about the "get_per_step_loss_importance_vector"

AntreasAntoniou / HowToTrainYourMAMLPytorch

The original code for the paper "How to train your MAML" along with a replication of the original "Model Agnostic Meta Learning" (MAML) paper in Pytorch.

https://arxiv.org/abs/1810.09502

Other

773 stars 137 forks source link

about the "get_per_step_loss_importance_vector" #33

Open liujianzhao6328057 opened 4 years ago

liujianzhao6328057 commented 4 years ago

thanks for your great work. I have some doubt about the "self.args.multi_step_loss_num_epochs" in function "get_per_step_loss_importance_vector". What does it mean?