1차 시도 docker image docker pull nvcr.io/nvidia/pytorch:19.10-py3 NVIDIA CUDA 10.1.243 docker container NV_GPU=1 docker run --runtime=nvidia --name bernice-romp --shm-size 4G -it -v $(pwd):/workspace -v $(readlink -f disk1):/disk1 nvcr.io/nvidia/pytorch:19.10-py3 /bin/bash python -V # python import torch print(torch.__version__) print(torch.version.cuda) Python 3.6.9 :: Anaconda, Inc. 1.3.0a0+24ae..

[개요] docker hub 에서 pull 한 image 로 container 실행하기 [정리] # dockerhub에서 이미지 pull docker pull pytorch/pytorch:1.6.0-cuda10.1-cudnn7-runtime # 1번 GPU, /bin/bash로 container 실행 NV_GPU=1 nvidia-docker run -it -v :/ /bin/bash # 실제 예시 NV_GPU=1 nvidia-docker run --rm --name bernice-FreezeG --user 1002:1002 -it \ -v /etc/group:/etc/group:ro -v /etc/passwd:/etc/passwd:ro \ -v $(pwd):/workspace -v $(readlink -..

[이미지가 삭제되지 않는 문제] what I did docker rmi a42 what I expected 이미지가 삭제됨 error >> Error response from daemon: conflict: unable to delete a427a50f1301 (must be forced) - image is being used by stopped container c650817e060a solution docker ps -a # exit한 컨테이너까지 모두 확인 docker rm 921 # 관련 컨테이너 삭제 [nvcc fatal : Value 'sm_86' is not defined for option 'gpu-architecture'] what I did & what I expected styleg..