Error using cuda in docker container #1771
Description
I used the Dockerfile to build pytorch
docker build -t pytorch .
and ran the image with
nvidia-docker run --rm -ti --ipc=host pytorch
Inside the container, I get an error using cuda:
import torch
x = torch.rand(1).cuda()
THCudaCheck FAIL file=/tmp/pip-b1neh3fv-build/torch/lib/THC/THCGeneral.c line=66 error=30 : unknown error
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/opt/conda/envs/pytorch-py35/lib/python3.5/site-packages/torch/_utils.py", line 65, in _cuda
return new_type(self.size()).copy_(self, async)
File "/opt/conda/envs/pytorch-py35/lib/python3.5/site-packages/torch/cuda/__init__.py", line 266, in _lazy_new
_lazy_init()
File "/opt/conda/envs/pytorch-py35/lib/python3.5/site-packages/torch/cuda/__init__.py", line 85, in _lazy_init
torch._C._cuda_init()
RuntimeError: cuda runtime error (30) : unknown error at /tmp/pip-b1neh3fv-build/torch/lib/THC/THCGeneral.c:66