- Increase speed of development by useful boilerplate code
- It should be extensible, hence not very specific
Build the image
docker build -t data-science-base .
Run the container using following command
docker run --publish {host_port}:8889 --volume {path_to_local_dir}:/app/notebooks --rm data-science-base
host_port
- port on host machine where jupyter lab should be accessiblepath_to_local_dir
- host directory path where notebooks will be saved
example:
docker run --rm --publish 8889:8889 --volume /Users/shubhamgarg/notebooks:/app/notebooks data-science-base