excluded _prior suffix from CLV models

[arthurmello]

Description

Removed the _prior suffix from parameters of CLV model config + adjacent files (tests, notebooks, etc.)

Related Issue

• Closes Exclude prior suffix from CLV model config #1428
• Related to #

Checklist

• Checked that the pre-commit linting/style checks pass. Feel free to comment pre-commit.ci autofix to auto-fix.
• Included tests that prove the fix is effective or that the new feature works
• Added necessary documentation (docstrings and/or example notebooks) using numpydoc format.
• If you are a pro: each commit corresponds to a relevant logical change

---

📚 Documentation preview 📚: https://pymc-marketing--1498.org.readthedocs.build/en/1498/

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.76%. Comparing base (b8b7b9d) to head (cd87a0a).
Report is 91 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1498      +/-   ##
==========================================
+ Coverage   92.75%   92.76%   +0.01%     
==========================================
  Files          52       52              
  Lines        6127     6138      +11     
==========================================
+ Hits         5683     5694      +11     
  Misses        444      444              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[williambdean]

Nice. Thanks for the PR @arthurmello

We should provide a deprecation for this. Likely in the clv/basic class. Maybe we can check for any model_config keys with "_prior" suffix then remove while having a warning

[williambdean]

Can we write a test to make sure that save / load is backwards compatible in this scenario

[arthurmello]

Sure, do you mean a test that ensures that models saved with the old configuration can still be loaded?
Something like:

def test_backward_compatibility_with_old_config(self, mocker):
        old_model_config = {
            "alpha_prior": ...
            "r_prior": ...
        }
        model = CLVModelTest(model_config=old_model_config)

        mocker.patch("pymc.sample", mock_sample)
        model.fit(...)
        model.save("old_model_config_test")

        loaded_model = CLVModelTest.load("old_model_config_test")

        assert loaded_model.model_config == {
            "alpha": ...,
            "r": ...
        }

[williambdean]

Yeah, something like that. Dont need to sample. Just load and make sure that a method that uses the posterior still works

We have an issue if the built model has variables without the prefix and posterior samples do have the prefix.

We might have to rename the variables upon load as well

[williambdean]

Let's test that there is a warning

[williambdean]

This isn't testing for loading from an old saved model. The idata.posterior will have the suffix. We want to check for that suffix in the posterior will not break using the model in the future

[williambdean]

pre-commit.ci autofix

[williambdean]

Looks good to me. Thanks for the PR!

@ColtAllen Can you take a look as well

[ColtAllen]

Looks good to me. Thanks for the PR!

@ColtAllen Can you take a look as well

I've started looking at this; will post my review tomorrow.

[ColtAllen]

Looks good; just some suggestions to update a unit test and add some removal reminders in addition to @wd60622's requested changes.

[ColtAllen]

Thanks for creating the issue for follow-up work!