PyMC3 Files

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 https://github.com/pymc-devs/pymc/pull/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 https://github.com/pymc-devs/pymc/pull/8248
• pip install pymc[nutpie] is now available
• Use arviz 1.0 by @aloctavodia in https://github.com/pymc-devs/pymc/pull/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 https://github.com/pymc-devs/pymc/pull/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 https://github.com/pymc-devs/pymc/pull/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 https://github.com/pymc-devs/pymc/pull/8204
• Remove experimental warning from dims module by @ricardoV94 in https://github.com/pymc-devs/pymc/pull/8137
• Remove several functions and objects from PyMC root namespace by @ricardoV94 in https://github.com/pymc-devs/pymc/pull/6973

New Features 🎉

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

Bugfixes 🪲

• Validate coordinate and variables don't have the same names by @Chirag3841 in https://github.com/pymc-devs/pymc/pull/8093

Documentation 📖

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

Maintenance 🔧

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

New Contributors

• @vkverma9534 made their first contribution in https://github.com/pymc-devs/pymc/pull/8123
• @drbenvincent made their first contribution in https://github.com/pymc-devs/pymc/pull/8205
• @vybhav72954 made their first contribution in https://github.com/pymc-devs/pymc/pull/8166
• @johncant made their first contribution in https://github.com/pymc-devs/pymc/pull/7380
• @KRYSTALM7 made their first contribution in https://github.com/pymc-devs/pymc/pull/8211
• @MukundaKatta made their first contribution in https://github.com/pymc-devs/pymc/pull/8251
• @anevolbap made their first contribution in https://github.com/pymc-devs/pymc/pull/8281

Full Changelog: https://github.com/pymc-devs/pymc/compare/v5.28.0...v6.0.0