search

titipata / paper-reviewer-matcher

Linear programming solver for paper-reviewer matching and mind-matching
Apache License 2.0
82 stars 29 forks source link

Output of mindmatch.py #6

Open inezpereira opened 4 years ago

inezpereira commented 4 years ago

Hello! I was just trying out mindmatch.py and there is something I don’t get from the output. If we take for example the provided data/output_match.csv :

user_id,match_ids 1,93;217;463;645;783;1101 […] 217,97;304;731;792;1010;1134

Shouldn't user_id 217 also have 1 in its match_ids?

I also tried the function on some data of mine and I get the same situation.

titipata commented 4 years ago

Hi @InezPereira, thanks for the feedback. And yes, it should be the same as you mentioned! I'll check the output for you soonish.

titipata commented 4 years ago

@InezPereira this is really a bug from the linear programming setup, it seems! I'm investigating on this issue. Mainly, there is a shifted solution that still satisfies the constraints that we gave.

inezpereira commented 4 years ago

Hi @titipata! I'm sure you have been super busy with Neuromatch 2.0, but I was wondering if you have had time to look at this bug or have any suspicion of what might be failing. :)