initiumsp / data-urchin

The boilerplate and utilities for Data Science in Python3

Data Urchin

The boilerplate and utilities for Data Science in Python3.

Features

• Python3
• Hand-picked, de facto standard Python libraries
• Use IPython Notebook
• Time-proven practices in Initium Lab
• Boilerplates, sample codes, cheat sheets, quick hacks, ...
• [TODO] Integration with production-ready Javascript libraries
• [TODO] Workflow for Continuous Deployment of mining results

Usage -- Docker Compose

• docker-compose up.
• Visit port 8888 on docker-machine ip XXXX to operate the notebook.
• Be default, this current dir is mapped into the container as working folder. You can pass code/data in directly.

Usage -- Docker

Get the docker image

Choose either way:

• docker pull hupili/urchin
• docker build -t hupili/urchin .

Run The Notebook

docker run -v urchin:/app -p 8888:8888 -it hupili/urchin ipython notebook --ip 0.0.0.0

You can find the volumen on your docker-machine in following folder: /var/lib/docker/volumes/urchin/_data

Usage -- Plain Format

Initial setup:

• Fork and Clone this repository
• Instal Python3
• virtualenv -p python3 venv or pyvenv venv
• source venv/bin/activate
• pip install -r requirements.txt

Following usage:

• source venv/bin/activate (save time if you use virtualenvwrapper. .venv is configured)
• Use ipython notebook to launch the environment
• Copy any interesting stuff from boilerplates to the root and hack away

Sample pages

• http://project.initiumlab.com/data-urchin/Sample.html