Python Data Science container

This is fully ready Docker container with:

• NumPy
• Pandas
• Sklearn
• Matplotlib
• Seaborn
• pyyaml
• h5py
• Jupyter
• Tensorflow
• Keras
• OpenCV 3

We're building this container on top of Ubuntu 16.04 Docker container (Dockerfile) and Anaconda Docker container (Dockerfile)

docker build -f Dockerfile . -t <tag name>

Push the new docker image to your docker hub

$ docker tag first-image {docker-hub-username}/{default-repo-folder-name}:first-image
$ docker push {docker-hub-username}/{default-repo-folder-name}:first-image

Running container

We'll use notebooks forlder to store Jupyter Notebooks:

mkdir notebooks

Run Docker container with the following command:

docker run -it -p 8888:8888 -p 6006:6006 -d -v $(pwd)/notebooks:/notebooks amaksimov/python_data_science

We're using following parameters:

• -p 8888:8888 to export Jupyter Web interface
• -p 6006:6006 to export TensorflowDashboard Web interface
• -d to run Docker container in background
• -v notebooks:/notebooks to mount just created notebooks folder Docker inside container

Remove all container

docker rm `docker ps --no-trunc -aq` 

Check docker container details

docker inspect -f '{{.Name}} - {{.NetworkSettings.IPAddress }}' $(docker ps -aq)

© Andrei Maksimov / ProgKids.ru