Rectify return type hints in logprob module rewrites

[AryanNanda17]

Description

Related Issue

• Related to Rectify Return type hint #6811

Checklist

• Checked that the pre-commit linting/style checks pass
• If you are a pro: each commit corresponds to a [relevant logical change]

Type of change

• Other (please specify):

def find_measurable_comparisons( 
    fgraph: FunctionGraph, node: Node 
) -> Optional[List[MeasurableComparison]]: 

Here, the return type hint is wrong. MeasurableComparision is an Op, but what is returned are TensorVariables whose node has that Op. This needs to be rectified for all the functions which intend to find a measurable op.

---

📚 Documentation preview 📚: https://pymc--7093.org.readthedocs.build/en/7093/

[welcome]

💖 Thanks for opening this pull request! 💖 The PyMC community really appreciates your time and effort to contribute to the project. Please make sure you have read our Contributing Guidelines and filled in our pull request template to the best of your ability.

[AryanNanda17]

@larryshamalama please have a look at the pull request.

[ricardoV94]

Looks good. Sorry for the delay in reviewing

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (7bb2ccd) 92.21% compared to head (4c73bd3) 76.34%.
Report is 18 commits behind head on main.

❗ Current head 4c73bd3 differs from pull request most recent head fd3ced5. Consider uploading reports for the commit fd3ced5 to get more accurate results

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #7093       +/-   ##
===========================================
- Coverage   92.21%   76.34%   -15.87%     
===========================================
  Files         101      101               
  Lines       16912    16889       -23     
===========================================
- Hits        15595    12894     -2701     
- Misses       1317     3995     +2678     

Files	Coverage Δ
pymc/logprob/binary.py	29.54% <100.00%> (-66.93%)	⬇️
pymc/logprob/censoring.py	61.29% <100.00%> (-37.59%)	⬇️
pymc/logprob/checks.py	42.59% <100.00%> (-57.41%)	⬇️
pymc/logprob/cumsum.py	88.23% <100.00%> (-11.77%)	⬇️
pymc/logprob/tensor.py	48.76% <100.00%> (-30.38%)	⬇️

... and 67 files with indirect coverage changes

[AryanNanda17]

@ricardoV94 , In this file:-
pymc/pymc/logprob/tensor.py

 def find_measurable_stacks( 
     fgraph, node 
 ) -> Optional[List[Union[TensorVariable, TensorVariable]]]:

This went wrong. What is wrong here?

[ricardoV94]

@ricardoV94 , In this file:- pymc/pymc/logprob/tensor.py

 def find_measurable_stacks( 
     fgraph, node 
 ) -> Optional[List[Union[TensorVariable, TensorVariable]]]:

This went wrong. What is wrong here?

The Union is not needed, since you have the same return type now

[AryanNanda17]

I updated it.
Thanks!

[AryanNanda17]

Is this a formatting issue?
@ricardoV94

[AryanNanda17]

@ricardoV94 @larryshamalama Please look into the issue. I researched a lot during the past few days, but did not get anything specific about the error here.
Thanks

[ricardoV94]

@AryanNanda17 yes that's a formatting issue due to pre-commit. You can read about it in our contributing guides: https://www.pymc.io/projects/docs/en/latest/contributing/pr_tutorial.html

Check the pre-commit instructions specifically

[AryanNanda17]

@ricardoV94, I made the changes and now the precommits tests are passing.

[AryanNanda17]

@ricardoV94 Why am I getting errors during mergeable check?

[larryshamalama]

@ricardoV94 Why am I getting errors during mergeable check?

Hi @AryanNanda17, sorry about such delays in replying - I am pushing through graduate studies, which have been time-consuming. I will look more carefully at this afternoon or tomorrow.

Have you:

• Rebased with respect to main? Currently, I see that this branch has conflicts. This is probably due to the addition of commits to the main branch
• Applied pre-commit? Our linter and pre-commit configurations recently changed (#7901)

[AryanNanda17]

My feature branch was some commits behind the main branch and I was trying to resolve merge conflicts, by mistake I discarded all the commits, and the pull request got closed, really sorry.
I will make a new pull request.

[larryshamalama]

Sounds good, do not hesitate to ping me in the new PR once you will have done so.

Edit: FYI, this was previously worked on in #7013. In principle, priority is given to preceding PRs, but it was my fault for providing late reviews. If possible, co-authorship will be given regardless of the merged PR (I will look into how to do this)