Data Science Software Stack Docker Prototype

Docker images providing the following data science software stacks for peronsal use:

• Anaconda Python with Jupyter Notebook
• Rstudio Server
• Apache Spark (Stand-alone)
• Tensorflow (cpu and gpu versions)
• h2o
• xgboost
• lightgbm

See wiki for additional information.

Built docker images can be found at dockerhub.com

This work inspired by these talks:

• Docker for Data Scientists
• Capital One Analytic Garage on Docker

System Requirements:

• MacOS
• Docker for Mac 18.06.1 ce (requires kubernetes enabled)
• Chrome Browser

Directories:

• bin scripts to start and stop containers
• h2oai h2o Flow server
• jpynb Anaconda Python with additional packages: h2o, xgboost, lightgbm
• pyspnb Ancaonda Python with stand-alone Spark
• rstudio Rstudio Server (Community Edition)
• tfcpu Tensorflow with Python 3 and Jupyter Notebook (cpu)
• tfgpu Tensorflow with Python 3 and Jupyter Notebook (gpu)