SciML / Surrogates.jl

Surrogate modeling and optimization for scientific machine learning (SciML)
https://docs.sciml.ai/Surrogates/stable/
Other
328 stars 70 forks source link

Remove unnecessary doc bits after unification #390

Closed ArnoStrouwen closed 2 years ago

codecov[bot] commented 2 years ago

Codecov Report

Merging #390 (39ccc49) into master (c1b8397) will decrease coverage by 0.29%. The diff coverage is n/a.

:exclamation: Current head 39ccc49 differs from pull request most recent head 0a92054. Consider uploading reports for the commit 0a92054 to get more accurate results

@@            Coverage Diff             @@
##           master     #390      +/-   ##
==========================================
- Coverage   79.53%   79.24%   -0.30%     
==========================================
  Files          16       16              
  Lines        2336     2327       -9     
==========================================
- Hits         1858     1844      -14     
- Misses        478      483       +5     
Impacted Files Coverage Δ
src/Wendland.jl 97.14% <0.00%> (-2.86%) :arrow_down:
src/Optimization.jl 71.61% <0.00%> (-0.45%) :arrow_down:
src/Radials.jl 88.07% <0.00%> (-0.22%) :arrow_down:
src/GEKPLS.jl 94.97% <0.00%> (-0.07%) :arrow_down:
src/Lobachevsky.jl 97.22% <0.00%> (-0.03%) :arrow_down:

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

vikram-s-narayan commented 2 years ago

I'm not sure why the contributing.md page is being removed as it is a good place to provide guidelines to new contributors.

At present it contains important information related to how to format code before making pull requests.

using JuliaFormatter, Surrogates
format(joinpath(dirname(pathof(Surrogates)), ".."))

In addition, I was planning on adding the following contributor guidelines to contributing.md:

###########

Contributor Guidelines:

  1. Please add more tests for the surrogates that you create. For your tests, ensure that you provide a baseline for accuracy. The idea behind this is that as other maintainers in the future change your code, the accuracy tests will ensure that the surrogate does not get worse on this important measure while getting better on some other measure such as performance.
  2. Please add paper references or code references as it will make maintenance easier in the long run.
  3. Add docstrings to functions
  4. If the surrogate you're adding has new dependencies not available in the main project.toml file, consider adding it to the libs folder. This enables you to have your own set of dependencies for your surrogate and will ensure that other surrogate models do not break.
  5. If possible, for unusual code-design choices made, please add comments describing the rationale for making that choice.

###########

One alternative is to retain the contributing.md and indicate how the contents of ColPrac relates to the CONTRIBUTING.md as prescribed in the last section of the ColPrac/README.md.

ArnoStrouwen commented 2 years ago

Most of those items sound like they should go to a general SciML contributions page and not be repeated for every package?

vikram-s-narayan commented 2 years ago

Most of those items sound like they should go to a general SciML contributions page and not be repeated for every package?

Yes. This applies to the docstrings and documenting rationale for making certain programming choices (items 3 and 5 from the list above).

However, I believe that the other items may be unique to surrogates (i.e. ensure that tests establish baseline accuracy, provide references to papers while writing code and add surrogates with unique package dependencies to the libs folder.)