GLM linear

[OriolAbril]

File: https://github.com/pymc-devs/pymc-examples/blob/main/examples/generalized_linear_models/GLM-linear.ipynb
Reviewers:

Please refer to the notebook updates overview for more detailed guidance on the points below

Known changes needed

Changes listed in this section should all be done at some point in order to get this
notebook to a "Best Practices" state. However, these are probably not enough!
Make sure to thoroughly review the notebook and search for other updates.

General updates

• Use new numpy random generator
• Check why manual model runs slower. Is the default prior of glm.from_formula a halfcauchy?

ArviZ related

• In code cell 7, pass figsize as kwarg to plot_trace, not as plt.figure argument.

Notes

Should that notebook mention bambi at some point? cc @aloctavodia

Exotic dependencies

I think it needs patsy, not completely sure if it' installed with pymc3 or if it's an optional dependency.

Computing requirements

Models run in less than 30 seconds.

[chiral-carbon]

Hello! I commented under issue #58 too but this issue is also one I'd like to take up and attempt to make the suggested improvements.

[OriolAbril]

I'll assign this to you too, as PRs are only on a per notebook basis here, it's no problem to work on 2-3 notebooks simultaneously

[chiral-carbon]

great, thank you!

[chiral-carbon]

hello! had a plots related doubt. apart from using figzise as kwargs in plot_trace in cell 7 (using ArviZ), is there any other change that should be done? it's a very minor change so I wanted to confirm in case I might have misunderstood.

the plot code looks like this after changing:
az.plot_trace(trace, figsize=(11,7));

is this sufficient?

[OriolAbril]

Yes, the notebook is short and concise and there aren't many things to update

[chiral-carbon]

okay, got it. thanks.

[OriolAbril]

Needs to be updated to use bambi instead of glm module