This repository contains Dockerfile
and deploy.sh
to
set up a container and an environment to perform machine learning experiments in.
This repository is assumed to be used in servers of AYYM lab.
Clone this repository and change currect directory.
$ git clone [email protected]:cheeseNA/docker-lab.git
$ cd docker-lab
Build the image.
$ docker build -t docker-lab-image .
Make ws folder to mount and start the container.
$ mkdir ws
$ docker run -it -v $(pwd)/ws:/root/ws --gpus all --name docker-lab docker-lab-image
If you want to enable ssh-agent forwarding to the container, add options below.
-v $(readlink -f $SSH_AUTH_SOCK):/ssh-agent -e SSH_AUTH_SOCK=/ssh-agent
Make sure that you are connecting a lab server with
-A
flag orForwardAgent yes
option in~/.ssh/config
and thatssh-agent
is running on your computer.
Then, attach shell to the container and execute the following command.
# bash deploy.sh
Refresh shell and setup is done!
$ exec $SHELL -l