A Curator for Managing Docker Library.
Many lab clusters are managed by docker for sharing resources. However, sometimes we may find that using docker can be annoying since we need to figure out which is the volume/port to be mounted, whether we need to use privileged containers, which docker version to use (docker has nvidia GPU supports only after 19.03), and so on.
This project aims at providing a united interface for docker container configuration. So you can configure once, and deploy everywhere.
Please refer to configurations.
To build a library:
python3 curator.py build /path/to/the/libraryTo start a container from a library:
python3 curator.py start /path/to/the/libraryTo attach to a library:
python3 curator.py attach /path/to/the/libraryTo clean the existing containers:
python3 curator.py clean /path/to/the/libraryThe code is released under the MIT License.