v0.13.0

v0.13.0 (2022 Oct 22)

Release highlights:

New features

• Add side argument to plot_violin to allow single-sided violin plots (1996)
• Added support for Bean Machine via the function from_beanmachine. (2107)
• Add support for warmup samples in from_pystan for PyStan 3. (2132)
• New arviz.extract function.

Maintenance and fixes

• Addressed netCDF writing issues. (2129)

Deprecation

• Removed fill_last, contour and plot_kwargs arguments from plot_pair function (2085)

See full CHANGELOG.md file for a detailed release changes description.

Beta release

v0.13.0rc1 (2022 Oct 22)

Release highlights:

New features

• Add side argument to plot_violin to allow single-sided violin plots (1996)
• Added support for Bean Machine via the function from_beanmachine. (2107)
• Add support for warmup samples in from_pystan for PyStan 3. (2132)
• New arviz.extract function.

Maintenance and fixes

• Addressed netCDF writing issues. (2129)

Deprecation

• Removed fill_last, contour and plot_kwargs arguments from plot_pair function (2085)

See full CHANGELOG.md file for a detailed release changes description.

Beta release

Version 0.12.1 introduces a new argument to arviz.summary: stat_focus to choose between using the mean and highest density intervals or the median and equal-tail intervals as summaries. It also has several doc improvements and bug fixes.

See detailed changelog

Beta release

Change log

v0.12.0 (2022 Mar 23)

New features

• Add new convenience function arviz.extract_dataset (1725)
• Add combine_dims argument to several functions (1676)
• [experimental] Enable dask chunking information to be passed to InferenceData.from_netcdf with regex support (1749)
• Allow kwargs to customize appearance of the mean in plot_lm
• Add dict option to from_cmdstan log_likelihood parameter (as in from_pystan)
• Unify model comparison API. Both plot_compare, plot_elpd can now take dicts of InferenceData or ELPDData (1690)
• Change default for rcParam stats.ic_pointwise to True (1690)
• Add new plot type: plot_ecdf (1753)

Maintenance and fixes

• Drop Python 3.6 support (1430)
• Bokeh 3 compatibility. (1919)
• Remove manual setting of 2d KDE limits (1939)
• Pin to bokeh<3 version (1954)
• Fix legend labels in plot_ppc to reflect prior or posterior. (1967)
• Change DataFrame.append to pandas.concat (1973)
• Fix axis sharing behaviour in plot_pair. (1985)
• Fix parameter duplication problem with PyStan (1962)
• Temporarily disable pyjags tests (1963)
• Fix tuple bug in coords (1695)
• Fix extend 'right' join bug (1718)
• Update attribute handling for InferenceData (1357)
• Fix R2 implementation (1666)
• Added warning message in plot_dist_comparison() in case subplots go over the limit (1688)
• Fix coord value ignoring for default dims (2001)
• Fixed plot_posterior with boolean data (1707)
• Fix min_ess usage in plot_ess (2002)

Deprecation

Documentation

• Fixed typo in Forestplot documentation
• Restructured contributing section and added several new pages to help contributing to docs (1903)

Beta release

Patch release 0.11.4

• Fix standard deviation code in density utils

Beta release

Patch release on v0.11.2.

Many small changes. Includes code from GSOC.

Beta release

Patch release on v0.11.1.

The highlight of the release is updating the from_cmdstanpy to work with cmdstanpy>=0.9.68. The other changes are listed below

New features

• Added to_zarr and from_zarr methods to InferenceData (1518)
• Added confidence interval band to auto-correlation plot (1535)

Maintenance and fixes

• Updated from_cmdstanpy, from_cmdstan, from_numpyro and from_pymc3 converters to follow schema convention (1550, 1541, 1525 and 1555)
• Fix calculation of mode as point estimate (1552)
• Remove variable name from legend in posterior predictive plot (1559)
• Added significant digits formatter to round rope values (1569)
• Updated from_cmdstan. csv reader, dtype problem fixed and dtype kwarg added for manual dtype casting (1565)

Deprecation

• Removed Geweke diagnostic (1545)
• Removed credible_interval and include_circ arguments (1548)

Documentation

• Added an example for converting dataframe to InferenceData (1556)
• Added example for coords argument in plot_posterior docstring (1566)

---

See also detailed change log of v0.11.0

Beta release

Patch release on v0.11.1. Changes are:

Maintenance and fixes

• Fixed regression in plot_pair labels that prevented coord names to be shown when necessary (1533)
• Fixed ovelapping titles and repeating warnings on circular traceplot (1517)
• Removed repetitive variable names from forest plots of multivariate variables (1527)

Documentation

• Use tabs in ArviZ example gallery (1521)

---

See also detailed change log of v0.11.0

Beta release

Highlights

• Added to_dataframe method, __getitem__ magic and copy method to InferenceData
• Add ref_line, bar, vlines and marker_vlines kwargs to plot_rank
• Add observed argument to (un)plot observed data in plot_ppc
• Add support for named dims and coordinates with multivariate observations
• Add support for discrete variables in rank plots and loo_pit
• Make stacking the default method to compute weights in compare (1438)
• [experimental] Modified SamplingWrapper base API

Maintenance and fixes

• Updated CmdStanPy interface
• Have from_pystan store attrs as strings to allow netCDF storage
• Use circular KDE function and fix tick labels in circular plot_trace
• Fix computation of weights in compare
• Add grid argument to manually specify the number of rows and columns
• Switch to compact=True by default in our plots
• Make from_pymc3 compatible with theano-pymc 1.1.0

Deprecations

• plot_khat deprecate annotate argument in favor of threshold.

---

See detailed change log

Beta release

Highlights

• Added support for circular variables to several functions
• Added InferenceData JSON converter and to_dict function
• Added plot_separation for binary data
• Extended InferenceData capabilities for computation and combination of multiple objects.
• Ridgeplot is now hdi aware
• Add support for discrete variables in plot_bpv

Fixes

• Fixed import error with xarray 0.16.1
• Restructured code to be compatible with matplotlib>=3.3

Deprecations

• This will be the last ArviZ version to support Python 3.6. ArviZ will follow NEP 29

---

See detailed change log