Sorry I'm a 1st timer here.
It says in the rule: "Before the end of the challenge you need to choose 2 submissions to be judged on for the private leaderboard. If you do not make a selection your 2 best public leaderboard submissions will be used to score on the private leaderboard."
But as I know the submission file is a csv with only prediction result of public test set. How can it be used to judge the private test set? Shouldn't we submit a notebook instead?
The public leaderboard is only 30% of what you have submitted, private is the whole thing
Yeah I get it. I'm only confused about how the private submission process works.
According to the rule (or at least how I understand it), upon deadline they will automatically choose our 2 best public submissions (30%) to evaluate the private test set (the whole thing). But since we only submit a csv file with prediction of public (30%) test set, how exactly can they do that?
Instead of csv file, should we submit the notebook with code so that they can run it with the full test set? Like some kaggle competitions
Or will we be given access to the private set to run it in our local environment (and a small extra time to submit the result)? I'm not sure about this, rule says private set is only available after challenge closed
Oh just read your reply again and now I got it lol. Thank you