Closed alalith3298 closed 11 months ago
@alalith3298 Could you please run a regular pytorch example on the GPU to ensure that your torch install is working: https://github.com/pytorch/examples/tree/main/mnist
There is no reason why you would get stuck in the trainer initialization unless there is some problem with torch I think. Or the other possibility is that you are misinterpreting where the script gets stuck. Make sure to record the output of your program in the bug report above.
@alalith3298 Could you please take another look?
Bug description
i am trying to load the trainer, but the code gets stuck at
trainer = Trainer(accelerator='gpu', devices=1)
. i am running this on a server with 4 NVIDIA RTX A6000s.What version are you seeing the problem on?
master
How to reproduce the bug
Error messages and logs here please