FAQ page first draft #657

abuzarmahmood · 2023-03-23T16:22:36Z

Adding questions for FAQ page per Issue #644

canyon289

This is look good to me. I'll let Tomas review as well!

Thank you for the PR

docs/faq.rst

tomicapretto · 2023-03-23T21:13:33Z

docs/faq.rst

+
+What is the difference between Bambi and PyMC3?
+----------------------------
+    * Bambi is a wrapper around PyMC3, NumPyro, and BlackJax. It provides a simple interface for


I would say Bambi is a library built on top of PyMC, it's not a wrapper of PyMC. Also, it's not a wrapper of NumPyro or BlackJax. Technically when PyMC compiles a PyMC model for sampling with NumPyro or Blackjax it's creating such models, but it's happening all automatically and it's not exposed to users

Yes, calling it a wrapper is incorrect. Thank you for the correction. I also took out mention of numpyro and blackjax here because they are mentioned as backends in multiple other questions.

tomicapretto · 2023-03-23T21:17:59Z

Hi @abuzarmahmood thanks a lot for this contribution!

I think the following questions fit really well in the FAQ

What is the difference between Bambi and PyMC3?
What sampling methods are available?
Can infernece in Bambi be sped up using GPUs/TPUs?
My data has a non-normal distributions, can I still use Bambi?
- Although I would phrase this in a different way.
- I want to use a non-normal likelihood model, can I still use Bambi?
How do I find out what priors are available?
Does bambi come with pre-specified regression models?

I'm not so sure about these other questions

Why have a Bayesian regression library?
I'm not familiar with Bayesian modelling, where can I learn more?
My sampler through errors/indicating divergences, what should I do?

Those are more conceptual/philosophical and modeling-related questions. I don't think we should try to answer them in the FAQ, since I think the FAQ is for questions about the library.

Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

codecov-commenter · 2023-03-23T22:13:45Z

Codecov Report

Merging #657 (2733fbd) into main (afcabb6) will increase coverage by 0.06%.
The diff coverage is 95.45%.

❗ Current head 2733fbd differs from pull request most recent head 4d829ab. Consider uploading reports for the commit 4d829ab to get more accurate results

@@            Coverage Diff             @@
##             main     #657      +/-   ##
==========================================
+ Coverage   87.56%   87.62%   +0.06%     
==========================================
  Files          40       40              
  Lines        2629     2650      +21     
==========================================
+ Hits         2302     2322      +20     
- Misses        327      328       +1

Impacted Files	Coverage Δ
bambi/defaults/distributions.py	`100.00% <ø> (ø)`
bambi/defaults/families.py	`87.50% <ø> (ø)`
bambi/defaults/utils.py	`87.50% <ø> (ø)`
bambi/families/likelihood.py	`85.36% <ø> (ø)`
bambi/families/univariate.py	`90.37% <95.00%> (+0.80%)`	⬆️
bambi/families/family.py	`93.25% <100.00%> (ø)`

... and 1 file with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

abuzarmahmood · 2023-03-23T22:37:31Z

Hi @abuzarmahmood thanks a lot for this contribution!

I think the following questions fit really well in the FAQ

What is the difference between Bambi and PyMC3?

What sampling methods are available?

Can infernece in Bambi be sped up using GPUs/TPUs?

My data has a non-normal distributions, can I still use Bambi?

Although I would phrase this in a different way.

I want to use a non-normal likelihood model, can I still use Bambi?

How do I find out what priors are available?

Does bambi come with pre-specified regression models?

I'm not so sure about these other questions

Why have a Bayesian regression library?

I'm not familiar with Bayesian modelling, where can I learn more?

My sampler through errors/indicating divergences, what should I do?

Those are more conceptual/philosophical and modeling-related questions. I don't think we should try to answer them in the FAQ, since I think the FAQ is for questions about the library.

That makes sense. I've updated it to include the questions you suggested.

docs/faq.rst

tomicapretto

Hi @abuzarmahmood sorry for the delay in the reply. I just requested one more small change, then we're good to merge. Thanks!

abuzarmahmood · 2023-04-18T15:26:21Z

@tomicapretto Got it! That was my bad for confusing families with priors. I've update the response to the priors question (67f52fb).

tomicapretto · 2023-04-18T17:32:59Z

@tomicapretto Got it! That was my bad for confusing families with priors. I've update the response to the priors question (67f52fb).

Last nit: It's PyMC instead of PyMC3 (we're using PyMC version 5, but we write just PyMC)

abuzarmahmood · 2023-04-18T18:53:35Z

@tomicapretto Got it! That was my bad for confusing families with priors. I've update the response to the priors question (67f52fb).

Last nit: It's PyMC instead of PyMC3 (we're using PyMC version 5, but we write just PyMC)

Sounds good. Latest commit uses "PyMC" instead of "PyMC3".

tomicapretto · 2023-04-18T23:24:37Z

The test failure is known. It's due to an update of the variable names in PyMC/PyTensor, which uses full names now. It's already updated in #668. I'm merging.

