Open WillLiGitHub opened 4 years ago
when you train multi-labels classification on you dataset, whether there is serious overfitting？ whether you novel loss function can resolve this problem?
when you train multi-labels classification on you dataset, whether there is serious overfitting？ whether you novel loss function can resolve this problem?