[REVIEW]: Minimalist And Customizable Optimization Package

[whedon]

Submitting author: @jbuisine (Jérôme BUISINE)
Repository: https://github.com/jbuisine/macop
Version: v1.2.0
Editor: @melissawm
Reviewer: @stsievert, @torressa
Archive: 10.5281/zenodo.4595986

⚠️ JOSS reduced service mode ⚠️

Due to the challenges of the COVID-19 pandemic, JOSS is currently operating in a "reduced service mode". You can read more about what that means in our blog post.

Status

Status badge code:

HTML: <a href="https://joss.theoj.org/papers/9ea7d55c4fa83808f96929cb87adff3e"><img src="https://joss.theoj.org/papers/9ea7d55c4fa83808f96929cb87adff3e/status.svg"></a>
Markdown: [![status](https://joss.theoj.org/papers/9ea7d55c4fa83808f96929cb87adff3e/status.svg)](https://joss.theoj.org/papers/9ea7d55c4fa83808f96929cb87adff3e)

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) by leaving comments 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

@stsievert & @torressa, 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.readthedocs.io/en/latest/reviewer_guidelines.html. Any questions/concerns please let @melissawm know.

✨ Please start on your review when you are able, and be sure to complete your review in the next six weeks, at the very latest ✨

Review checklist for @stsievert

Conflict of interest

• I confirm that I have read the JOSS conflict of interest (COI) policy and that: I have no COIs with reviewing this work or that any perceived COIs have been waived by JOSS for the purpose of this review.

Code of Conduct

• I confirm that I read and will adhere to the JOSS code of conduct.

General checks

• Repository: Is the source code for this software available at the repository url?
• License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?
• Contribution and authorship: Has the submitting author (@jbuisine) made major contributions to the software? Does the full list of paper authors seem appropriate and complete?
• Substantial scholarly effort: Does this submission meet the scope eligibility described in the JOSS guidelines

Functionality

• Installation: Does installation proceed as outlined in the documentation?
• Functionality: Have the functional claims of the software been confirmed?
  • Specifically, this claim is made in paper.md: "Solutions modeling continuous problems can also be created by the anyone who wants to model his own problem", but I'm not seeing a continuous optimization problem in the examples or tests.
• Performance: If there are any performance claims of the software, have they been confirmed? (If there are no claims, please check off this item.)

Documentation

• A statement of need: Do the authors clearly state what problems the software is designed to solve and who the target audience is?
• Installation instructions: Is there a clearly-stated list of dependencies? Ideally these should be handled with an automated package management solution.
• Example usage: Do the authors include examples of how to use the software (ideally to solve real-world analysis problems).
• Functionality documentation: Is the core functionality of the software documented to a satisfactory level (e.g., API method documentation)?
• Automated tests: Are there automated tests or manual steps described so that the functionality of the software can be verified?
• Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Software paper

• Summary: Has a clear description of the high-level functionality and purpose of the software for a diverse, non-specialist audience been provided?
• A statement of need: Do the authors clearly state what problems the software is designed to solve and who the target audience is?
• State of the field: Do the authors describe how this software compares to other commonly-used packages?
• Quality of writing: Is the paper well written (i.e., it does not require editing for structure, language, or writing quality)?
• References: Is the list of references complete, and is everything cited appropriately that should be cited (e.g., papers, datasets, software)? Do references in the text use the proper citation syntax?

Review checklist for @torressa

Conflict of interest

• I confirm that I have read the JOSS conflict of interest (COI) policy and that: I have no COIs with reviewing this work or that any perceived COIs have been waived by JOSS for the purpose of this review.

Code of Conduct

• I confirm that I read and will adhere to the JOSS code of conduct.

General checks

• Repository: Is the source code for this software available at the repository url?
• License: Does the repository contain a plain-text LICENSE file with the contents of an OSI approved software license?
• Contribution and authorship: Has the submitting author (@jbuisine) made major contributions to the software? Does the full list of paper authors seem appropriate and complete?
• Substantial scholarly effort: Does this submission meet the scope eligibility described in the JOSS guidelines

Functionality

• Installation: Does installation proceed as outlined in the documentation?
• Functionality: Have the functional claims of the software been confirmed?
• Performance: If there are any performance claims of the software, have they been confirmed? (If there are no claims, please check off this item.)

