Arviz

New features

- Initial bokeh support.
- ArviZ.jl a Julia interface to ArviZ (sethaxen )

Maintenance and fixes

- Fully support `numpyro` (fehiepsi )
- log_likelihood and observed data from pyro
- improve rcparams
- fix `az.concat` functionality (anzelpwj )

Documentation

- distplot docstring plotting example (jscarbor )

Maintenance and fixes

- Comment dev requirements in setup.py

New features

- Add from_numpyro Integration ([811](https://github.com/arviz-devs/arviz/pull/811))
- Numba Google Summer of Code additions (https://ban-zee.github.io/jekyll/update/2019/08/19/Submission.html)
- Model checking, Inference Data, and Convergence assessments (https://github.com/OriolAbril/gsoc2019/blob/master/final_work_submission.md)

New features

- Reorder stats columns ([695](https://github.com/arviz-devs/arviz/pull/695))
- Plot Forest reports ess and rhat by default([685](https://github.com/arviz-devs/arviz/pull/685))
- Add pointwise elpd ([678](https://github.com/arviz-devs/arviz/pull/678))

Maintenance and fixes

- Fix io_pymc3 bug ([693](https://github.com/arviz-devs/arviz/pull/693))
- Fix io_pymc3 warning ([686](https://github.com/arviz-devs/arviz/pull/686))
- Fix 0 size bug with pystan ([677](https://github.com/arviz-devs/arviz/pull/677))

New features

- Add plot_dist ([592](https://github.com/arviz-devs/arviz/pull/592))
- New rhat and ess ([623](https://github.com/arviz-devs/arviz/pull/623))
- Add plot_hpd ([611](https://github.com/arviz-devs/arviz/pull/611))
- Add plot_rank ([625](https://github.com/arviz-devs/arviz/pull/625))

Deprecation

- Remove load_data and save_data ([625](https://github.com/arviz-devs/arviz/pull/625))

New features

- Plot ppc supports multiple chains ([526](https://github.com/arviz-devs/arviz/pull/526))
- Plot titles now wrap ([441](https://github.com/arviz-devs/arviz/pull/441))
- plot_density uses a grid ([379](https://github.com/arviz-devs/arviz/pull/379))
- emcee reader support ([550](https://github.com/arviz-devs/arviz/pull/550))
- Animations in plot_ppc ([546](https://github.com/arviz-devs/arviz/pull/546))
- Optional dictionary for stat_funcs in summary ([583](https://github.com/arviz-devs/arviz/pull/583))
- Can exclude variables in selections with negated variable names ([574](https://github.com/arviz-devs/arviz/pull/574))

Maintenance and fixes

- Order maintained with xarray_var_iter ([557](https://github.com/arviz-devs/arviz/pull/557))
- Testing very improved (multiple)
- Fix nan handling in effective sample size ([573](https://github.com/arviz-devs/arviz/pull/573))
- Fix kde scaling ([582](https://github.com/arviz-devs/arviz/pull/582))
- xticks for discrete variables ([586](https://github.com/arviz-devs/arviz/pull/586))
- Empty InferenceData saves consistent with netcdf ([577](https://github.com/arviz-devs/arviz/pull/577))
- Removes numpy pinning ([594](https://github.com/arviz-devs/arviz/pull/594))

Documentation

- JOSS and Zenodo badges ([537](https://github.com/arviz-devs/arviz/pull/537))
- Gitter badge ([548](https://github.com/arviz-devs/arviz/pull/548))
- Docs for combining InferenceData ([590](https://github.com/arviz-devs/arviz/pull/590))