RuntimeError: CUDA error: invalid device ordinal CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

[MingyuYang-Yale]

Hi,

I got another error:

RuntimeError: CUDA error: invalid device ordinal
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

This is my installation code:

conda create -y -n stDiff python=3.6
conda activate stDiff
pip install stDiff-sc
#  I got the error you mentioned in your github, so I run the following line as you suggested.
pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html  

This is my GPU information

Could you please provide guidance on how to resolve this issue?

Thank you!

[lkmdeer]

Hi，you could try with python 3.8, and the newer torch. It might work.

[MingyuYang-Yale]

Hi，you could try with python 3.8, and the newer torch. It might work.

Thank you for your reply! However, even I tried python 3.8 and with newer torch, it still got the same error.

[dyyjenny]

Same issue here with H100 gpus, tried pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html but didn't resolve the issue? Did you find a way to fix? Thanks!