search

openjournals / joss-reviews

Reviews for the Journal of Open Source Software
Creative Commons Zero v1.0 Universal
696 stars 36 forks source link

[REVIEW]: SpatialGEV: Fast Bayesian inference for spatial extreme value models in R #6878

Open editorialbot opened 1 month ago

editorialbot commented 1 month ago

Submitting author: !--author-handle-->@meixichen<!--end-author-handle-- (Meixi Chen) Repository: https://github.com/meixichen/SpatialGEV Branch with paper.md (empty if default branch): joss Version: v1.0.1 Editor: !--editor-->@vissarion<!--end-editor-- Reviewers: @fabian-s, @fernandomayer Archive: Pending

Status

status

Status badge code:

HTML: <a href="https://joss.theoj.org/papers/e13a0debdab2d59bb8bc0eefb55aa8ac"><img src="https://joss.theoj.org/papers/e13a0debdab2d59bb8bc0eefb55aa8ac/status.svg"></a>
Markdown: [![status](https://joss.theoj.org/papers/e13a0debdab2d59bb8bc0eefb55aa8ac/status.svg)](https://joss.theoj.org/papers/e13a0debdab2d59bb8bc0eefb55aa8ac)

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

@fabian-s & @fernandomayer, your review will be checklist based. Each of you will have a separate checklist that you should update when carrying out your review. First of all you need to run this command in a separate comment to create the checklist:

@editorialbot generate my checklist

The reviewer guidelines are available here: https://joss.readthedocs.io/en/latest/reviewer_guidelines.html. Any questions/concerns please let @vissarion 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 ✨

Checklists

📝 Checklist for @fabian-s

editorialbot commented 1 month ago

Hello humans, I'm @editorialbot, a robot that can help you with some common editorial tasks.

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

@editorialbot commands

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

@editorialbot generate pdf
editorialbot commented 1 month ago

Software report:

github.com/AlDanial/cloc v 1.90  T=0.03 s (2017.8 files/s, 242489.1 lines/s)
-------------------------------------------------------------------------------
Language                     files          blank        comment           code
-------------------------------------------------------------------------------
R                               37            146            932           2086
C/C++ Header                    15            234           1139           1342
Markdown                         2             61              0            327
TeX                              2             52              0            304
YAML                             2             12              6             58
C++                              2              4              2             53
Rmd                              2            138            511             44
-------------------------------------------------------------------------------
SUM:                            62            647           2590           4214
-------------------------------------------------------------------------------

Commit count by author:

   156  meixichen
    23  m372chen
     9  mlysy
     3  Meixi Chen
editorialbot commented 1 month ago

Paper file info:

📄 Wordcount for paper.md is 2092

✅ The paper includes a Statement of need section

editorialbot commented 1 month ago

License info:

🟡 License found: GNU General Public License v3.0 (Check here for OSI approval)

editorialbot commented 1 month ago 
Reference check summary (note 'MISSING' DOIs are suggestions that need verification):

OK DOIs

- None

MISSING DOIs

- No DOI given, and none found for title: Fast and Scalable Approximate Inference for Spatia...
- 10.2307/2287970 may be a valid DOI for title: Accurate Approximations for Posterior Moments and ...
- No DOI given, and none found for title: Inference and computation with generalized additiv...
- 10.1016/j.spasta.2022.100599 may be a valid DOI for title: The SPDE approach for Gaussian and non-Gaussian fi...
- 10.1016/s0167-4730(98)00015-0 may be a valid DOI for title: Extreme Value Modelling of Hurricane Wind Speeds
- 10.1198/016214506000000780 may be a valid DOI for title: Bayesian Spatial Modeling of Extreme Precipitation...
- 10.1007/s10687-009-0098-2 may be a valid DOI for title: A Comparison Study of Extreme Precipitation from S...
- 10.1002/env.2301 may be a valid DOI for title: Bayesian Hierarchical Modeling of Extreme Hourly P...
- 10.1016/j.jhydrol.2015.09.023 may be a valid DOI for title: A Spatial Model to Examine Rainfall Extremes in Co...
- 10.1201/b10905-6 may be a valid DOI for title: MCMC Using Hamiltonian Dynamics
- No DOI given, and none found for title: The No-U-Turn Sampler: Adaptively Setting Path Len...
- No DOI given, and none found for title: RStan: the R interface to Stan
- 10.1007/s13253-009-0010-1 may be a valid DOI for title: Continuous Spatial Process Models for Spatial Extr...
- No DOI given, and none found for title: Bayesian Spatial Modelling with R-INLA
- 10.1111/j.1467-9868.2011.00777.x may be a valid DOI for title: An explicit link between Gaussian fields and Gauss...
- No DOI given, and none found for title: TMB: Automatic Differentiation and Laplace Approxi...
- No DOI given, and none found for title: A Unifying View of Sparse Approximate Gaussian Pro...
- No DOI given, and none found for title: Understanding the Stochastic Partial Differential ...
- No DOI given, and none found for title: SpatialExtremes: Modelling Spatial Extremes
- No DOI given, and none found for title: \textttmgcv: Mixed GAM Computation Vehicle with Au...
- No DOI given, and none found for title: \textttevgam: An R Package for Generalized Additiv...

