Rename zerosum_axes to n_zerosum_axes

[michaelraczycki]

What is this PR about?
-added n_zerosum_axes parameter
-added backwards compatibility for previous parameter name

Major / Breaking Changes

New features

Bugfixes

Documentation

• slightly altered ZeroSumNormal inline documentation

Maintenance

[AlexAndorra]

LGTM, thanks @michaelraczycki ! I think it's your very first PR on this repo, so congrats 🍾

[codecov]

Codecov Report

Merging #6522 (e85ecf3) into main (28bac77) will decrease coverage by 19.10%.
The diff coverage is 8.51%.

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #6522       +/-   ##
===========================================
- Coverage   94.74%   75.64%   -19.10%     
===========================================
  Files         146      147        +1     
  Lines       27807    27867       +60     
===========================================
- Hits        26346    21081     -5265     
- Misses       1461     6786     +5325     

Impacted Files	Coverage Δ
pymc/tests/distributions/test_multivariate.py	0.00% <0.00%> (-99.46%)	⬇️
pymc/distributions/multivariate.py	57.36% <16.00%> (-34.92%)	⬇️
pymc/tests/sampling/test_forward.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/stats/test_convergence.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/distributions/test_bound.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/sampling/test_population.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/stats/test_log_likelihood.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/distributions/test_censored.py	0.00% <0.00%> (-100.00%)	⬇️
pymc/tests/distributions/test_continuous.py	0.00% <0.00%> (-99.78%)	⬇️
pymc/tests/distributions/test_simulator.py	0.00% <0.00%> (-99.50%)	⬇️
... and 32 more

[AlexAndorra]

The doc fail is expected?

[michaelraczycki]

@AlexAndorra I'm not sure. I'm not so familiar with this check, and I can't find a manual way to trigger it locally. From what I can see here in the description, it looks like it hits some kind of timeout. Could you give some advice on how to check it to ensure everything is ok?

[ricardoV94]

The ReadTheDocs has been failing with timeout in other places, so don't worry about it here. This failing test however has to be fixed: https://github.com/pymc-devs/pymc/actions/runs/4174655801/jobs/7321891536

[michaelraczycki]

It might be that the docs check was failing because the solution had a typo, which resulted in the zerosum_axes param not to be correctly saved into its new version (n_zerosum_axes). That caused the support_shape etc. not to get correct values and the corresponding tests to fail. If readthedocs checks test coverage then indeed it would fail because the typo caused 'new' parameter to appear and the tests were not testing stuff that was actually created. I'm not sure if I understand the check correctly, so in any case any suggestions that would help me get it better are highly appreciated!

[michaelraczycki]

second mistake was not adapting unit tests, I noticed that they use the internal functions (ofc!) and I didn't switch the params used there to n_zerosum_axes, where inside the multivariate.py I did. Thus creating the "unexpected kwarg" error

[michaelraczycki]

@AlexAndorra could you approve the checks again? I think it should work now