What's Changed

New Features 🎉

• Add Poisson and NegativeBinomial to pymc.dims by @williambdean in #8305

Bugfixes 🪲

• Eagerly clean rv shape references introduced by maybe_resize by @ricardoV94 in #8302
• Fix marimo progress bar issues by @ricardoV94 in #8298

New Contributors

• @mcwanza made their first contribution in #8107

Full Changelog: v6.0.0...v6.0.1

What's Changed

See the announcement post for an overview of what's new in PyMC 6.0.

Major Changes 🛠

• Depend on PyTensor 3.0 major release by @ricardoV94 in #8199
  • Default backend is now numba
    • set pytensor.config.linker = "cvm" to return to the old default.
  • Several imports have changed location and already deprecated functions were removed
  • Low level changes that shouldn't impact most users:
    • Autodiff L_op and R_op methods are now deprecated in favor of pull_back and push_forward
    • test_value machinery is deprecated
    • low-level random API changed
• PyMC is now safely pip install-able
• Make nutpie the default nuts sampler (if installed) by @ricardoV94 in #8248
  • pip install pymc[nutpie] is now available
• Use arviz 1.0 by @aloctavodia in #8019
  • arviz.InferenceData replaced by xarray.DataTree
  • plot_trace replaced by plot_trace_dist (although plot_rank_dist is recommended instead)
  • default credible interval changed from 0.94 highest-density interval to 0.89 equal-tailed interval
  • InferenceData.extend replaced by DataTree.update
  • See the full migration guide for more details
• Allow samplers to choose default number of tuning steps by @ricardoV94 in #8289
  • Default pymc nuts sampler remains at 1000 tuning steps. Nutpie defaults to 400. Set tune explicitly to control it.
• Change sample_posterior_predictive API wrt to volatility by @ricardoV94 in #8209
  • Introduce sample_vars and freeze_vars argument to explicitly control which variables are resampled and which are reused from trace
  • Variables that are deemed "volatile" because they depend on data that has changed or variables being sampled will issue a warning instead of being resampled by default. Users should assign these variables to either sample_vars or freeze_vars to suppress the warning.
  • var_names has no effect on sampling semantics and only controls what variables are saved in the trace.
  • See docstrings for detailed examples.
• Remove deprecated samples argument in sample_prior_predictive by @williambdean in #8204
• Remove experimental warning from dims module by @ricardoV94 in #8137
• Remove several functions and objects from PyMC root namespace by @ricardoV94 in #6973

New Features 🎉

• Richer textual representation: show Data variables, fold constants, and use semantic separators by @drbenvincent in #8205
• Add Model.table() method by @fonnesbeck in #8254
• Implement correct unconstraining transform for LKJCorr by @johncant in #7380
• Implement unconstraining transform for Wishart by @ricardoV94 in #8246
• Add logccdf to LogNormal distribution by @vybhav72954 in #8166
• Apply PyMC brand theme to Model.display mermaid diagram by @williambdean in #8294

Bugfixes 🪲

• Validate coordinate and variables don't have the same names by @Chirag3841 in #8093

Documentation 📖

• compile_pymc -> compile by @williambdean in #8168
• docs: fix RST in Model.debug docstring numbered list by @anevolbap in #8281
• Updated pymc overview notebook to v6 by @fonnesbeck in #8284
• Improve HalfFlat class docstring by @KRYSTALM7 in #8211

Maintenance 🔧

• Support blackjax 1.5 progress bars by @zaxtax in #8241
• VI Approximation's sample method uses model contexts by @zaxtax in #7940
• Log external sampler and variables by @ricardoV94 in #8275
• Speedup PyMC import by @ricardoV94 in #8276
• compute_log_prior: make subsequent arguments keyword only by @vkverma9534 in #8123

New Contributors

• @vkverma9534 made their first contribution in #8123
• @drbenvincent made their first contribution in #8205
• @vybhav72954 made their first contribution in #8166
• @johncant made their first contribution in #7380
• @KRYSTALM7 made their first contribution in #8211
• @MukundaKatta made their first contribution in #8251
• @anevolbap made their first contribution in #8281

Full Changelog: v5.28.0...v6.0.0

What's Changed

Bugfixes 🪲

• Delay attaching ShapeFeature to logprob rewrites by @ricardoV94 in #8274

Maintenance 🔧

• Reject variable names with / by @Chirag3841 in #8256

Full Changelog: v5.28.4...v5.28.5

What's Changed

Maintenance 🔧

• Throttle marimo progress bars by @ricardoV94 in #8242

Full Changelog: v5.28.3...v5.28.4

What's Changed

Bugfixes 🪲

• Fix off-by-one progress bar bugs by @jessegrabowski in #8201
• Fix crash in vectorize_over_posterior when using ZeroSumNormal distributions by @Chirag3841 in #8220

Maintenance 🔧

• Clean up marimo progress bar by @fonnesbeck in #8184

New Contributors

• @Chirag3841 made their first contribution in #8220

Full Changelog: v5.28.2...v5.28.3

What's Changed

Bugfixes 🪲

• Fix broadcast check on log_jac_det by @velochy in #8174
• Remove tune stat from steps and fix non-discarding of tuning draws from trace by @eclipse1605 in #8015

