I‘m glad to read your work. However, I have some questiones as fllow:
Firstly, in the named "trainer.py" file, what's the means of the function of train_Bi_stage1_epoch/train_Bi_stage1_epoch?
Secondly, according to the design of CMKD framework loss function in your paper, I did not find its definition in the file named train.py
If you can answer my questiones, I will appreciate you very much.
I‘m glad to read your work. However, I have some questiones as fllow: Firstly, in the named "trainer.py" file, what's the means of the function of train_Bi_stage1_epoch/train_Bi_stage1_epoch? Secondly, according to the design of CMKD framework loss function in your paper, I did not find its definition in the file named train.py If you can answer my questiones, I will appreciate you very much.