Bambi 0.18.0

What's Changed

Maintenance and fixes

• Fix tests and update CI to run Pixi v0.62.2 and setup-pixi v0.9.2 by @tomicapretto in #960
• Test on numba backend by @ricardoV94 in #955
• Fix horseshoe prior by @aloctavodia in #969
• Interpret refactor by @GStechschulte in #971
• Remove pps and update target params in interpret effects signatures by @GStechschulte in #977
• Update to work with last version of arviz, pymc and pytensor by @aloctavodia in #980
• Use pymc 6 by @aloctavodia in #984

Documentation

• Wording correction: a HSGP -> an HSGP by @star1327p in #963
• Render unordered lists in docstrings and fix changelog update by @tomicapretto in #964
• Fixed, Random Effects and Mundlak Machines Example by @juanitorduz in #965
• Add discrete time notebook by @NathanielF in #967
• Minor correction to documentation by @star1327p in #975

New Contributors

• @ricardoV94 made their first contribution in #955

Full Changelog: 0.17.1...0.18.0

Bambi 0.17.2

Patch release since the previous one failed :')

The sparse bambino

What's Changed

New features

• Support sparse design matrix for group specific effects by @tomicapretto in #950
• Add SPARSE_DOT configuration variable by @tomicapretto in #950

Maintenance and fixes

• Add interlinks by @tomicapretto in #946
• Begin migration to "new arviz" by @aloctavodia in #945
• Make quartodoc build verbose and copy objects.json to _site by @tomicapretto in #949
• Fix non-deterministic predictions when sample_new_groups=True and random_seed is set by @Delogon in #954
• Update how data is read from figshare by @tomicapretto in #956
• Make twine verbose by @tomicapretto in 641a938

Documentation

• Correct typos in Polynomial Regression notebook by @star1327p in #940
• Updated Hierarchical Linear Regression (Pigs dataset) example (#484) by @pranavduraisamy in #939
• Add entry in the example gallery linking to kulprit documentation by @aloctavodia in #944
• Correct grammar issues regarding a/an usage by @star1327p in #952

Deprecation

• Remove tan_2 link function for vonmises by @aloctavodia in #943

New Contributors

• @pranavduraisamy made their first contribution in #939
• @Delogon made their first contribution in #954

Full Changelog: 0.16.0...0.17.0

The awakening bambino

What's Changed

• Fix broken links by @B-Deforce in #866
• Fix random seed handling for "vi" by @B-Deforce in #869
• typo fix -> SETTINGS_FAMILIES to BUILTIN_FAMILIES by @Schefflera-Arboricola in #873
• Add post-release.yml to update changelogs automatically by @rohanbabbar04 in #882
• Adjustments to the negative binomial tutorial by @tomicapretto in #890
• Changing to nutpie sampler on mr p example by @NathanielF in #895
• Correct a name accent in the LaTeX citation by @star1327p in #897
• Add random_seed argument to predict by @aloctavodia in #892
• Add link to the R package modelr by @star1327p in #899
• Add R2 by @aloctavodia in #896
• Update the link for "Keep it Maximal" paper by @star1327p in #901
• Add link to the book "Causal Inference: What if" by @star1327p in #902
• Updated links to PyMC and Bambi pages by @star1327p in #903
• Fix data links to fish and marginaleffects by @star1327p in #904
• Plot predictions categorical error by @GStechschulte in #905
• DOC: Update three URLs in the Jupyter Notebooks by @star1327p in #909
• Added the link to Strack RRR paper by @star1327p in #910
• DOC Updated links in Robust Linear Regression by @star1327p in #913
• DOC: Correct a few typos in the API Reference by @star1327p in #915
• DOC: Clean up a few Bambi Example Notebooks by @star1327p in #916
• DOC: Continue cleaning up the notebooks by @star1327p in #917
• Remove bayeux for accessing alternative sampler backends by @GStechschulte in #919
• Drop Python 3.10 add 3.13 by @aloctavodia in #922
• DOC: Correct more typos in the Jupyter Notebooks by @star1327p in #924
• Use mock sampling in tests and incorporate other improvements to the test suite. by @tomicapretto in #923
• Modernize development infraestructure by @tomicapretto in #925
• Update docs by @tomicapretto in #926
• Fix build status badge by @tomicapretto in #929
• DOC: Make a long equation into two lines by @star1327p in #931
• DOC: Correct typos in the narrative of Examples by @star1327p in #933
• Double-check existing docstrings + start preparing release by @tomicapretto in #935
• DOC: Improve the formatting of Bambi examples by @star1327p in #936
• Prepare release by @tomicapretto in #938

New Contributors

• @B-Deforce made their first contribution in #866
• @Schefflera-Arboricola made their first contribution in #873
• @rohanbabbar04 made their first contribution in #882

Full Changelog: 0.15.0...0.16.0

Release 0.15.0

New features

• Add default priors for binomial and bernoulli families with logit link (#830)
• Add horseshoe prior (#836)
• Handle multivariate responses with HSGP (#856)

Maintenance and fixes

• Change the JAX random number generator key for 32 bit systems (#833)
• Change rename to replace in pre-render.py (#843)
• Fix out of sample prediction for multivariate families. It would not work for tables where the
  number of rows was different from the one used to fit the model (#847)
• Check variables before trying to access them in posterior predictive sampling (#851)
• Pass kwargs to nutpie + create env.yml file (#855)

