Request for Details on Teacher Model Training and Accuracy

[levunn]

Hi,

Thank you for providing this amazing repository! I have a question regarding the teacher model:

Could you please share some details about how the teacher model was trained? Specifically:

1. What techniques (e.g., data augmentation or scheduling, lr...) were used during training?
2. What is the accuracy or performance of the teacher model on the validation or test set?

Additionally, if possible, could you release the code used for training the teacher model?

Thank you so much for your time and support!