Open WowThankyou opened 6 years ago
Hello everyone,
I have some problem about label, can the nodes in our own dataset have multi-label? In other word, the node could belong to two or more classes. If it can, could you please tell me how to implement ! Thank you very much!
Yes, simply replace the loss function with a multi-label loss function (e.g. sigmoid cross entropy instead of softmax cross entropy).
Thanks a lot!
Hello everyone,
I have some problem about label, can the nodes in our own dataset have multi-label? In other word, the node could belong to two or more classes. If it can, could you please tell me how to implement ! Thank you very much!