search

openjournals / joss-reviews

Reviews for the Journal of Open Source Software
Creative Commons Zero v1.0 Universal
721 stars 38 forks source link

[REVIEW]: ArviZ a unified library for exploratory analysis of Bayesian models in Python #1143

Closed whedon closed 5 years ago

whedon commented 5 years ago

Submitting author: @canyon289 (Ravin Kumar) Repository: https://github.com/arviz-devs/arviz Version: v.31 Editor: @arfon Reviewer: @malmaud , @mattpitkin Archive: 10.5281/zenodo.2540945

Status

status

Status badge code:

HTML: <a href="http://joss.theoj.org/papers/83e0e4048aa30a256a89f3b35b90f065"><img src="http://joss.theoj.org/papers/83e0e4048aa30a256a89f3b35b90f065/status.svg"></a>
Markdown: [![status](http://joss.theoj.org/papers/83e0e4048aa30a256a89f3b35b90f065/status.svg)](http://joss.theoj.org/papers/83e0e4048aa30a256a89f3b35b90f065)

Reviewers and authors:

Please avoid lengthy details of difficulties in the review thread. Instead, please create a new issue in the target repository and link to those issues (especially acceptance-blockers) in the review thread below. (For completists: if the target issue tracker is also on GitHub, linking the review thread in the issue or vice versa will create corresponding breadcrumb trails in the link target.)

Reviewer instructions & questions

@malmaud & @mattpitkin, please carry out your review in this issue by updating the checklist below. If you cannot edit the checklist please:

  1. Make sure you're logged in to your GitHub account
  2. Be sure to accept the invite at this URL: https://github.com/openjournals/joss-reviews/invitations

The reviewer guidelines are available here: https://joss.theoj.org/about#reviewer_guidelines. Any questions/concerns please let @arfon know.

✨ Please try and complete your review in the next two weeks ✨

Review checklist for @malmaud

Conflict of interest

Code of Conduct

General checks

Functionality

Documentation

Software paper

Review checklist for @mattpitkin

Conflict of interest

Code of Conduct

General checks

Functionality

Documentation

Software paper

whedon commented 5 years ago

Hello human, I'm @whedon, a robot that can help you with some common editorial tasks. @canyon289, it looks like you're currently assigned as the reviewer for this paper :tada:.

:star: Important :star:

If you haven't already, you should seriously consider unsubscribing from GitHub notifications for this (https://github.com/openjournals/joss-reviews) repository. As a reviewer, you're probably currently watching this repository which means for GitHub's default behaviour you will receive notifications (emails) for all reviews 😿

To fix this do the following two things:

  1. Set yourself as 'Not watching' https://github.com/openjournals/joss-reviews:

watching

  1. You may also like to change your default settings for this watching repositories in your GitHub profile here: https://github.com/settings/notifications

notifications

For a list of things I can do to help you, just type:

@whedon commands
whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

arfon commented 5 years ago

@malmaud , @mattpitkin - please carry out your review in this issue by updating the checklist above and giving feedback in this issue. The reviewer guidelines are available here: https://joss.readthedocs.io/en/latest/reviewer_guidelines.html

Any questions/concerns please let me know.

malmaud commented 5 years ago

OK, I've played with all the functionality of the software on my own device - everything works well, the utility is clear, and the documentation is comprehensive (albeit with a few minor typos). I think this is an easy accept after just a few tiny fixes:

canyon289 commented 5 years ago

@malmaud

Thanks for the feedback. A PR is open, just waiting for feedback internally before we submit it again for review to you.

https://github.com/arviz-devs/arviz/pull/483

canyon289 commented 5 years ago

@whedon generate pdf

whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

canyon289 commented 5 years ago

@malmaud The content is ready to review but the example plots seems to have rendered oddly. Is there a way I can see what command whedon is running to compile the pdf so I can test locally?

