Add ChiSquared moment

[patel-zeel]

Add moments and tests for the below distributions as part of #5078

• pymc.distributions.continuous.ChiSquared

[ricardoV94]

Thanks for opening a PR. I left a small suggestion. Looks great by the way!

[patel-zeel]

Agree. Implemented the suggestion :)

[ricardoV94]

Looks good. Waiting to see if the tests pass

[codecov]

Codecov Report

Merging #5154 (ebc1c25) into main (310a6b9) will decrease coverage by 0.06%.
The diff coverage is 100.00%.

❗ Current head ebc1c25 differs from pull request most recent head 1f81e22. Consider uploading reports for the commit 1f81e22 to get more accurate results

@@            Coverage Diff             @@
##             main    #5154      +/-   ##
==========================================
- Coverage   77.99%   77.92%   -0.07%     
==========================================
  Files          88       88              
  Lines       14094    14099       +5     
==========================================
- Hits        10992    10987       -5     
- Misses       3102     3112      +10     

Impacted Files	Coverage Δ
pymc/distributions/continuous.py	95.82% <100.00%> (+0.02%)	⬆️
pymc/step_methods/hmc/integration.py	78.84% <0.00%> (-3.85%)	⬇️
pymc/backends/report.py	89.51% <0.00%> (-2.10%)	⬇️
pymc/parallel_sampling.py	86.33% <0.00%> (-1.00%)	⬇️
pymc/tests/sampler_fixtures.py	90.51% <0.00%> (-0.73%)	⬇️
pymc/step_methods/hmc/nuts.py	95.00% <0.00%> (-0.63%)	⬇️

[patel-zeel]

Not sure why test_exec_nuts_init failed. Can you please help me with this @ricardoV94?

[ricardoV94]

Not sure why test_exec_nuts_init failed. Can you please help me with this @ricardoV94?

Can you replicate the failure locally?

[patel-zeel]

I tried and it seems to work locally.

[ricardoV94]

Probably it was a fluke, because find_MAP is still starting from prior draws instead of moments, so that test is not even deterministic. The changes to the Chisquared from this PR look completely unrelated. I'll merge this already.

[ricardoV94]

Thanks for your help @patel-zeel!

[patel-zeel]

Thank, you @ricardoV94 for the clarification and for accepting the PR :)