Closed RouChan closed 1 year ago
And for those under Pytorch1.5.1, you may need to add a line of code as follows at the beginning of train.py
.
torch.backends.cudnn.enabled = False
Hi Rouchan, thanks for reporting the bug. I fixed it and pushed the new-version code. Please take a look and let me know if doesn't work.
The code goes well now. Thanks so much.
Hi there. Thanks for your brilliant work and for your generosity to make this repository open source. I successfully trained my model. But when I try to test, seems some files missing.