Documentation

• A statement of need: Do the authors clearly state what problems the software is designed to solve and who the target audience is?
• Installation instructions: Is there a clearly-stated list of dependencies? Ideally these should be handled with an automated package management solution.
• Example usage: Do the authors include examples of how to use the software (ideally to solve real-world analysis problems).
• Functionality documentation: Is the core functionality of the software documented to a satisfactory level (e.g., API method documentation)?
• Automated tests: Are there automated tests or manual steps described so that the functionality of the software can be verified?
• Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

Software paper

• Summary: Has a clear description of the high-level functionality and purpose of the software for a diverse, non-specialist audience been provided?
• A statement of need: Do the authors clearly state what problems the software is designed to solve and who the target audience is?
• State of the field: Do the authors describe how this software compares to other commonly-used packages?
• Quality of writing: Is the paper well written (i.e., it does not require editing for structure, language, or writing quality)?
• References: Is the list of references complete, and is everything cited appropriately that should be cited (e.g., papers, datasets, software)? Do references in the text use the proper citation syntax?

[whedon]

Hello human, I'm @whedon, a robot that can help you with some common editorial tasks. @stsievert, @torressa it looks like you're currently assigned to review this paper 🎉.

⚠️ JOSS reduced service mode ⚠️

Due to the challenges of the COVID-19 pandemic, JOSS is currently operating in a "reduced service mode". You can read more about what that means in our blog post.

⭐ Important ⭐

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:

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

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

@whedon commands

For example, to regenerate the paper pdf after making changes in the paper's md or bib files, type:

@whedon generate pdf

[whedon]

Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- 10.1007/0-306-48056-5_11 is OK
- 10.1109/TEVC.2007.892759 is OK
- 10.1016/j.cor.2006.02.008 is OK
- 10.1109/TEVC.2013.2239648 is OK
- 10.1007/s00158-004-0465-1 is OK
- 10.1109/ICMLA.2007.35 is OK
- 10.3390/rs10071117 is OK

MISSING DOIs

- None

INVALID DOIs

- None

[whedon]

👉📄 Download article proof 📄 View article proof on GitHub 📄 👈

[melissawm]

👋🏼 @jbuisine @stsievert @torressa this is the review thread for the paper. All of our communications will happen here from now on.

Both reviewers have checklists at the top of this thread with the JOSS requirements. As you go over the submission, please check any items that you feel have been satisfied. There are also links to the JOSS reviewer guidelines.

The JOSS review is different from most other journals. Our goal is to work with the authors to help them meet our criteria instead of merely passing judgment on the submission. As such, the reviewers are encouraged to submit issues and pull requests on the software repository. When doing so, please mention openjournals/joss-reviews#REVIEW_NUMER so that a link is created to this thread (and I can keep an eye on what is happening). Please also feel free to comment and ask questions on this thread. In my experience, it is better to post comments/questions/suggestions as you come across them instead of waiting until you've reviewed the entire package.

We aim for reviews to be completed within about 2-4 weeks. Please let me know if any of you require some more time (@torressa has already done so). We can also use Whedon (our bot) to set automatic reminders if you know you'll be away for a known period of time.

Please feel free to ping me (@melissawm) if you have any questions/concerns.

[whedon]

👋 @stsievert, please update us on how your review is going.

[whedon]

👋 @torressa, please update us on how your review is going.

[stsievert]

I'll try to have a review in by this weekend.

[stsievert]

@whedon generate pdf

[whedon]

👉📄 Download article proof 📄 View article proof on GitHub 📄 👈

[stsievert]

@whedon commands

[whedon]

Here are some things you can ask me to do:

# List Whedon's capabilities
@whedon commands

# List of editor GitHub usernames
@whedon list editors

# List of reviewers together with programming language preferences and domain expertise
@whedon list reviewers

EDITORIAL TASKS

# Compile the paper
@whedon generate pdf

# Compile the paper from alternative branch
@whedon generate pdf from branch custom-branch-name

# Ask Whedon to check the references for missing DOIs
@whedon check references

# Ask Whedon to check repository statistics for the submitted software
@whedon check repository

[stsievert]

@melissawm I have a review waiting. It looks like I didn't accept the invitation to collaborate on this repo in time, so I can't edit #2812 (comment). Could you resend that invitation?

[melissawm]

Sure, no problem!

[melissawm]

@whedon re-invite @stsievert as reviewer

[whedon]

OK, the reviewer has been re-invited.

