[Paper] add comparison tables and benchmark

[Santymax98]

Hi Oskar, I don't know what you think about this, I used "BenchmarkTools.jl" to get the results. Can it be helpful for correction?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (b9bc7e4) 83.42% compared to head (5f1b355) 87.31%.
Report is 21 commits behind head on main.

❗ Current head 5f1b355 differs from pull request most recent head 2db2884. Consider uploading reports for the commit 2db2884 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #127      +/-   ##
==========================================
+ Coverage   83.42%   87.31%   +3.89%     
==========================================
  Files          29       29              
  Lines         748      749       +1     
==========================================
+ Hits          624      654      +30     
+ Misses        124       95      -29     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[lrnv]

First table looks very good. The second one is surprising to me, this is the first time I ran BenchmarkTools on these things, and i NEVER optimized the code, this is wild that we beat DataGenCopulaBased like that. What about BivariateCopulas why dont you have a result ?

[Santymax98]

With "bivariatecopula.jl" it is only possible for the bivariate case and on the other hand I was not able to execute the code correctly. I was trying for quite some time using the documentation but I didn't succeed. Maybe it's a problem with my machine. On the other hand, do you think we should cite "BenchmarkTools" and put the characteristics of the machine I used to obtain those results?

[lrnv]

That is good ! Yes you can cite benchmarktools like i cited other packages: with a link to their repository +, if they have some bibtex citation in their readme and/or docs, these bibtex entries. Look at what I did for BivariateCopulas to see how it works.

Nice numbers. The statement should be somthing like

Since our primary target is maintainability and readability of the implementation, we did not consider the efficiency and the perfromance of the code yet. However, a (limited in scope) benchmark on Copulas.jl's implementation of Clayton's pdf is availiable in Table 1 and shows competitve behavior of our implementation.

[lrnv]

You could add here a line "Classic Bivariate copulas sampling" with Yes/Yes/Yes.

Also a line "Obscure bivariate copulas sampling" with "Yes/No/No ?" ^^

Maybe the first column could be refactored to :
Sampling
- Classic Bivariate Copulas Yes Yes Yes
- Obscure Bivariate Copulas Yes No No
- Archimedean copulas All / Classic Only / Classic Only
- Multivariate Copulas Yes Yes No (You exchanged the two i think)
- Archimedean Chains (this is the real name of "nested" : No No Yes
Fitting No No Yes
Plotting No No Yes
Dependence metrics Partial / ? / ? (check)

[lrnv]

It would also be nice to add this benchmark to the documentation (with the code to run it) since it is pretty :)

[lrnv]

For the reference to work you need to add it to the joss/paper.bib file, taking it from their CITATION file there : https://github.com/JuliaCI/BenchmarkTools.jl/blob/master/CITATION.bib

@ARTICLE{BenchmarkTools,
  author =	 {{Chen}, Jiahao and {Revels}, Jarrett},
  title =	 "{Robust benchmarking in noisy environments}",
  journal =	 {arXiv e-prints},
  keywords =	 {Computer Science - Performance, 68N30, B.8.1, D.2.5},
  year =	 2016,
  month =	 "Aug",
  eid =		 {arXiv:1608.04295},
  archivePrefix ={arXiv},
  eprint =	 {1608.04295},
  primaryClass = {cs.PF},
  adsurl =	 {https://ui.adsabs.harvard.edu/abs/2016arXiv160804295C},
  adsnote =	 {Provided by the SAO/NASA Astrophysics Data System}
}

[Santymax98]

Hello, I added the code, however I only added the one I used for copulas.jl but it works with any package. Additionally, it must be taken into account that the methodology for obtaining samples is different for each package.

[lrnv]

Thanks, this is Ok for now I will finsh this PR and merge it myself on monday. very good job.

Can you now focus on #98 from now on ? I am sure this is not very hard to solve but this needs to work correctly.

[Santymax98]

Do you think I can add a thank you to the foundation that finances my studies? El El vie, 9 feb 2024 a la(s) 15:11, Oskar Laverny ***@***.***> escribió:

…

Thanks, this is Ok for now I will finsh this PR and merge it myself on monday. very good job. Can you now focus on #98 <#98> from now on ? I am sure this is not very hard to solve but this needs to work correctly. — Reply to this email directly, view it on GitHub <#127 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7PTWNXEVSRDNMIU5IYVNGTYSZ7GRAVCNFSM6AAAAABC4YIQXGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMZWGUZTSNRUGM> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[lrnv]

Yes of course, add it under a section at the end of the paper before the references like so :

# Acknowledgement

Santiago Jiménez Ramos thanks XXX for the funding 

[Santymax98]

Thanks, I will. El El vie, 9 feb 2024 a la(s) 15:24, Oskar Laverny ***@***.***> escribió:

…

Yes of course, add it under a section at the end of the paper before the references like so : # Acknowledgement Santiago Jiménez Ramos thanks XXX for the funding — Reply to this email directly, view it on GitHub <#127 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7PTWNTB7CRSW7ZADGO3463YS2AYPAVCNFSM6AAAAABC4YIQXGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMZWGU2TKMJSGY> . You are receiving this because you authored the thread.Message ID: ***@***.***>