INVALID DOIs

- None
editorialbot commented 1 month ago

:point_right::page_facing_up: Download article proof :page_facing_up: View article proof on GitHub :page_facing_up: :point_left:

fabian-s commented 1 month ago

Review checklist for @fabian-s

Conflict of interest

Code of Conduct

General checks

Functionality

Documentation

Software paper

fabian-s commented 1 month ago

@meixichen could you try finding/adding the missing DOIs? (see above)

vissarion commented 1 month ago

Thanks @fabian-s for noting this. @meixichen please also try to reduce the length of the submission (following this pre-review-comment https://github.com/openjournals/joss-reviews/issues/6861#issuecomment-2157993938)

meixichen commented 1 month ago

Thanks for reviewing this manuscript and package. I will shorten the paper and add the missing DOIs.

meixichen commented 1 month ago

@editorialbot generate pdf

editorialbot commented 1 month ago

:point_right::page_facing_up: Download article proof :page_facing_up: View article proof on GitHub :page_facing_up: :point_left:

meixichen commented 1 month ago

@editorialbot check repository

editorialbot commented 1 month ago

Software report:

github.com/AlDanial/cloc v 1.90  T=0.03 s (2015.3 files/s, 237836.6 lines/s)
-------------------------------------------------------------------------------
Language                     files          blank        comment           code
-------------------------------------------------------------------------------
R                               37            146            932           2086
C/C++ Header                    15            234           1139           1342
TeX                              2             54              0            314
Markdown                         2             46              0            196
YAML                             2             12              6             58
C++                              2              4              2             53
Rmd                              2            138            511             44
-------------------------------------------------------------------------------
SUM:                            62            634           2590           4093
-------------------------------------------------------------------------------

Commit count by author:

   158  meixichen
    23  m372chen
     9  mlysy
     3  Meixi Chen
editorialbot commented 1 month ago

Paper file info:

📄 Wordcount for paper.md is 847

✅ The paper includes a Statement of need section

editorialbot commented 1 month ago

License info:

🟡 License found: GNU General Public License v3.0 (Check here for OSI approval)

meixichen commented 1 month ago

@editorialbot generate pdf

meixichen commented 1 month ago

@editorialbot check repository

editorialbot commented 1 month ago

Software report:

github.com/AlDanial/cloc v 1.90  T=0.03 s (1911.5 files/s, 227994.9 lines/s)
-------------------------------------------------------------------------------
Language                     files          blank        comment           code
-------------------------------------------------------------------------------
R                               37            146            932           2086
C/C++ Header                    15            234           1139           1342
TeX                              2             54              0            314
Markdown                         2             55              0            265
YAML                             2             12              6             58
C++                              2              4              2             53
Rmd                              2            138            511             44
-------------------------------------------------------------------------------
SUM:                            62            643           2590           4162
-------------------------------------------------------------------------------

Commit count by author:

   159  meixichen
    23  m372chen
     9  mlysy
     3  Meixi Chen
editorialbot commented 1 month ago

Paper file info:

📄 Wordcount for paper.md is 1494

✅ The paper includes a Statement of need section

editorialbot commented 1 month ago

License info:

🟡 License found: GNU General Public License v3.0 (Check here for OSI approval)

editorialbot commented 1 month ago

:point_right::page_facing_up: Download article proof :page_facing_up: View article proof on GitHub :page_facing_up: :point_left:

meixichen commented 1 month ago

Hello, I wonder if 1494 word count is still considered too long? Thanks!

fabian-s commented 1 month ago

Contribution and authorship:

Seems like only @meixichen has a meaningful number of commits here -- what are the contributions of the other two authors? The only commit of @mlysy from 4 years ago (https://github.com/meixichen/SpatialGEV/commit/47be58e10ef277a041d2c60817e10252939bad63) seems more like housekeeping to me and, in any case, git blame (https://github.com/meixichen/SpatialGEV/blame/master/inst/include/SpatialGEV/utils.hpp) seems to indicate that very little none of the code committed by them back then still remains part of the codebase?

