c2s-docker

Dockerfile for the c2s toolbox.

The image is available as an automated build on Docker Hub.

If you need IPython, consider c2s-ipython-docker instead.

How to use

Please note that you might need to prefix all docker commands with sudo depending on your docker setup.

After installing Docker, get the c2s Docker image:

docker pull jonasrauber/c2s

Alternatively, you can build the image yourself, however the build process may take some time.

docker build --rm -t jonasrauber/c2s https://github.com/jonasrauber/c2s-docker.git

To simplify running c2s in a Docker container, create an alias:

alias c2s='docker run -it --rm -v "$PWD:/data/workdir" jonasrauber/c2s'

This creates a temporary alias. In order to make it persistent, just append this line to your ~/.bashrc, ~/.zshrc or similar files. If you are using bash, this can be done like this:

alias c2s >> ~/.bashrc

If you are using zsh, use this instead:

alias -L c2s >> ~/.zshrc

You can then use c2s as if it would be installed on the host:

c2s predict data.mat predictions.mat

The only difference is that c2s can only access files in your current working directory and its subdirectories.

To get more information on c2s itself, have a look at the c2s repository and the c2s documentaiton.

c2s visualize workaround

c2s visualize from within a Docker container requires a workaround as described in issue #3.

The following steps work on OS X:

brew install socat
socat TCP-LISTEN:6000,reuseaddr,fork UNIX-CLIENT:\"$DISPLAY\"
# in another window
docker run -it --rm -e DISPLAY=192.168.1.2:0 -v $PWD:/data/workdir jonasrauber/c2s visualize data.0.mat

The ip address 192.168.1.2 needs to be replaced with the respective host ip (i.e. the ip of OS X).