ByL&YY

About MHSA PROBLEMS.

1

Hello. In your paper, you said that you use multi head self attention, and this head num is four. But in this code, I only see 1 head in mhsa...

dear author: I have a question that training on gopro, have 2 blur folder, that blur and blur_gamma, I want to know you used is blur or blur_gamma? thankyou

Some question about the mindspore code different from paper

1

Dear noah: In your IPT paper, writer that " The body has 12 encoder layers and 12 decoder layers.", but in the mindspore code, the num_layers in encoder and decoder...

About datasets

3

dear author: I have a question that training on gopro, have 2 blur folder, that blur and blur_gamma, I want to know you used is blur or blur_gamma? thankyou

Dear author: could you release the train code about this paper, thank you and looking forward your reply`

In this code main.py, I dont see the implementation of dice loss, could you please provide this part! thank you