Add CLVWrapper

[wd60622]

Description

Building upon CLV MLflow integration. WIP

CC: @ColtAllen

Related Issue

• Closes #
• Related to #

Checklist

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

Modules affected

• MMM
• CLV
• Customer Choice

Type of change

• New feature / enhancement
• Bug fix
• Documentation
• Maintenance
• Other (please specify):

---

📚 Documentation preview 📚: https://pymc-marketing--1377.org.readthedocs.build/en/1377/

[codecov]

Codecov Report

Attention: Patch coverage is 45.45455% with 12 lines in your changes missing coverage. Please review.

Project coverage is 93.73%. Comparing base (16db682) to head (308ec37).
Report is 23 commits behind head on main.

Files with missing lines	Patch %	Lines
pymc_marketing/mlflow.py	45.45%	12 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1377      +/-   ##
==========================================
- Coverage   93.92%   93.73%   -0.20%     
==========================================
  Files          48       48              
  Lines        5133     5152      +19     
==========================================
+ Hits         4821     4829       +8     
- Misses        312      323      +11     

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

[wd60622]

General question for @ColtAllen

Are you using the save load much? What do you do after loading?

[ColtAllen]

General question for @ColtAllen

Are you using the save load much? What do you do after loading?

I did monthly model runs at my last company. With about 1.5 million customers the idata file was about 500MB, hence the need for #1356. Before loading a model, metrics and number of new customers were checked first to see if retraining was needed. Otherwise, run & rewrite predictions.

[ColtAllen]

Note #1356 is more of a nice-to-have than a blocker for this. Model files exceeding 1GB are huge by scikit-learn standards, but not uncommon for Hierarchical MMM models, and downright tiny for LLMs.