Speed up notebook run action

[wd60622]

Closes #1141

This runs into some issues with sampling Flat distribution, loading the model, and some sample_stats

This potentially could reduce the run time of the notebooks to 1/3 of the original time. One local notebook run when from 2.5 minutes to 19 seconds.

CC: @juanitorduz

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.61%. Comparing base (31def91) to head (dfd1792).
Report is 13 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1147   +/-   ##
=======================================
  Coverage   95.61%   95.61%           
=======================================
  Files          39       39           
  Lines        4063     4063           
=======================================
  Hits         3885     3885           
  Misses        178      178           

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

[juanitorduz]

Thanks @wd60622 ! I left some small questions and comments! Looks cool!

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[wd60622]

I am looking at some of the logs. Seems like it was taken down from 40 minutes to 10 minutes 😆

[juanitorduz]

@wd60622 This is a new trick for me, hehe. How does the injected code "replace" the sampling step in the notebook? Can you maybe push an error in a notebook and see if it get caught (we can then revert the commit). This seems like magic XD

[wd60622]

@wd60622 This is a new trick for me, hehe. How does the injected code "replace" the sampling step in the notebook? Can you maybe push an error in a notebook and see if it get caught (we can then revert the commit). This seems like magic XD

The injected first cell will run the "injected.py". This is before any imports which makes python think that pm.sample is mock_sample and pm.Flat = pm.Normal and pm.HalfFlat = pm.HalfFlat because they are cached even when they are imported from our modules

[juanitorduz]

ok! I thnik I get it now :D Very nice! After removing the commented lines (see comment above), feel free to merge 🚀