@stsievert please accept the invite by clicking this link: https://github.com/openjournals/joss-reviews/invitations

[stsievert]

My initial review: I do not believe this software project meets the JOSS review requirements. I welcome questions and comments from @jbuisine so provide clarification or be corrected on points below.

The most significant comments/questions are below. I have also edited #2812 (comment) to check various boxes* and add some notes on specific items.

• The tests are not sufficient. I only found a single test in setup.py, and not a single doctest as claimed. In this one test, there are many untested classes.
• Target audience and statement of need. The main documentation do not identify a target audience or a statement of need. Only one feature of the software, maximum flexibility, is described. Who would want that and why? In paper.md, I see that there's a need for maximum flexibility mentioned in the paper, but why would a user want if they're not doing "thesis work"?

The documentation has some specific problems: it's very general and not detailed, is missing a good API description.

• Generality: optimization is very general. When is Macop not appropriate?
  • For example, could I use it for my favorite deep learning problem? Could it be used for a continuous convex optimization?
  • Or what if the optimization problem I'm using requires sparse arrays or GPUs?
• Detailed: the only information core classes have is their name – past that, there's not much detail.
  • For example, the documentation for IteratedLocalSearch is basically "[this class is] used to avoid local optima and increave EvE (Exploration vs Exploitation) compromise". What's "EvE", what effect does it have, and how do the various parameters balance it?
  • For example, what do C and exp_rate do in UCBPolicy? Does increasing C increase or decrease the importance of exploration? How does changing the value of exp_rate change the exploration/exploitation tradeoff?
• API: it's not clear what different policies, mutators and crossovers are used for. What's the data flow of one example? Detailing the input/output of these modules would be useful. I'd expect to see a clear definitions for non-experts.

Some specific questions/suggestions/typos for paper.md:

• "generic and implemented OR algorithms" This is the first definition of "OR".
• "Tools for modelling and solving discrete and continuous problems are proposed in the literature." This needs at least one citation.
• "Allowing students to quickly develop their own algorithms." This sentence is a sentence fragment: it doesn't express a complete thought.
• "finding a point $x \in X$ en that has" What does "en" mean?
• "implement the whole available algorithms in the literature" -> "which doesn't implement every algorithm in the literature"
• "but let you the possibility to quickly develop and test" -> "but provides the ability to ..."
• "The main objective of this package is to be the most flexible as possible and hence, to offer a maximum of implementation possibilities." This sentence doesn't make sense. I think you're trying to convey "The main objective of this package is to provide maximum flexibility, which allows for easy experimentation in implementation."
• The second paragraph in motivation needs cleaning up -- I'm having a hard time deciphering "Binary solution was used as appreciation for selected or non-selected feature from the available set of features. The solution was therefore a new model obtained and its fitness which is the score obtained on the test data set. "
• $\mathbb{N}$ should be used instead of $\mathbf{N}$

^{* a checked box ([x]) means "satisfied", an unchecked box ([ ]) means "not reviewed" and a crossed box ([ ]) means "not satisfied". A partially satisfied requirement is indicated with "[.]"}

[torressa]

Hi @jbuisine thanks for the submission!
Here's my review, I welcome responses and questions in any of the following points. Also, I may have more comments in the future.

Major Issues (general)

• Documentation: bad.

  • The api is present but badly documented (e.g. "Iterated Local Search Algorithm implementation") and not up to date with the master branch. It needs more elaborate explanations of the implementations and citations where appropriate.
  • The README and paper do not present an accurate description of what Macop actually is. For this you have to go through the code. Clearer technical description (in paper too)
  • The uses of the toolbox has to be better defined, where is it better to use Macop over exact or other methods.

• Testing: bad. Unit tests missing (there is one test in the setup.py file)

• LICENSE not completed.

• Possibly not a substantial scholarly effort

  • I've found the bulk of the algorithmic implementation to be mostly contained in July 2019 and August-October 2020. The rest of the work is re-structuring and packaging. Therefore, raising the question of whether the time spent on this project was greater than 3 months.
  • To add to this, as reported by whedon there's 1114 lines of python code. Given the structure of the code (with the nested inheritance and redefinition of functions), it is hard to say whether this ticks the "Substantial scholarly effort" box.

• Software paper:

  • State of the field: there is no mention of other similar software packages. The author comments on well-known general optimisation packages (which need a proper citation), but nothing related to the methods used in Macop. In a quick google-search I found a more closely related repo: Solid which provides multiple derivate-free algorithms with user-defined objectives. The author probably knows more about this but I suspect there must be others, either way, it should be stated in the paper.
  • Quality of writing: the paper needs rereading to correct minor grammatical / orthographical errors. I will provide a list once the other comments have been addressed.

