v0.2.1

What's Changed

Other Changes

• GHA: use dedicated docs environment by @martinfleis in #121
• TST: ignore warnings about OOB scores by @martinfleis in #122
• DOC: update examples for sklearn 1.9 by @martinfleis in #124
• Bump codecov/codecov-action from 6 to 7 in the github-actions group by @dependabot[bot] in #126
• rename package to spatialml by @martinfleis in #127

Full Changelog: v0.2.0...v0.2.1

v0.2.0

What's Changed

Bug Fixes

• BUG: fix failure of BandwidthSearch in case of an invariant y by @FirePheonix in #86
• BUG: fix batch fitting with non-default index by @martinfleis in #101

Enhancements

• Expose local_class_support_ as fitted attribute by @AshishRaj97 in #82
• Add structured fit-time input validation by @jigyasaba in #83
• ENH: expose local_metric on regressors as well by @martinfleis in #112
• Add validation for bandwidth and kernel parameters by @jigyasaba in #114

Other Changes

• REF: Add type hints to internal _get_score_data method by @GeneralAumsum07 in #71
• MAINT: update docstring and clean typing comments by @martinfleis in #75
• DOC: fix outdated BaseRegressor prediction docstring by @AshishRaj97 in #78
• FMT: Ruff format ipynb files by @FirePheonix in #81
• Remove outdated TODO block for performance metrics in parallel validation test by @jigyasaba in #89
• Bump actions/upload-artifact from 6 to 7 in the github-actions group by @dependabot[bot] in #98
• Fix inconsistent bandwidth validation for numpy scalar and non-scalar inputs by @AshishRaj97 in #94
• compute information criteria only for linear models by @martinfleis in #100
• REF: rely on sklearn wrappers around joblib Parallel by @martinfleis in #102
• DOC: use markdown instead of RST in references by @martinfleis in #104
• DOC: link to source and new libpysal location by @martinfleis in #105
• ENH: expose handling of coplanar points by @martinfleis in #106
• cancel doc builds in progress by @jGaboardi in #108
• source version info in docs from packaging by @jGaboardi in #107
• Bump the github-actions group with 2 updates by @dependabot[bot] in #109
• DOC: update config by @martinfleis in #110
• open-ended license by @jGaboardi in #111
• Bump actions/github-script from 8 to 9 in the github-actions group by @dependabot[bot] in #117
• reup SPEC000 - 2026-04-18 by @jGaboardi in #119
• TST: use linear models in tests where possible for speed up by @martinfleis in #120

New Contributors

• @GeneralAumsum07 made their first contribution in #71
• @jigyasaba made their first contribution in #83
• @dependabot[bot] made their first contribution in #98

Full Changelog: v0.1.1...v0.2.0

v0.1.1

What's Changed

Other Changes

• GHA: ignore pre-commit-ci[bot] in release notes by @martinfleis in #65
• GHA: build docs for PRs by @martinfleis in #67
• Fix strict type annotation in GWLogisticRegression by @AshishRaj97 in #70
• DOC: fix docs building by @martinfleis in #72
• COMPAT: pandas 3 compatibility by @martinfleis in #73

New Contributors

• @AshishRaj97 made their first contribution in #70

Full Changelog: v0.1.0...v0.1.1

v0.1.0

What's Changed

Other Changes

• MAINT: infrastructure enhancements by @martinfleis in #62
• DOC: make notebooks a bit faster to execute by @martinfleis in #63
• DOC: various documentation enhancements by @martinfleis in #64

Full Changelog: v0.0.4...v0.1.0

v0.0.4

Full Changelog: v0.0.3...v0.0.4

v0.0.3

What's Changed

• DOC: automatically build docs for stable (on version) and latest (on push) by @martinfleis in #59
• TST: make sure that codecov tracks undersampling by @martinfleis in #60
• ENH: add GWGradientBoostingRegressor by @FirePheonix in #57
• DOC: automatised release update by @martinfleis in #61

Full Changelog: v0.0.2...v0.0.3

v0.0.2

What's Changed

• clean up n_jobs warning by @jGaboardi in #44
• MAINT: eliminate warnings from CI by @martinfleis in #46
• ENH: support predict method on regressors by @martinfleis in #47
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in #50
• ENH: add GWRandomForestRegressor by @martinfleis in #49
• ENH: prediction based on nearest model only or a custom bandwidth by @martinfleis in #52
• ENH: implement fusion with the global model in prediction by @martinfleis in #54
• API: move geometry back to fit, implement score and backbone for metadata routing by @martinfleis in #55
• DOC: add comparison with mgwr by @FirePheonix in #51
• document and test metadata routing by @martinfleis in #56

New Contributors

• @pre-commit-ci[bot] made their first contribution in #50
• @FirePheonix made their first contribution in #51

Full Changelog: v0.0.1...v0.0.2

v0.0.1

What's Changed

• ENH: allow keeping focal in local models by @martinfleis in #4
• implement GWLinearRegression by @martinfleis in #5
• adjust kernels by @martinfleis in #6
• Compute AIC, AICc, BIC for both regression and classification to expose in bandwidth search by @martinfleis in #7
• fixup the attributes API and document it by @martinfleis in #8
• verbosity, repr fix by @martinfleis in #9
• subclass BaseEstimator by @martinfleis in #10
• ENH: allow custom Graph object by @martinfleis in #13
• pin oldest supported dependency versions & test by @jGaboardi in #15
• Allow pulling additional metrics from bandwidth search by @martinfleis in #16
• DOC: double check docstrings by @martinfleis in #17
• API: Move geometry from fit to init by @martinfleis in #18
• more granular control over perf metrics by @martinfleis in #19
• Allow bandwidth search based on a custom criterion by @martinfleis in #21
• Update build_docs.yml - bump setup-micromamba version by @jGaboardi in #22
• correct number of observations when models are left out by @martinfleis in #24
• Capture out-of-sample log loss by @martinfleis in #23
• properly report prediction rate in bandwidth search by @martinfleis in #25
• ensure bandwidth can be none with custom graph by @martinfleis in #26
• compute global metrics only for non-unique focal set by @martinfleis in #27
• fix incorrect sorting by @martinfleis in #28
• CI: replace deprecated macos-13 with macos-15-intel by @martinfleis in #29
• CI: attempt testing on Python 3.14 by @martinfleis in #31
• API: try to make sure that API for metrics is not confusing by @martinfleis in #30
• API: do not compute performance metrics, return arrays for users by @martinfleis in #32
• custom undersampling by @martinfleis in #33
• DOC: Add basic user guide by @martinfleis in #34
• use immaterial theme by @martinfleis in #35
• fix API rendering by @martinfleis in #36
• beef up docstrings by @martinfleis in #37
• fix some type hints by @martinfleis in #38
• full type hints (checked by ty) by @martinfleis in #39
• minor touch of docs by @martinfleis in #40

New Contributors

• @martinfleis made their first contribution in #4
• @jGaboardi made their first contribution in #15

Full Changelog: https://github.com/pysal/gwlearn/commits/v0.0.1