix bug where parameter callbacks is not set in LinearGAM constructor by @deniseschmitz42 in #359
- [BUG] modify order of param validation in gridsearch by @dswah in #251
- [BUG] fix incorrect scale parameter in Distribution by @dswah in #404
- [BUG] Fix error with data ownership in cholesky by @craymichael in #288
- [BUG] remove dense zeros by @dswah in #414
- [BUG] truncate SVD by @dswah in #417
- [BUG] Sparse array by @jonathan-taylor in #373
Docs
- [DOC] Update the stringdoc of FactorTerm in terms.py by @nicoleta-kyo in #286
- [DOC] Update version number to 0.11.0 in README by @fkiraly in #407
- [DOC] Fix Documentation Builds by @dswah in #405
- [DOC] Fixed docstring of GAM classes - index of intercept estimate by @utopianf in #324
- [DOC] corrections to pygam.py docstrings by @deniseschmitz42 in #361
- [DOC] Use sphinx-autodoc by @dswah in #410
- [DOC] remove notebook prompts by @dswah in #413
- [DOC] Fix typo (LogitLink --> LogLink) by @craymichael in #287
- [DOC] TensorTerm construction, Readme, Docs by @dswah in #415
New Contributors
- @nicoleta-kyo made their first contribution in #286
- @deniseschmitz42 made their first contribution in #359
- @utopianf made their first contribution in #324
- @craymichael made their first contribution in #288
- @jonathan-taylor made their first contribution in #373
All Contributors
Full Changelog: v0.11.0...v0.12.0
v0.11.0
What's Changed
- compatibility up to python 3.14
- bugfixes and minor improvements
Contents
- [DOC] replace svg TeX plugin in readme by GitHub native TeX by @fkiraly in #399
- [ENH] unit tests for
gen_imgs.pyby @Yasserelhaddar in #396 - [MNT] python 3.14 compatibility and python 3.9 end-of-life by @fkiraly in #400
- [BUG] fix missing datasets and dataset loaders crashing; also impacts quickstart tutorial by @Kraego in #401
New Contributors
- @Yasserelhaddar made their first contribution in #396
- @Kraego made their first contribution in #401
All Contributors
@fkiraly,
@Kraego,
@Yasserelhaddar
Full Changelog: v0.10.1...v0.11.0
v0.10.1
v0.10.0
Highlights
- compatibility up to python 3.13
- compatibility with most recent
scipyversions, from previouslyscipy<1.14 - compatibility with
numpy 2
What's Changed
- [MNT] Fix typos throughout the code base by @szepeviktor in #375
- [MNT] extend CI test matrix to ubuntu, macos, windows, python 3.9-3.13 by @fkiraly in #380
- [MNT] fix and upgrade linting CI job by @fkiraly in #382
- [MNT] make
pyproject.tomlpipcompatible by @fkiraly in #383 - [MNT] widen
scipydependency range by @fkiraly in #384 - [MNT] compatibility with
scipy>=1.14.0andpython 3.13by @ouslan in #366 - [MNT] additional
python 3.13compatibility actions by @Lyessi78 in #372 - [MNT] add CI concurrency exclusion for the same PR job by @fkiraly in #386
- [MNT] Apply pre-commit formatting to all files by @Shushankranjan in #389
- [MNT] 0.10.0 release,
pypirelease workflow using trusted publishers by @fkiraly in #390 - [DOC] README badges and links by @fkiraly in #391
- [DOC] fix GAM interaction term notation and broken reference links in README by @UmbertoFasci in #377
All contributors
@dswah,
@fkiraly,
@Lyessi78,
@ouslan,
@Shushankranjan,
@szepeviktor,
@UmbertoFasci
New Contributors
- @szepeviktor made their first contribution in #375
- @fkiraly made their first contribution in #380
- @ouslan made their first contribution in #366
- @Lyessi78 made their first contribution in #372
- @Shushankranjan made their first contribution in #389
- @UmbertoFasci made their first contribution in #377
Full Changelog: v0.9.1...v0.10.0
v0.9.1
v0.9.0
v0.8.0
New Features
- cyclic p-splines: you can now train models with periodic features by using the
'cp'basis like so:
GAM(s(0, basis='cp'))- factor smooths now allow dummy coding, via:
GAM(f(0, coding='dummy'))Models using this coding scheme are more statistically interpretable , and computationally less expensive than those using one-hot encodings.
Bug Fixes
- models can mix constrained terms and un-constrained tensor-terms
- tensor terms can be constrained