Docker 基础用法

创建镜像

1. create a Dockerfile

FROM tensorflow/tensorflow:nightly-devel-gpu-py3
RUN pip install opencv-python 
RUN ["apt-get", "update"]
RUN ["apt-get", "install", "-y", "vim-tiny"]
RUN pip3 install tqdm
RUN pip3 install seaborn
WORKDIR /project_name/
COPY . /project_name/
CMD []

2. create the image using the provided docker file

docker build <PATH-TO-THE-ROOT-OF-DOCKERFILE> -t <TAG-NAME>

3. create a container and enter it in the terminal

docker run -it <TAG-NAME> bash

3.1. you may share home folder with the container

docker run -v <home-path>:<map-to-container-path> -it <TAG-NAME> bash

3.2. If set the CMD in the dockerfile, you could run the image directly

docker run <TAG-NAME>

3. pull data from local files

scp <path-to-your-data/code> .

other usages

• display all images

docker images

• show running containers

docker ps

• close a running container

docker rm -f <CONTAINER_NAME>

• enter a runing container

docker exec -it <CONTAINER_NAME> bash

• restart a container

docker container restart <CONTAINER_NAME>