* Merge close remote connection

* Manually pickle step method in multiprocess sampling

* Fix tests for extra divergence info

* Add test for remote process crash

* Better formatting in test_parallel_sampling
Co-authored-by: Junpeng Lao <junpenglao@gmail.com>

* Use mp_ctx forkserver on MacOS

* Add test for pickle with dill
Co-authored-by: Junpeng Lao <junpenglao@gmail.com>

Co-authored-by: Adrian Seyboldt <aseyboldt@users.noreply.github.com>

This reverts commit 1bf867ed.

* Add more info to divergence warnings

* Add dataclasses as requirement for py3.6

* Fix tests for extra divergence info

* Remove py3.6 requirements

Closes #3994.

* Drop support for py3.6

* Update RELEASE-NOTES.md
Co-authored-by: Colin <ColCarroll@users.noreply.github.com>
Co-authored-by: Colin <ColCarroll@users.noreply.github.com>

* Add time values as sampler stats for NUTS

* Use float time counters for nuts stats

* Add timing sampler stats to release notes

* Improve doc of time related sampler stats
Co-authored-by: Alexandre ANDORRA <andorra.alexandre@gmail.com>
Co-authored-by: Alexandre ANDORRA <andorra.alexandre@gmail.com>

* Honor discard_tuned_samples during KeyboardInterrupt

* Do not compute convergence checks without samples

* first attempt to vectorize smc kernel

* add ess, remove multiprocessing

* run multiple chains

* remove unused imports

* add more info to report

* minor fix

* test log

* fix type_num error

* remove unused imports update BF notebook

* update notebook with diagnostics

* update notebooks

* update notebook

* update notebook

* rewrite radon notebook using ArviZ and xarray

Roughly half notebook has been updated

* add comments on xarray usage

* rewrite 2n half of notebook

* minor fix

* rerun notebook and minor changes

* rerun notebook on pymc3.9.2 and ArviZ 0.9.0

* remove unused import

* add change to release notes

* update gp-latent nb to use arviz

* rerun, run black

* rerun after fixes from comments

* rerun black

* Updated RELEASE NOTES
* Bumped version number

* Add simple test case for coords behavior

* Install arviz directly from github as temporary fix

* Revert attempt to point to latest arviz github

* Bump arviz requirement
Co-authored-by: Michael Osthege <m.osthege@fz-juelich.de>

* add regression test for issue 3965
* pass on the tune argument to fix #3965

closes #3965

+ removes docs/* from the tarball to fix PyPI size limits (setup.py and MANIFEST.in)
+ updates the release notes with 3.9.0/3.9.1 sections
+ black applied to setup.py

Should reduce the frequency of getting "Mass matrix contains zeros on the diagonal" during warmup and fix a bunch of issue in https://github.com/pymc-devs/pymc3/issues/3959

* add long description content type
* bump to 3.9.0
* add GitHub release pipeline definition
* be conservative while waiting for PyPI
* address TODOs to make merge-ready

* Re-run and fix many notebooks...
Co-authored-by: Michael Osthege <m.osthege@fz-juelich.de>

* distribution on new line in GraphViz node

* updating tests to check newline instead of space in graphviz graphs

* changed ar1 logp to use ar1 precision instead of innovation precision

* fixed tests for AR1

* making AR1 scipy test behave like pymc3 version

* Update pymc3/distributions/timeseries.py
Co-authored-by: Thomas Wiecki <thomas.wiecki@gmail.com>

* Update pymc3/distributions/timeseries.py
Co-authored-by: Thomas Wiecki <thomas.wiecki@gmail.com>

* updated AR notebook to use new AR1 model

* Fixed release notes

* Black

* Updated AR NB

* Blackified AR NB
Co-authored-by: Thomas Wiecki <thomas.wiecki@gmail.com>
Co-authored-by: AlexAndorra <andorra.alexandre@gmail.com>

* reparameterized TP chi2 parameter has unique name now

* also fix TP chi2 name in tests

* remove path that somehow got to top of file...

* Allow specification of dims instead of shape

* Add pm.TidyData

* Create coords for pm.Data(ndarray)

* empty commit to trigger CI

* Apply suggestions from code review
Co-authored-by: Alexandre ANDORRA <andorra.alexandre@gmail.com>

* apply black formatting

* address review comments & formatting

* Add demonstration of named coordinates/dims

* don't require dim names to be identifiers

* sort imports

* raise ShapeError instead of ValueError

* formatting

* robustify Dtype and ShapeError

* Removed TidyData and refined dims and coords implementation

* Changed name of kwarg export_dims and improved docstrings

* Add link to ArviZ in docstrings

* Removed TidyData from __all__

* Polished Data container NB

* Fixed line break in data.py

* Fix inference of coords for dataframes

* Refined Data container NB

* Updated getting started NB with new dims and coords features

* Reran getting started NB

* Blackified NBs

* rerun with ArviZ branch

* use np.shape to be compatible with tuples/lists

* add tests for named coordinate handling

* Extended tests for data container
Co-authored-by: Michael Osthege <m.osthege@fz-juelich.de>
Co-authored-by: Michael Osthege <michael.osthege@outlook.com>
Co-authored-by: Alexandre ANDORRA <andorra.alexandre@gmail.com>