Major Issues (code):

• Software usage: the usage of library is not very clear. How does one extract the final solution without looking at the logs? Not the user defined criteria like the score/fitness, but the final solution (i.e. in the knapsack example, the items in the knapsack).
• Why is abstract class inheritance only sometimes used? This seems like the right type of inheritance for all the classes.
• macop/evaluators: contains a single example and not an implementation
• macop/operators: triple nested inheritance makes it hard to determine where functions are actually being implemented (e.g. UCBPolicy inherits from Policy which inherits from Operator). For such simple functionality, e.g definining the method apply, this seems a bit overkill. Additionally, inheritance is not used e.g. macop/operators/policies/UCBPolicy.py could be a standalone class.
• macop/solutions/: Same as above, inheritance doesn't do much as functions are overloaded and arguments not passed to the parent. The purpose of these functions is really not clear to me.
• In general, the module like structure is really confusing, specially given the small and simple implementations. This makes the imports list to extend a fair bit. I understand that this structure may have been laid out for additional functionality to be introduced. However, if no more work is to be done, to help people understand the code and encourage contributions I suggest a major rework of the code to simplify the structure.

Minor comments

• Please get rid of the build files for the documentation. readthedocs can build the docs on commits to master if you set it up.
• Relative imports are discouraged by PEP 328 "With the shift to absolute imports, the question arose whether relative imports should be allowed at all".
• Remove leading _ for all input parameters, this is common practice for private parameters.
• macop/callbacks: they are technically a callback, but it seems more like a custom logger
• Is there any way to disable the progress output being printed to screen? I can imagine this needs disabling in normal applications
• The knapsack is a nice an simple example, however, I don't think it showcases the best use of this library. As it takes longer to arrive at the optimal solution with 30 elements with Macop than traditional exact methods.
• In a standard python package, the init file in the main directory is not needed. init file in macop/ should contain module imports
• Contributions file is really long while explanations on how to contribute are short.
• Examples docs page is really long (half of it is imports and multi-line comments)
• In README:
  • Tags in readme should link to ci results and external urls, not images.
  • "OR algorithms"? list them.

[melissawm]

Thank you both @stsievert and @torressa for the considerations. @jbuisine do you think this is something you can work on with the reviewers comments? Please let me know in case any of you need clarifications or more information.

[jbuisine]

Thanks to @stsievert and @torressa for your remarks. @melissawm Yes it is something I can work on. I will see to take into consideration each of the remarks and update the Python package accordingly (documentation, examples, structural problems...). I will also take the time to answer each of your questions and queries about the package during the week.

[melissawm]

Hello, all! Any updates, @jbuisine ?

[jbuisine]

Hello @melissawm, I will work on it very soon! Sorry for the delay of the updates

[jbuisine]

Hello,

I wanted to inform you that I have been working on updating the package and all the answers for some time. Everything will be available in the next 2 days. I am really sorry for the delay.

[whedon]

👋 @openjournals/joss-eics, this paper is ready to be accepted and published.

Check final proof 👉 openjournals/joss-papers#2141

If the paper PDF and Crossref deposit XML look good in openjournals/joss-papers#2141, then you can now move forward with accepting the submission by compiling again with the flag deposit=true e.g.

@whedon accept deposit=true

[melissawm]

Looks like I missed a DOI for an ISBN in the references, sorry about this!

[kyleniemeyer]

No worries @melissawm!

Hi @jbuisine, I'm doing some final checks before publishing—can you add that missing DOI that the bot pointed out above?

[jbuisine]

@whedon generate pdf

[whedon]

👉📄 Download article proof 📄 View article proof on GitHub 📄 👈

[jbuisine]

@whedon check references

[whedon]

Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- 10.1007/0-306-48056-5_11 is OK
- 10.1109/TEVC.2007.892759 is OK
- 10.1016/j.cor.2006.02.008 is OK
- 10.1109/TEVC.2013.2239648 is OK
- 10.1007/s00158-004-0465-1 is OK
- 10.1109/ICMLA.2007.35 is OK
- 10.3390/rs10071117 is OK
- 10.1007/s00158-011-0666-3 is OK
- 10.1145/3321707.3321800 is OK

MISSING DOIs

- 10.1007/978-3-319-42432-3_37 may be a valid DOI for title: PySCIPOpt: Mathematical Programming in Python with the SCIP Optimization Suite

INVALID DOIs

- None

[jbuisine]

@melissawm, @kyleniemeyer Hi!

