[ECCV-2022] Official implementation of MixSKD: Self-Knowledge Distillation from Mixup for Image Recognition && Pytorch Implementations of some Self-Knowledge Distillation and data augmentation methods
Hi Wingcg!
Your idea has inspired me a lot. May I ask whether logits self KD can be used alone, instead of combining logits self KD and feature self KD, because the size of the model and the amount of calculation are taken into account.Thanks a lot!
Hi!
I am grateful for your good work.
Now I'm learning FRSKD with your code, but the performance is about 62.xx, which is different from what is specified in github.
How can I fix this error?
(The code is nothing changed)