Also do you have a suggestion on where to place the example gallery, relative to the other headings?

mattpitkin commented 5 years ago

@canyon289 you can see a Makefile with an example of how you can locally generate a LaTeX file and pdf from the markdown doc here.

canyon289 commented 5 years ago

Tried getting the PDF to render correctly I think we might have figured it out

PR for reference https://github.com/arviz-devs/arviz/pull/484

Here it goes

@whedon generate pdf

canyon289 commented 5 years ago

@whedon generate pdf

whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

canyon289 commented 5 years ago

Well this failed. Will try something else. Sorry for all the spam guys

canyon289 commented 5 years ago

@whedon generate pdf

whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

canyon289 commented 5 years ago

Still not correct. I will revisit the way I'm rendering these to see if I can test more effectively in my local environment.

canyon289 commented 5 years ago

@malmaud and @mattpitkin

All feedback has been addressed, except example gallery which I'm struggling with tremendously.

I'm having issues generating a PDF with the make file, my tex output displays errors.

I can get a PDF to render with the command 

pandoc --filter pandoc-citeproc --bibliography=paper.bib  --variable papersize=a4paper -s paper.md -o paper.pdf

But while the output of this command may look good, the whedon generated pdf does not.

Lastly I tried looking for other JOSS papers with two or more figures but my random search was fruitless.

Here are two versions of the paper with Example gallery. Any suggestions on how to get them to render nicely? I realize this may be outside of the normal review role, I apologize for the trouble.

Multiple Figures https://github.com/arviz-devs/arviz/commit/4d5d6319c8b80872292ab2589606ca6d4a9b183a

Two Figures https://github.com/arviz-devs/arviz/commit/714bc2d3d0b852e477ef87c6460c3413411aa128

arfon commented 5 years ago

@canyon289 - could you share the PDF you've compiled locally here? I'd like to bette understand what you're trying to accomplish.

canyon289 commented 5 years ago

We're trying to add a local gallery per the review feedback In other words we'd like to meet the intent of the review, but struggling to get nice formatting while doing so.

Here's the two approaches we tried.

2+ Figures Here's a local render with multiple figures. Unfortunately this render required a "hack" to get it to work locally, so not a great solution https://github.com/arviz-devs/arviz/files/2722319/paper.pdf

Header hack https://github.com/arviz-devs/arviz/pull/484/files#diff-e2778e7674f45b806282a9611faa7220R100

Two Figures We also tried a version with just two figures. In the local render it seems to work, but not in JOSS. https://github.com/arviz-devs/arviz/files/2722350/paper.pdf https://github.com/arviz-devs/arviz/pull/485/files

Let me know if I explained this poorly and I can clarify any details

arfon commented 5 years ago

Ah OK. The problem here is that Pandoc + Markdown offers very little in terms of document flow/layout control.

Based on version https://github.com/arviz-devs/arviz/commit/4d5d6319c8b80872292ab2589606ca6d4a9b183a, I've generated the PDF locally by moving the # Example plots section below the # Acknowledgments section and slightly tweaking the width of the figures i.e.:

# Example plots  

A portion of ArviZ's functionality is shown in Figure 2 through Figure 5.  
ArviZ supports more plotting and inference diagnostics, in addition to the ones shown here. Additional examples can be found ArviZ documentation gallery

![Bivariate hexbin plot with marginal distributions](plot_joint.png){width=80%}

![2D Kernel Density estimation](plot_kde_2d.png){width=80%}

![Markov Chain Monte Carlo Trace Plot](plot_trace.png){width=80%}  

![John Kruschke styled posterior distribution plots](plot_posterior.png){width=80%}

This gives the resulting output: 10.21105.joss.01143.pdf

What do you think?