Thanks @abuzarmahmood !

abuzarmahmood · 2023-04-18T23:44:36Z

Lovely. Thank you @tomicapretto @canyon289 for your guidance

* FAQ page first draft * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Remove general bayesian modelling questions + minor fixes * Update faq.rst * Update faq.rst --------- Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

* plot_cap using post. pred. samples * add mtcars dataset and test plot_cap using post. pred. samples * plot_cap show predictions at obs. level * remove unused code and formatting * Add mtcars as a dataset. Bump PyMC to 5.3.0 * FAQ page first draft (#657) * FAQ page first draft * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Remove general bayesian modelling questions + minor fixes * Update faq.rst * Update faq.rst --------- Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * black reformatting * add tests for pps=bool using pytest parameterization --------- Co-authored-by: Tomas Capretto <tomicapretto@gmail.com> Co-authored-by: Abuzar Mahmood <abuzarmahmood@gmail.com> Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

…inos#668) * plot_cap using post. pred. samples * add mtcars dataset and test plot_cap using post. pred. samples * plot_cap show predictions at obs. level * remove unused code and formatting * Add mtcars as a dataset. Bump PyMC to 5.3.0 * FAQ page first draft (bambinos#657) * FAQ page first draft * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Update docs/faq.rst Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * Remove general bayesian modelling questions + minor fixes * Update faq.rst * Update faq.rst --------- Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com> * black reformatting * add tests for pps=bool using pytest parameterization --------- Co-authored-by: Tomas Capretto <tomicapretto@gmail.com> Co-authored-by: Abuzar Mahmood <abuzarmahmood@gmail.com> Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

FAQ page first draft

2733fbd

abuzarmahmood marked this pull request as draft March 23, 2023 16:28

canyon289 reviewed Mar 23, 2023

View reviewed changes

docs/faq.rst Outdated Show resolved Hide resolved

docs/faq.rst Outdated Show resolved Hide resolved

docs/faq.rst Show resolved Hide resolved

tomicapretto reviewed Mar 23, 2023

View reviewed changes

docs/faq.rst Outdated Show resolved Hide resolved

tomicapretto reviewed Mar 23, 2023

View reviewed changes

Update docs/faq.rst

1ef4a42

Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

abuzarmahmood and others added 3 commits March 23, 2023 18:23

Update docs/faq.rst

e96fb31

Co-authored-by: Ravin Kumar <7213793+canyon289@users.noreply.github.com>

Remove general bayesian modelling questions + minor fixes

18ac6a9

Merge branch 'faqs' of https://github.com/abuzarmahmood/bambi into faqs

a4a91ef

abuzarmahmood marked this pull request as ready for review March 23, 2023 22:37

tomicapretto reviewed Apr 18, 2023

View reviewed changes

docs/faq.rst Outdated Show resolved Hide resolved

tomicapretto requested changes Apr 18, 2023

View reviewed changes

abuzarmahmood added 3 commits April 18, 2023 11:18

Update faq.rst

a658bb1

Update faq.rst

67f52fb

Merge branch 'faqs' of https://github.com/abuzarmahmood/bambi into faqs

4d829ab

tomicapretto merged commit 6f2f399 into bambinos:main Apr 18, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

FAQ page first draft #657

FAQ page first draft #657

abuzarmahmood commented Mar 23, 2023

canyon289 left a comment

tomicapretto Mar 23, 2023

abuzarmahmood Mar 23, 2023

tomicapretto commented Mar 23, 2023 •

edited

Loading

codecov-commenter commented Mar 23, 2023 •

edited

Loading

abuzarmahmood commented Mar 23, 2023

tomicapretto left a comment

abuzarmahmood commented Apr 18, 2023 •

edited

Loading

tomicapretto commented Apr 18, 2023

abuzarmahmood commented Apr 18, 2023

tomicapretto commented Apr 18, 2023

abuzarmahmood commented Apr 18, 2023

FAQ page first draft #657

FAQ page first draft #657

Conversation

abuzarmahmood commented Mar 23, 2023

canyon289 left a comment

Choose a reason for hiding this comment

tomicapretto Mar 23, 2023

Choose a reason for hiding this comment

abuzarmahmood Mar 23, 2023

Choose a reason for hiding this comment

tomicapretto commented Mar 23, 2023 • edited Loading

codecov-commenter commented Mar 23, 2023 • edited Loading

Codecov Report

abuzarmahmood commented Mar 23, 2023

tomicapretto left a comment

Choose a reason for hiding this comment

abuzarmahmood commented Apr 18, 2023 • edited Loading

tomicapretto commented Apr 18, 2023

abuzarmahmood commented Apr 18, 2023

tomicapretto commented Apr 18, 2023

abuzarmahmood commented Apr 18, 2023

tomicapretto commented Mar 23, 2023 •

edited

Loading

codecov-commenter commented Mar 23, 2023 •

edited

Loading

abuzarmahmood commented Apr 18, 2023 •

edited

Loading