Open ShuchunXu opened 2 days ago
Dear Authers: I found that the training loss value is 'nan' after using the default training parameters in the source code, what are the training parameters used by the author, and the focus is on what is the learning rate?
How do I download the supplementary materials?
Dear Authers: I found that the training loss value is 'nan' after using the default training parameters in the source code, what are the training parameters used by the author, and the focus is on what is the learning rate?