Closed youyuanyi closed 11 months ago
Hi, I have not faced this issue, but after reading your error message feels like this is an issue with DataLoader (the fork()
was called most likely because num_workers=1
is present here). Now, if there is any data loader issue, I recommend setting num_workers=0
to understand better.
Finally, can you check if this discussion is relevant to your situation or not?
Thank you, it does work!
OS: Ubuntu 22.04 Graphic: RTX 3090 Python 3.10 mpi4py: 3.5.1 train_bash.sh
I encountered the following problem while training a diffuison model on cifar-10 datasest. Who also encountered this problem and how to solve it?