fabian-s commented 1 month ago

Automated tests: test coverage seems to be rather low, see meixichen/SpatialGEV#14

fabian-s commented 1 month ago

Community guidelines: you should add a CONTRIBUTING.md or similar to the repo, see https://contributing.md/example/ e.g.

fabian-s commented 1 month ago

Functionality: Neither the package vignette nor the JOSS paper provide examples or details on how covariate information is included in the model. Are these linear effects? If so, how are the coefficients sampled and based on which priors? Do the covariates need to be scaled/centered?

fabian-s commented 1 month ago

Documentation: Your github README.md and your vignette use y as the first argument to spatialGEV_fit, the package and the JOSS paper use data, this seems to have diverged.

fabian-s commented 3 weeks ago

@meixichen any news re. my questions/requests above?

meixichen commented 3 weeks ago

@meixichen any news re. my questions/requests above?

@fabian-s Regarding https://github.com/meixichen/SpatialGEV/issues/14 I have merged joss with master and added more unittests. Test coverage is close to 90% now https://app.codecov.io/github/meixichen/SpatialGEV/tree/master

I am still working on the other requests and will reply here after I am done with them. For authorship, many conversations about functionalities and code design were done off-line or over emails. Some of mlysy's code were in the joss branch and were not reflected in master, but now after merging the joss into master we should see it.

meixichen commented 1 week ago

Hello @fabian-s , regarding the issues you flagged, I have made the following changes:

Any comments or suggestions? Thank you very much!

vissarion commented 1 week ago

Hi @fernandomayer just checking if there are any issues that blocks you to start with your review.

fabian-s commented 1 week ago

Thanks, I closed the issue.

The example with covariates in your vignette does not work: https://github.com/meixichen/SpatialGEV/blob/61466b5dde6e5f36965f2244fd8034224076cac3/vignettes/SpatialGEV-vignette.Rmd#L314C1-L340C1 is all just error messages...? If I re-compile it locally, it works fine, though:

image

I don't really think it makes sense to include this vignette "pre-compiled" as you're doing now, it doesn't take that long to run and it's bad practice to do so anyway, to avoid exactly this kind of problem. You also provide the wrong paths to your image files in that pre-compiled fake vignette (the paths are `figures/bla.png', but the files are not in that subdirectory, it does not exist....)

@vissarion not sure what your editorial standards are re. R packages. This is currently not fully standards compliant and would not pass R CMD check --as-cran, see https://github.com/meixichen/SpatialGEV/issues/22, but I can check off all checklist items in good conscience here.

vissarion commented 1 week ago

@fabian-s there are no editorial standard specifically for R packages. The general rule "packaged appropriately according to common community standards for the programming language being used", could mean that R CMD check should pass but not necessarily R CMD check --as-cran. But still I am not saying that R CMD check is a requirement for JOSS acceptance but rather a useful tool to indicate if the submitted software is appropriately packaged.

fabian-s commented 1 week ago

sure, I know, I also do editor-stuff for JOSS - that's why I said your standards ;)

FWIW, the NOTES/WARNINGS/ERROS don't seem that serious to me, the package can be installed in the standard way and it runs, including examples etc.

If it were my call, I would still request to make it CRAN-compliant s.t. the version of the package presented in the paper is more similar/identical to the respective CRAN version.

vissarion commented 1 week ago

I agree that it would be nice to make it CRAN-compliant. Though I would not like this to be a blocking issue. To my understanding there is not much work towards this goal (and in any case maybe that work will be done anyways to update the current CRAN package). @meixichen do you agree to make this package pass R CMD check --as-cran in the course of JOSS submission?

meixichen commented 1 week ago

@fabian-s @vissarion I agree with you both. I have pushed the new changes on the vignette, which is now no longer pre-compiled. Passed the check for me:

checking files in 'vignettes' ... OK
* checking examples ... OK
* checking for unstated dependencies in 'tests' ... OK
* checking tests ... [41s] OK
  Running 'testthat.R' [41s]
* checking for unstated dependencies in vignettes ... OK
* checking package vignettes ... OK
* checking re-building of vignette outputs ... [410s] OK
* checking PDF version of manual ... [21s] OK
* checking HTML version of manual ... OK
* DONE
fabian-s commented 1 week ago

Passed the check for me:

for me as well

fernandomayer commented 1 week ago

Hi @fernandomayer just checking if there are any issues that blocks you to start with your review.

@vissarion Sorry for the delay. I'll be starting my review this week, planning to finalise it mid-next week.