Add ability to generate ER hypergraphs without multiedges

[nwlandry]

This PR fixes #339.

[codecov]

Codecov Report

Attention: Patch coverage is 84.21053% with 6 lines in your changes missing coverage. Please review.

Project coverage is 93.17%. Comparing base (5dc6f4f) to head (d2994fb).
Report is 5 commits behind head on main.

Files with missing lines	Patch %	Lines
xgi/generators/uniform.py	82.85%	6 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##             main     #596    +/-   ##
========================================
  Coverage   93.16%   93.17%            
========================================
  Files          60       62     +2     
  Lines        4523     4701   +178     
========================================
+ Hits         4214     4380   +166     
- Misses        309      321    +12     

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

[nwlandry]

@arashbm --- does this fix the issue you raised?

[nwlandry]

@maximelucas --- are you able to review this?

[maximelucas]

Mainly a few comments about clarifying the code with comments!

[nwlandry]

@maximelucas --- I believe I addressed all of your comments!

[maximelucas]

But there's no option to allow (or not) self-loops is this function, right? I'm a bit confused. In any case, since that singleton {0} is always removed by the method, does it matter?

The multiedges option is not linked to that correct?

[nwlandry]

I will try to clarify this sentence!

[maximelucas]

Thanks for the explanations! I'm still confused about some things, see above. Sorry for being a pain 😅 As long as the advantages are clear (and ideally why they are) we're good I guess. And that each method is clear from its own documentation.

If I understand a bit better, the index in "artificial/internal" in the sense that is just runs over all possible edges of size m among n nodes, with/without caring for the order (and hence allowing or not multiedges).

[nwlandry]

Thanks for the explanations! I'm still confused about some things, see above. Sorry for being a pain 😅 As long as the advantages are clear (and ideally why they are) we're good I guess. And that each method is clear from its own documentation.

If I understand a bit better, the index in "artificial/internal" in the sense that is just runs over all possible edges of size m among n nodes, with/without caring for the order (and hence allowing or not multiedges).

Yes, that's correct! I'll take a look at your more involved comments (hopefully today!) Thanks for the review!

[nwlandry]

Thanks for all the comments!!