v0.0.8

What's Changed

• Updates for new valued/bound variables in AePPL by @brandonwillard in #91
• Fix handling of constant inputs to SubsumingElemwise by @brandonwillard in #93
• Update pre-commit versions by @rlouf in #97
• Set up documentation by @rlouf in #95
• Update the README by @rlouf in #98
• Add Gitter badge and links to README by @rlouf in #100
• Add link to the roadmap in the documentation by @rlouf in #104
• Update CI setup and remove Python 3.7 by @brandonwillard in #94
• Add negative-binomial conjugate rewrite by @xjing76 in #105
• Update pre-commit hooks by @github-actions in #111
• Updates for new Aesara deprecations and AePPL version by @brandonwillard in #110

Full Changelog: v0.0.7...v0.0.8

v0.0.7

What's Changed

• Add closed-form posterior for gamma-poisson observation model by @rlouf in #58
• Update ShapeFeature imports by @brandonwillard in #55
• Fix the mathematical expression for the loc-scale transform by @rlouf in #60
• Add the transformation between the inverse gamma and the exponential distributions by @rlouf in #67
• Sample remaining variables with the NUTS sampler by @rlouf in #68
• Make construct_sampler available in the aemcmc namespace by @rlouf in #72
• Update pre-commit, black, flake8, isort, mypy versions in pre-commit hooks by @rlouf in #74
• Add Student's t distribution to the location-scale family by @rlouf in #75
• Update NUTS to work with AePPL's new joint_logprob interface by @rlouf in #78
• Add examples to the README by @rlouf in #82
• Add support for nightly builds by @rlouf in #83
• Encapsulate the sampling steps using OpFromGraph by @rlouf in #76
• Update README with new construct_sampler interface by @rlouf in #87
• Update copyright years and author info in setup.py by @brandonwillard in #92

Full Changelog: v0.0.6...v0.0.7

v0.0.6

What's Changed

• Change use of "opt[imize]" to "rewrite" by @brandonwillard in #53

Full Changelog: v0.0.5...v0.0.6

v0.0.5

What's Changed

• Make horseshoe_posterior take and return non-reciprocal lambda and tau values by @brandonwillard in #51

Full Changelog: v0.0.4...v0.0.5

v0.0.4

What's Changed

• Use the rule notation in docstring by @rlouf in #39
• Make deterministic_vars and updates optional in ModelInfo by @rlouf in #41
• Add relation for recentering and rescaling by @rlouf in #31
• Extend unittests for some distributions. by @zoj613 in #42
• Add a forward-filter backward-sampler by @brandonwillard in #44
• Add the NUTS sampler constructor by @rlouf in #35
• Add a function that constructs samplers by @brandonwillard in #45

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

v0.0.3

What's Changed

• Fix flaky tests by @rlouf in #32
• Add relation for the Beta-Binomial observation model by @rlouf in #29
• Add some general utilities and a README by @brandonwillard in #34
• Add a Gibbs sampler for the negative-binomial dispersion parameter by @xjing76 in #28

New Contributors

• @brandonwillard made their first contribution in #34
• @xjing76 made their first contribution in #28

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

v0.0.2

What's Changed

• Add a Gibbs sampler for the Horseshoe prior and a Polya-gamma normal scale-mixture expansion by @zoj613 in #1
• Document the Negative binomial gibbs sampler by @rlouf in #9
• Add mypy to pre-commit checks by @rlouf in #15
• Bump black version to 22.3.0 in pre-commit by @rlouf in #18
• Add a Gibbs sampler for the regularized logistic regression by @rlouf in #13
• Remove Python 3.6 and add 3.10 to CI test matrix by @rlouf in #21
• Remove the intercept assumption in Gibbs samplers by @rlouf in #23
• Build nbinom_horseshoe_gibbs sampler from the graph by @rlouf in #25

New Contributors

• @zoj613 made their first contribution in #1
• @rlouf made their first contribution in #9

Full Changelog: https://github.com/aesara-devs/aemcmc/commits/v0.0.2