I have the same problem. I also want to know how to deal with it.
I agree with @konstantin_nikolayev. I think it is better to evaluate the submissions by more precise log loss score instead of rounded one.
In this competition, to improve the log loss by 0.000x (it may be ignored if rounded to the 3rd decimal) requires some effort.
if scores are rounded, it is more likely that the difference between better solution and others is just when it is created rather than what it is.
Though 0.000x improvement may not be so important for practical use, I think it is better to take the difference into consideration to make this competition fair.
Thank you for your reply.
It seems that the leaderboard has changed, but the order may still not be correct. (for example, there are 3 teams in rank#3, and the team that has the second highest score and submitted the earliest is placed at the bottom among them.)
Is it still being fixed?
Is it really ok to consider all submissions? if so, I think a solution that just happen to overfit the test dataset is more likely to be selected as the best.
how are the submissions sorted on the Leaderboard? I think they are sorted only based on the log-loss score. But now(5/8 13:35 UTC) on the Leaderboard, it seems that #1 public log-loss score is 0.60837 and #2 is 0.60834. These submissions are tie if they are rounded on 3 decimal place, however, #2 was submitted earlier than #1. Is that an expected behavior of the Leaderboard?
in addition, do we need to select the final submission that is evaluated on the private leaderboard? or all submissions will be included on the private leaderboard?
i’m sorry if this is not appropriate place where we ask questions to competition admins.