canyon289 commented 5 years ago

Looks good. We made the changes in our paper. Hopefully this works

canyon289 commented 5 years ago

@whedon generate pdf

whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

canyon289 commented 5 years ago

This looks ready for rereview. Thank you all for the help

mattpitkin commented 5 years ago

Hi @canyon289, everything looks good, but I just have a few comments:

canyon289 commented 5 years ago

@whedon generate pdf

whedon commented 5 years ago 
Attempting PDF compilation. Reticulating splines etc...
whedon commented 5 years ago

:point_right: Check article proof :page_facing_up: :point_left:

canyon289 commented 5 years ago

@mattpitkin Thank you for the feedback. We made some changes based on your suggestions. Let me know if there's anything else you notice

mattpitkin commented 5 years ago

@canyon289 thanks for the changes. Would it be possible to include the InferenceData class in the API documentation, or is there a specific reason for not doing this? I think it would be good to have it included even if the docstring mainly just adds a link to the page https://arviz-devs.github.io/arviz/notebooks/XarrayforArviZ.html. Also, could that notebook also contain the plot from Figure 1 of the paper?

canyon289 commented 5 years ago

@mattpitkin My (subjective) reasoning for not including az.InferenceData in the api docs was because all the methods in the api docs are intended to be callable by the user. For InferenceData the users don't directly instantiate an az.InferenceData. This is why I after some debate I opted not to include it in api docs, however this reasoning is largely subjective but would love to get your thoughts.

In terms of notebook, here is an updated version in a PR https://github.com/arviz-devs/arviz/pull/510. If you like it I'll merge so we can close that particular item!

https://github.com/arviz-devs/arviz/blob/8b9528ac046843ffd84f2429caff6fe0ce1c76da/doc/notebooks/XarrayforArviZ.ipynb

canyon289 commented 5 years ago

Please let me know if there's any remaining actions that I should take. I don't mean to rush this, just want to be sure I'm not holding anyone up

mattpitkin commented 5 years ago

Hi @canyon289, @arfon, sorry for the delay. I'm happy to sign-off on the review and have completed all the tick boxes.

canyon289 commented 5 years ago

@mattpitkin No problem, I didn't feel delayed! What are next steps?

arfon commented 5 years ago

Thanks @mattpitkin.

@canyon289 - At this point could you make an archive of the reviewed software in Zenodo/figshare/other service and update this thread with the DOI of the archive? I can then move forward with accepting the submission.

canyon289 commented 5 years ago

Thank you. We're prepping a release this week which will generate a DOI

canyon289 commented 5 years ago

@arfon

Please see the DOI archive here https://zenodo.org/record/2540945#.XD4csN-YXmE

arfon commented 5 years ago

@whedon set 10.5281/zenodo.2540945 as archive

whedon commented 5 years ago

OK. 10.5281/zenodo.2540945 is the archive.

arfon commented 5 years ago

@whedon accept

whedon commented 5 years ago 
Attempting dry run of processing paper acceptance...
whedon commented 5 years ago

Check final proof :point_right: https://github.com/openjournals/joss-papers/pull/433

If the paper PDF and Crossref deposit XML look good in https://github.com/openjournals/joss-papers/pull/433, then you can now move forward with accepting the submission by compiling again with the flag deposit=true e.g.

@whedon accept deposit=true
arfon commented 5 years ago

@whedon accept deposit=true

whedon commented 5 years ago 
Doing it live! Attempting automated processing of paper acceptance...
whedon commented 5 years ago

🚨🚨🚨 THIS IS NOT A DRILL, YOU HAVE JUST ACCEPTED A PAPER INTO JOSS! 🚨🚨🚨

Here's what you must now do:

  1. Check final PDF and Crossref metadata that was deposited :point_right: https://github.com/openjournals/joss-papers/pull/434
  2. Wait a couple of minutes to verify that the paper DOI resolves https://doi.org/10.21105/joss.01143
  3. If everything looks good, then close this review issue.

  4. Party like you just published a paper! 🎉🌈🦄💃👻🤘

    Any issues? notify your editorial technical team...