Teaching Container
A Docker container I use for my machine learning lectures.
Includes
- Miniconda Python 3
- scikit-learn 0.18.1
- keras 2 with Tensorflow backend
- matplotlib and seaborn
- Jupyter notebook (launched at start)
The container is also prepared for mapping the container X11 output to that of the host, so applications with a Graphical Interface can be used.
Prerequisites
You will need Docker Community Edition. Working under a Linux operative system is strongly recommended, but if you wish to work under Windows or Mac it might be best for you to install Docker Toolbox.
Usage
Run
docker run --rm -it -p 8888:8888 -v $(pwd):/home/developer -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix teaching-container
in a terminal while standing at the folder with the downloaded lab assignments. Then follow the link to the Jupyter notebook service that will appear in the terminal.