Docker Python Data Webservices

A base docker image for building python webservices that perform data analysis.

Contents of image

The image is based on official python:3.6-slim image, with the following additions:

• numpy >=1.12
• pandas >=0.19
• netCDF4 >=1.2.9
• h5py >=2.7.1
• Flask >=0.12
• gunicorn >=19.7.1

Usage

This image should not be used as a standalone container, but rather as a base image for your own applications/containers. The image is bundled with an example flask application stored in /srv/app/ directory. The application is booted by gunicorn HTTP server, with configuration stored at /srv/gunicorn_cfg.py.

To use image, simply base your Dockerfile on it with:

FROM mastropinguino/python-data-ws:3.6

copy the application into /srv/app/ folder:

COPY . /srv/app/

and launch pip install if more packages are needed:

RUN cd /srv/app/ && pip install -r requirements.txt

Build and run the container

$ docker build -t awesome-app-image:latest .

$ docker run -d -p 80:8000 awesome-app-image:latest

A complete skeleton of app could be found in example folder

Contributions

Contributions are always welcome! Feel free to let me know your thoughts or improvements.