Maintenance 🔧

• Stream progress bar to stderr by @williambdean in #8117
• Emit scalar DEMetropolis "scaling" stat by @michaelosthege in #8178
• Fix Wishart incompatibility with newer scipy by @ricardoV94 in #8179
• Better str_repr for models with dims variables by @ricardoV94 in #8170

Full Changelog: v5.28.1...v5.28.2

What's Changed

New Features 🎉

• Implement dims.Censored by @ricardoV94 in #8133

Maintenance 🔧

• str_repr: Accept non tensor-variables by @ricardoV94 in #8124
• Pin upper bound arviz by @aloctavodia in #8125
• compute_log_prior: make subsequent arguments keyword only by @vkverma9534 in #8123
• Fix logp shape of transformed ZeroSumNormal and dims.ZeroSumNormal with batch sigma by @ricardoV94 in #8136

New Contributors

• @vkverma9534 made their first contribution in #8123

Full Changelog: v5.28.0...v5.28.1

What's Changed

New Features 🎉

• observe(sum) for Normal via rewrite to equivalent NormalRV by @eclipse1605 in #8067
• Implement dims.Weibull by @ricardoV94 in #8081
• Support logcdf, logccdf and icdf for dim RVs by @ricardoV94 in #8083
• SMC Multiprocessing and Progress Bar Refactor by @jessegrabowski in #8047

Maintenance 🔧

• add action yml to populate issue by @williambdean in #7807
• Broadcast probability functions with value and size by @ricardoV94 in #8105
• dims.Data creates XTensorSharedVariables by @ricardoV94 in #8096

Full Changelog: v5.27.1...v5.28.0

What's Changed

New Features 🎉

• Added quiet argument to sample() to suppress all output. by @fonnesbeck in #8023
• Implement logccdf helper for numerically stable log survival function by @maresb in #7996
• Add logprob support for leaky-ReLU switch transforms by @eclipse1605 in #7995
• Allow constant branches in logprob of switch/join/make_vector by @eclipse1605 in #8008
• Implement Uniform and TruncatedNormal dims distributions by @ricardoV94 in #8065

Documentation 📖

• Add Docstrings for pymc.dims.distributions.scalar by @MSK-005 in #8041
• Update links and remove reference to pymc 4 by @aloctavodia in #8068

Maintenance 🔧

• Fix NaN gradients in Hurdle distributions for continuous components by @pirzada-ahmadfaraz in #8057
• Only use threadpool_limits when start method is not fork by @fonnesbeck in #8044
• Adapt progress bars for use in marimo by @fonnesbeck in #8055
• Clarify dims error message to reference as_xtensor by @AdishreeB in #8066
• Updated pymc.distributions.transforms docstring by @Cassiebastress in #8028

New Contributors

• @jyotihotwani25 made their first contribution in #8031
• @eclipse1605 made their first contribution in #7995
• @pirzada-ahmadfaraz made their first contribution in #8057
• @MSK-005 made their first contribution in #8041
• @AdishreeB made their first contribution in #8066
• @Jiya873 made their first contribution in #8006
• @Cassiebastress made their first contribution in #8028

Full Changelog: v5.27.0...v5.27.1

What's Changed

Major Changes 🛠

• Bump PyTensor dependency and support Python 3.14 by @ricardoV94 in #8010
• Replace VarName with built-in str by @osyuksel in #7855

New Features 🎉

• Automatically derive icdf for negative power transforms by @ricardoV94 in #7956
• Added icdf for HalfStudentT by @dalai444 in #7919
• Add ICDF for the Kumaraswamy distribution by @gokuld in #6642
• Add power_spectral_density to RatQuad covariance kernel for HSGP support by @fonnesbeck in #7973
• Adding Dirichlet and SimplexTransform to pymc.dims by @zaxtax in #7921

Bugfixes 🪲

• Fix independent_rvs determination in vectorize_over_posterior by @lucianopaz in #7890

Documentation 📖

• Remove MutableData by @sotagg in #7900
• Use correct coords in Model docstring by @williambdean in #7970
• Clarify ZeroSumNormal covariance documentation by @swarnaliiiiii in #7988
• Readme updates by @OriolAbril in #7941

Maintenance 🔧

• Add support for sample_stats group in pymc.testing.mock_sample by @williambdean in #7887
• Fix print name of LogitNormal and DiscreteWeibull by @aloctavodia in #7935
• Change s/draws to s/draw by @premshaw04 in #7953
• Import math module in root by @williambdean in #7965

New Contributors

• @Copilot made their first contribution in #7928
• @sotagg made their first contribution in #7900
• @premshaw04 made their first contribution in #7953
• @dalai444 made their first contribution in #7919
• @immortal71 made their first contribution in #7962
• @swarnaliiiiii made their first contribution in #7988

Full Changelog: v5.26.1...v5.27.0