So I’ve noticed that the score I get remains the same regardless of what value I predict for samples ‘v5fsfs’ and ‘lrvqwb’. To confirm this, I downloaded the submission_format.csv file again, submitted it with all values default (0.5). Then I changed the values for the above two samples to 0 and the score did not change. Would someone be able to confirm that this is an issue with the scoring please (and it’s not me who is making a mistake)?
Thanks a lot!
Not every row ends up affecting your public score.