Documentation

• Fix typos and incomplete doc strings (#765)
• Clarify elpd differences interepretation (#825)
• Fix the contributing readme link (#837)
• Add example using offset (#842)
• Fix model formula in negative binomial notebook (#859)
• Fix formatting in t-test examples (#861)
• Fix issue 812 Broken link (#862)
• Update repository documentation files (#865)

Release 0.14.0

New features

• Add configuration facilities to Bambi (#745)
• Interpet submodule now outputs informative messages when computing default values (#745)
• Bambi supports weighted responses (#761)
• Bambi supports constrained responses (#764)
• Implement compute_log_likelihood() method to compute the log likelihood on a model (#769)
• Add a class InferenceMethods that allows users to access the available inference methods and kwargs (#795)

Maintenance and fixes

• Fix bug in predictions with models using HSGP (#780)
• Fix get_model_covariates() utility function (#801)
• Use pm.compute_deterministics() to compute deterministics when bayeux based samplers are used (#803)
• Wrap all the parameters of the response distribution (the likelihood) with a pm.Deterministic (#804)
• Keep bayeux-ml as the single direct JAX-related dependency (#804)
• The response component only holds response information about the response, not about predictors of the parent parameter (#804)
• Resolve import error associated with bayeux (#822)

Documentation

• Our Code of Conduct now includes how to send a report (#783)
• Add polynomial regression example (#809)
• Add Contact form to our webpage (#816)

Deprecation

• f"{response_name}_obs" has been replaced by "__obs__" as the dimension name for the observation index (#804)
• f"{response_name}_{parameter_name}" is no longer the name for the name of parameters of the likelihood. Now Bambi uses "{parameter_name}" (#804)
• kind in Model.predict() now use "response_params" and "response" instead of "mean" and "pps" (#804)
• include_mean has been replaced by include_response_params in Model.fit() (#804)

Bambi 0.13.0

This is the first version of Bambi that is released with a Governance structure. Added in #709.
The highlights are the shiny interpret subpackage and the implementation of support for censored models.

New features

• Bambi now supports censored responses (#697)
• Implement "exponential" and "weibull" families (#697)
• Add "kidney" dataset (#697)
• Add interpret submodule (#684, #695, #699, #701, #732, #736)
  • Implements comparisons, predictions, slopes, plot_comparisons, plot_predictions, and plot_slopes
• Support censored families

Maintenance and fixes

• Replace univariate_ordered with ordered (#724)
• Add missing docstring for center_predictors (#726)
• Fix bugs in plot_comparison (#731)

Documentation

• Add docstrings to utility functions (#696)
• Migrate documentation to Quarto (#712)
• Add case study for MRP (#716)
• Add example about ordinal regression (#719)
• Add example about zero inflated models (#725)
• Add example about predictions for new groups (#734)

Deprecation

• Drop official suport for Python 3.8 (#720)
• Change plots submodule name to interpret (#705)

Bambi 0.12.0: Ordinal models and predictions on new groups

0.12.0

New features

• Implement new families "ordinal" and "sratio" for modeling of ordinal responses (#678)
• Allow families to implement a custom create_extra_pps_coord() (#688)
• Allow predictions on new groups (#693)

Maintenance and fixes

• Robustify how Bambi handles dims (#682)
• Fix links in FAQ (#686)
• Update additional dependencies install command (#689)
• Update predict pps docstring (#690)
• Add warning for aliases athat aren't used (#691)

Documentation

• Add families to the Getting Started guide (#683)

Bambi 0.11.0: The family grows

0.11.0

New features

• Add support for Gaussian Processes via the HSGP approximation (#632)
• Add new families: "zero_inflated_poisson", "zero_inflated_binomial", and "zero_inflated_negativebinomial" (#654)
• Add new families: "beta_binomial" and "dirichlet_multinomial" (#659)
• Allow plot_cap() to show predictions at the observation level (#668)
• Add new families: "hurdle_gamma", "hurdle_lognormal", "hurdle_negativebinomial", and "hurdle_poisson" (#676)

Maintenance and fixes

• Modify how HSGP is built in PyMC when there are groups (#661)
• Modify how Bambi is imported in the tests (#662)
• Prevent underscores from being removed in dim names (#664)
• Bump sphinx dependency to a version greater than 7 (#672)

Documentation

• Document how to use custom priors (#656)
• Fix name of arviz traceplot function in the docs (#666)
• Add example that shows how plot_cap() works (#670)

Bambi 0.10.0

New features

• Refactored the codebase to support distributional models (#607)
• Added a default method to handle posterior predictive sampling for custom families (#625)
• plot_cap() gains a new argument target that allows to plot different parameters of the response distribution (#627)

Maintenance and fixes

• Moved the tests directory to the root of the repository (#607)
• Don't pass dims to the response of the likelihood distribution anymore (#629)
• Remove requirements.txt and replace with pyproject.toml config file to distribute the package (#631)

Documentation

• Update examples to work with the new internals (#607)
• Fixed figure in the Sleepstudy example (#607)
• Add example using distributional models (#641)

Deprecation

• Removed versioned documentation webpage (#616)
• Removed correlated priors for group-specific terms (#607)
• Dictionary with tuple keys are not allowed for priors anymore (#607)