boundless-zebra-79556
05/02/2023, 3:44 AM$ docker run --rm -it --gpus all --entrypoint=bash nvidia/cuda:11.7.1-devel-ubuntu22.04
# Inside docker env:
$ apt update && apt install -y python3.10 pip
$ pip install pex
$ pex torch==2.0.0
>>> import torch
Traceback (most recent call last):
File "<console>", line 1, in <module>
File "/root/.pex/installed_wheels/7a9319a67294ef02459a19738bbfa8727bb5307b822dadd708bc2ccf6c901aca/torch-2.0.0-cp310-cp310-manylinux1_x86_64.whl/torch/__init__.py", line 229, in <module>
from torch._C import * # noqa: F403
ImportError: libcudnn.so.8: cannot open shared object file: No such file or directory
$ pip install torch==2.0.0
$ python3.10
>>> import torch
>>> torch.cuda.is_available()
# etc etc Works fine
Any idea why this might be the case? I need to use this image so that it will work well with EKS' rather outdated GPU driverboundless-zebra-79556
05/02/2023, 3:45 AMLD_LIBRARY_PATH
set to /usr/local/nvidia/lib:/usr/local/nvidia/lib64
. any chance its not propagated to pex environment?enough-analyst-54434
05/02/2023, 4:20 AMpex torch==2.0 --venv --venv-site-packages-copies --pip-version latest --resolver-version pip-2020-resolver
enough-analyst-54434
05/02/2023, 4:29 AMboundless-zebra-79556
05/02/2023, 4:30 AMpex torch==2.0.0 --venv --venv-site-packages-copies
seems to do the trick. Could you quickly explain how --venv-site-packages-copies
might have helped in this scenario please?boundless-zebra-79556
05/02/2023, 4:36 AM