Closed yianzhongguo closed 5 years ago
Please read the paper carefully: "Whenever possible, we initialize these techniques using aggressive parameters that we subsequently attenuate over the course of the training." Data augmentation is made to be less aggressive at some point
Thank you!
@FabianIsensee Hi, sir. In the train_network.py when the eporch==100 and epoch==175 the data_gen_train is different generated form create_data_gen_train with different parameters. Why this happend? Thank you!