Arviz

New features

- Add `stat_focus` argument to `arviz.summary` ([1998](https://github.com/arviz-devs/arviz/pull/1998))

Maintenance and fixes

- `psislw` now smooths log-weights even when shape is lower than `1/3`([2011](https://github.com/arviz-devs/arviz/pull/2011))
- Fixes `from_cmdstanpy`, handles parameter vectors of length 1 ([2023](https://github.com/arviz-devs/arviz/pull/2023))
- Fix typo in `BaseLabeller` that broke `NoVarLabeller` ([2018](https://github.com/arviz-devs/arviz/pull/2018))

Documentation

- Adding plotting guides ([2025](https://github.com/arviz-devs/arviz/pull/2025))
- Update links to use new domain ([2013](https://github.com/arviz-devs/arviz/pull/2013))

New features

- Add new convenience function `arviz.extract_dataset` ([1725](https://github.com/arviz-devs/arviz/pull/1725))
- Add `combine_dims` argument to several functions ([1676](https://github.com/arviz-devs/arviz/pull/1676))
- [experimental] Enable dask chunking information to be passed to `InferenceData.from_netcdf` with regex support ([1749](https://github.com/arviz-devs/arviz/pull/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](https://github.com/arviz-devs/arviz/pull/1690))
- Change default for rcParam `stats.ic_pointwise` to True ([1690](https://github.com/arviz-devs/arviz/pull/1690))
- Add new plot type: plot_ecdf ([1753](https://github.com/arviz-devs/arviz/pull/1753))

Maintenance and fixes

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

Deprecation

Documentation

- Fixed typo in `Forestplot` documentation
- Restructured contributing section and added several new pages to help contributing to docs ([1903](https://github.com/arviz-devs/arviz/pull/1903))

Maintenance and fixes

- Fix standard deviation code in density utils by replacing it with `np.std`. ([1833](https://github.com/arviz-devs/arviz/pull/1833))

New features

- Change order of regularization in `psislw` ([1943](https://github.com/arviz-devs/arviz/pull/1943))
- Added `labeller` argument to enable label customization in plots and summary ([1201](https://github.com/arviz-devs/arviz/pull/1201))
- Added `arviz.labels` module with classes and utilities ([1201](https://github.com/arviz-devs/arviz/pull/1201) and [1605](https://github.com/arviz-devs/arviz/pull/1605))
- Added probability estimate within ROPE in `plot_posterior` ([1570](https://github.com/arviz-devs/arviz/pull/1570))
- Added `rope_color` and `ref_val_color` arguments to `plot_posterior` ([1570](https://github.com/arviz-devs/arviz/pull/1570))
- Improved retrieving or pointwise log likelihood in `from_cmdstanpy`, `from_cmdstan` and `from_pystan` ([1579](https://github.com/arviz-devs/arviz/pull/1579) and [1599](https://github.com/arviz-devs/arviz/pull/1599))
- Added interactive legend to bokeh `forestplot` ([1591](https://github.com/arviz-devs/arviz/pull/1591))
- Added interactive legend to bokeh `ppcplot` ([1602](https://github.com/arviz-devs/arviz/pull/1602))
- Add more helpful error message for HDF5 problems reading `InferenceData` from NetCDF ([1637](https://github.com/arviz-devs/arviz/pull/1637))
- Added `data.log_likelihood`, `stats.ic_compare_method` and `plot.density_kind` to `rcParams` ([1611](https://github.com/arviz-devs/arviz/pull/1611))
- Improve error messages in `stats.compare()`, and `var_name` parameter. ([1616](https://github.com/arviz-devs/arviz/pull/1616))
- Added ability to plot HDI contours to `plot_kde` with the new `hdi_probs` parameter. ([1665](https://github.com/arviz-devs/arviz/pull/1665))
- Add dtype parsing and setting in all Stan converters ([1632](https://github.com/arviz-devs/arviz/pull/1632))
- Add option to specify colors for each element in ppc_plot ([1769](https://github.com/arviz-devs/arviz/pull/1769))

