Thank you very much for your great work. I am currently trying to reproduce the training and evaluation performance using the released code. However, only test samples are provided, and there is little information in the paper about the training/evaluation datasets and the detailed protocol. Could you please help clarify these points?
Thank you very much for your great work. I am currently trying to reproduce the training and evaluation performance using the released code. However, only test samples are provided, and there is little information in the paper about the training/evaluation datasets and the detailed protocol. Could you please help clarify these points?