I'm sorry but I do not understand why this reference is not checked. The DOI refers to this article.

Could you please tell me what I might have forgotten?

[melissawm]

You need to include the DOI link in the reference just as you did for the others. You have included an ISBN, but not the DOI.

[jbuisine]

@whedon check references

[whedon]

Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- 10.1007/0-306-48056-5_11 is OK
- 10.1109/TEVC.2007.892759 is OK
- 10.1016/j.cor.2006.02.008 is OK
- 10.1109/TEVC.2013.2239648 is OK
- 10.1007/s00158-004-0465-1 is OK
- 10.1109/ICMLA.2007.35 is OK
- 10.3390/rs10071117 is OK
- 10.1007/s00158-011-0666-3 is OK
- 10.1007/978-3-319-42432-3_37 is OK
- 10.1145/3321707.3321800 is OK

MISSING DOIs

- None

INVALID DOIs

- None

[jbuisine]

@melissawm Thank you for your reply!

@kyleniemeyer I have now added the DOI.

[kyleniemeyer]

@whedon generate pdf

[whedon]

👉📄 Download article proof 📄 View article proof on GitHub 📄 👈

[kyleniemeyer]

@whedon accept

[whedon]

Attempting dry run of processing paper acceptance...

[whedon]

Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- 10.1007/0-306-48056-5_11 is OK
- 10.1109/TEVC.2007.892759 is OK
- 10.1016/j.cor.2006.02.008 is OK
- 10.1109/TEVC.2013.2239648 is OK
- 10.1007/s00158-004-0465-1 is OK
- 10.1109/ICMLA.2007.35 is OK
- 10.3390/rs10071117 is OK
- 10.1007/s00158-011-0666-3 is OK
- 10.1007/978-3-319-42432-3_37 is OK
- 10.1145/3321707.3321800 is OK

MISSING DOIs

- None

INVALID DOIs

- None

[whedon]

👋 @openjournals/joss-eics, this paper is ready to be accepted and published.

Check final proof 👉 openjournals/joss-papers#2142

If the paper PDF and Crossref deposit XML look good in openjournals/joss-papers#2142, then you can now move forward with accepting the submission by compiling again with the flag deposit=true e.g.

@whedon accept deposit=true

[kyleniemeyer]

@whedon accept deposit=true

[whedon]

Doing it live! Attempting automated processing of paper acceptance...

[whedon]

🐦🐦🐦 👉 Tweet for this paper 👈 🐦🐦🐦

[whedon]

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

Here's what you must now do:

0. Check final PDF and Crossref metadata that was deposited 👉 Creating pull request for 10.21105.joss.02812 joss-papers#2143
1. Wait a couple of minutes to verify that the paper DOI resolves https://doi.org/10.21105/joss.02812
2. If everything looks good, then close this review issue.
3. Party like you just published a paper! 🎉🌈🦄💃👻🤘

Any issues? Notify your editorial technical team...

[kyleniemeyer]

Congrats @jbuisine on your article's publication in JOSS!

Many thanks to @stsievert and @torressa for reviewing this, and @melissawm for editing.

[whedon]

🎉🎉🎉 Congratulations on your paper acceptance! 🎉🎉🎉

If you would like to include a link to your paper from your README use the following code snippets:

Markdown:
[![DOI](https://joss.theoj.org/papers/10.21105/joss.02812/status.svg)](https://doi.org/10.21105/joss.02812)

HTML:
<a style="border-width:0" href="https://doi.org/10.21105/joss.02812">
  <img src="https://joss.theoj.org/papers/10.21105/joss.02812/status.svg" alt="DOI badge" >
</a>

reStructuredText:
.. image:: https://joss.theoj.org/papers/10.21105/joss.02812/status.svg
   :target: https://doi.org/10.21105/joss.02812

This is how it will look in your documentation:

We need your help!

Journal of Open Source Software is a community-run journal and relies upon volunteer effort. If you'd like to support us please consider doing either one (or both) of the the following:

• Volunteering to review for us sometime in the future. You can add your name to the reviewer list here: https://joss.theoj.org/reviewer-signup.html
• Making a small donation to support our running costs here: https://numfocus.org/donate-to-joss