Docker - Tensorflow 1.8-gpu 셋팅

1. Docker Pull

​x
docker pull tensorflow/tensorflow:1.8.0-gpu-py3       // 태그명이 1.8.0-gpu-py3인 텐서플로우 이미지 다운로드
docker images  
​

2. Docker Run

xxxxxxxxxx
nvidia-docker run -it \
-p 8888:8888 -p 6006:6006 \
--name ailab-yurim \
-v /home/zayden/Desktop/test_garbage:/notebooks \
-e PASSWORD="0000" \
--restart always \
tensorflow/tensorflow:1.8.0-gpu-py3

3. Docker Commit

xxxxxxxxxx
docker commit ailab-yurim riverzayden/ailab-yurim:0.1

4. Docker stop, remove

xxxxxxxxxx
docker ps -a  
​
docker stop eade3e95f777
​
docker rm eade3e95f777

5. Commit된것으로 다시 실행 해보기

xxxxxxxxxx
nvidia-docker run -it \
-p 8888:8888 -p 6006:6006 \
--name ailab-yurim \
-v /home/zayden/Desktop/test_garbage:/notebooks \
-e PASSWORD="0000" \
--restart always \
riverzayden/ailab-yurim:0.1

6. Docker Login and Push

xxxxxxxxxx
docker login
#docker tag local-image:tagname new-repo:tagname
docker tag riverzayden/ailab-yurim:0.1
docker push riverzayden/ailab-yurim:0.1
docker pull riverzayden/ailab-yurim:0.1
​

출처

https://devyurim.github.io/python/tensorflow/development%20enviroment/docker/2018/05/25/tensorflow-3.html