Replacing PyMC3 plots w/ Arviz plots & sigma Param change

[CloudChaoszero]

Description

Replace PyMC3 dependent plots with arviz plots in case studies & examples.

• The issue is a 3-part step for the following:
  • Initiative via Issue Remove Plots & Diagnostics from PyMC Code Base pymc#4371 (Leave open)

Replace parameter sd with sigma (e.g. some examples have pm.Normal(...sd=...)

• Closes issue Use sigma instead of sd #2

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[twiecki]

Woah, @CloudChaoszero is this done / ready for review? Did you do it manually? Did you re-execute the NBs?

[CloudChaoszero]

@CloudChaoszero is this done / ready for review? Did you do it manually? Did you re-execute the NBs?

Yeah, @twiecki I would say this is done! haha (I was watching some tv shows while working on this, over time). This was done manually, and I re-executed most books***.

I do have some notes:

Notes

1. I'm confident I replaced all Pymc3 plots & diagnostics, regarding Remove Plots & Diagnostics from PyMC Code Base pymc#4371 (except for plot_posterior_predictive_glm, that stays)
2. ***There were some files with hiccups/errors, so I did not re-run them. (Errors & issues like what are listed here https://github.com/pymc-devs/pymc3/issues/3959) However, I did replace functions like pm.traceplot w/ az.plot_trace, etc
3. Added a couple warnings.ignore() because the notebooks that ran had my local path printed on the Warnings (e.g. Warning: C://User/Username/....ipynb etc)

[twiecki]

Amazing and much appreciated!

[AlexAndorra]

Thanks a lot for this @CloudChaoszero 🤩
I have one request though: the PR is currently huge, so could you chunk it off in several different PRs, each having, let's say 10 notebooks, and then a last one with the .py files?
That way, it'll be much easier and faster to review for us, and it'll be easier for you to update the files according to our comments. How does that sound?

[CloudChaoszero]

I have one request though: the PR is currently huge, so could you chunk it off in several different PRs, each having, let's say 10 notebooks, and then a last one with the .py files?
That way, it'll be much easier and faster to review for us, and it'll be easier for you to update the files according to our comments. How does that sound?

@AlexAndorra That works! haha

• I know what you mean.
  The PR's will be available, shortly. I broke them down by folder, which is an ~10 files.
  (This article helped me break them out)

[MarcoGorelli]

Thanks @CloudChaoszero - OK, shall we close this PR then in favour of the others? (I'll take a look at the during the coming week)

[AlexAndorra]

Yeah let's do that. Thanks @CloudChaoszero