Maintenance and fixes

- Fix conversion for numpyro models with ImproperUniform latent sites ([1713](https://github.com/arviz-devs/arviz/pull/1713))
- Fixed conversion of Pyro output fit using GPUs ([1659](https://github.com/arviz-devs/arviz/pull/1659))
- Enforced using coordinate values as default labels ([1201](https://github.com/arviz-devs/arviz/pull/1201))
- Integrate `index_origin` with all the library ([1201](https://github.com/arviz-devs/arviz/pull/1201))
- Fix pareto k threshold typo in reloo function ([1580](https://github.com/arviz-devs/arviz/pull/1580))
- Preserve shape from Stan code in `from_cmdstanpy` ([1579](https://github.com/arviz-devs/arviz/pull/1579))
- Updated `from_pystan` converters to follow schema convention ([1585](https://github.com/arviz-devs/arviz/pull/1585)
- Used generator instead of list wherever possible ([1588](https://github.com/arviz-devs/arviz/pull/1588))
- Correctly use chain index when constructing PyMC3 `DefaultTrace` in `from_pymc3` ([1590](https://github.com/arviz-devs/arviz/pull/1590))
- Fix bugs in CmdStanPyConverter ([1595](https://github.com/arviz-devs/arviz/pull/1595) and [1598](https://github.com/arviz-devs/arviz/pull/1598))
- Fix `c` argument in `plot_khat` ([1592](https://github.com/arviz-devs/arviz/pull/1592))
- Fix `ax` argument in `plot_elpd` ([1593](https://github.com/arviz-devs/arviz/pull/1593))
- Remove warning in `stats.py` compare function ([1607](https://github.com/arviz-devs/arviz/pull/1607))
- Fix `ess/rhat` plots in `plot_forest` ([1606](https://github.com/arviz-devs/arviz/pull/1606))
- Fix `from_numpyro` crash when importing model with `thinning=x` for `x > 1` ([1619](https://github.com/arviz-devs/arviz/pull/1619))
- Upload updated mypy.ini in ci if mypy copilot fails ([1624](https://github.com/arviz-devs/arviz/pull/1624))
- Added type checking to raise an error whenever `InferenceData` object is passed using `io_pymc3`'s `trace` argument ([1629](https://github.com/arviz-devs/arviz/pull/1629))
- Fix `xlabels` in `plot_elpd` ([1601](https://github.com/arviz-devs/arviz/pull/1601))
- Renamed `sample` dim to `__sample__` when stacking `chain` and `draw` to avoid dimension collision ([1647](https://github.com/arviz-devs/arviz/pull/1647))
- Removed the `circular` argument in `plot_dist` in favor of `is_circular` ([1681](https://github.com/arviz-devs/arviz/pull/1681))
- Fix `legend` argument in `plot_separation` ([1701](https://github.com/arviz-devs/arviz/pull/1701))
- Removed testing dependency on http download for radon dataset ([1717](https://github.com/arviz-devs/arviz/pull/1717))
- Fixed plot_kde to take labels with kwargs. ([1710](https://github.com/arviz-devs/arviz/pull/1710))
- Fixed xarray related tests. ([1726](https://github.com/arviz-devs/arviz/pull/1726))
- Fix Bokeh deprecation warnings ([1657](https://github.com/arviz-devs/arviz/pull/1657))
- Fix credible inteval percentage in legend in `plot_loo_pit` ([1745](https://github.com/arviz-devs/arviz/pull/1745))
- Arguments `filter_vars` and `filter_groups` now raise `ValueError` if illegal arguments are passed ([1772](https://github.com/arviz-devs/arviz/pull/1772))
- Remove constrained_layout from arviz rcparams ([1764](https://github.com/arviz-devs/arviz/pull/1764))
- Fix plot_elpd for a single outlier ([1787](https://github.com/arviz-devs/arviz/pull/1787))

Deprecation

- Deprecated `index_origin` and `order` arguments in `az.summary` ([1201](https://github.com/arviz-devs/arviz/pull/1201))

Documentation

- Language improvements of the first third of the "Label guide" ([1699](https://github.com/arviz-devs/arviz/pull/1699))
- Added "Label guide" page and API section for `arviz.labels` module ([1201](https://github.com/arviz-devs/arviz/pull/1201) and [1635](https://github.com/arviz-devs/arviz/pull/1635))
- Add "Installation guide" page to the documentation ([1551](https://github.com/arviz-devs/arviz/pull/1551))
- Improve documentation on experimental `SamplingWrapper` classes ([1582](https://github.com/arviz-devs/arviz/pull/1582))
- Added example to `plot_hdi` using Inference Data ([1615](https://github.com/arviz-devs/arviz/pull/1615))
- Removed `geweke` diagnostic from `numba` user guide ([1653](https://github.com/arviz-devs/arviz/pull/1653))
- Restructured the documentation sections to improve community and about us information ([1587](https://github.com/arviz-devs/arviz/pull/1587))

New features

- Added `to_zarr` and `from_zarr` methods to InferenceData ([1518](https://github.com/arviz-devs/arviz/pull/1518))
- Added confidence interval band to auto-correlation plot ([1535](https://github.com/arviz-devs/arviz/pull/1535))

Maintenance and fixes

- Updated CmdStanPy converter form compatibility with versions >=0.9.68 ([1558](https://github.com/arviz-devs/arviz/pull/1558) and ([1564](https://github.com/arviz-devs/arviz/pull/1564))
- Updated `from_cmdstanpy`, `from_cmdstan`, `from_numpyro` and `from_pymc3` converters to follow schema convention ([1550](https://github.com/arviz-devs/arviz/pull/1550), [1541](https://github.com/arviz-devs/arviz/pull/1541), [1525](https://github.com/arviz-devs/arviz/pull/1525) and [1555](https://github.com/arviz-devs/arviz/pull/1555))
- Fix calculation of mode as point estimate ([1552](https://github.com/arviz-devs/arviz/pull/1552))
- Remove variable name from legend in posterior predictive plot ([1559](https://github.com/arviz-devs/arviz/pull/1559))
- Added significant digits formatter to round rope values ([1569](https://github.com/arviz-devs/arviz/pull/1569))
- Updated `from_cmdstan`. csv reader, dtype problem fixed and dtype kwarg added for manual dtype casting ([1565](https://github.com/arviz-devs/arviz/pull/1565))

Deprecation

- Removed Geweke diagnostic ([1545](https://github.com/arviz-devs/arviz/pull/1545))
- Removed credible_interval and include_circ arguments ([1548](https://github.com/arviz-devs/arviz/pull/1548))

Documentation

- Added an example for converting dataframe to InferenceData ([1556](https://github.com/arviz-devs/arviz/pull/1556))
- Added example for `coords` argument in `plot_posterior` docstring ([1566](https://github.com/arviz-devs/arviz/pull/1566))

Maintenance and fixes

- Fixed ovelapping titles and repeating warnings on circular traceplot ([1517](https://github.com/arviz-devs/arviz/pull/1517))
- Removed repetitive variable names from forest plots of multivariate variables ([1527](https://github.com/arviz-devs/arviz/pull/1527))
- Fixed regression in `plot_pair` labels that prevented coord names to be shown when necessary ([1533](https://github.com/arviz-devs/arviz/pull/1533))

Documentation

- Use tabs in ArviZ example gallery ([1521](https://github.com/arviz-devs/arviz/pull/1521))