It seems that we can do the following columns computation: sum, avg, min and max.
See https://github.com/codalab/codabench/wiki/Leaderboard-Functionality and the screenshot below.
However, we can't compute average rank? It is a very important computation. Also, as Codabench is supposed to be v18 compatible, I though average rank was working already. We should implement it.
https://github.com/codalab/codabench/blob/develop/src/apps/leaderboards/models.py
In the long term, having organizers columns computation is what we want (#406), or at least being able to chose from the ranking functions of Ranky.
It seems that we can do the following columns computation:
sum,avg,minandmax.See https://github.com/codalab/codabench/wiki/Leaderboard-Functionality and the screenshot below.
However, we can't compute average rank? It is a very important computation. Also, as Codabench is supposed to be v18 compatible, I though average rank was working already. We should implement it.
https://github.com/codalab/codabench/blob/develop/src/apps/leaderboards/models.py
In the long term, having organizers columns computation is what we want (#406), or at least being able to chose from the ranking functions of Ranky.