My local results is the same as the online's results. use score.py to evaulate and the auc result is 0.0***, buy i see my prediction by meself, my acc is about 80%. I hope the organizer could fix this bug and update the STARTING_KIT.
Big thanks to you.
Sorry, 60%, not 80%, For example, 1-20:
prediction:
0 0 0 0 1 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0
Correct:
1 0 1 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0
Hi,
There is no error in the scoring function, we have used it in AutoML and AutoML2 challenges. Please note that a high acc does not imply a proportional Auc (besides, please note that we are using a normalized auc).
Best
Posted by: hugo.jair @ Aug. 10, 2018, 2:22 p.m.Big thanks for you reply!!I understand it, so sorry for this problem.
Posted by: Lean-Y @ Aug. 10, 2018, 2:25 p.m.