Plotting American Community Survey data with Python Bokeh

This is a simple demo created for PyData 2014 showing how to package a simple Python Bokeh application using Docker and Ferry. This demo uses the following technologies:

• Docker
• Ferry
• Bokeh

To make things interesting, it also uses data from the American Community Survey. This survey contains economic data at the county and zip code levels for the United States. The ACS and U.S. Census provides a nice API to access this data. You can find the scripts that downloads/parses that data under census.

Getting Started

This demo assumes that you have both Docker and Ferry installed. If neither are true, you can grab the Ferry Vagrant box like this:

        $ vagrant box add opencore/ferry https://s3.amazonaws.com/opencore/ferry.box
        $ vagrant init opencore/ferry
        $ vagrant up

This box has both Docker and Ferry pre-installed. Now onto the demos!

There are actually two demos, one relying solely on Docker and another that uses Ferry and incorporates Cassandra. The Docker demo can be run using the command:

	$ docker pull ferry/pydata
	$ docker run -p 8000:8000 -d ferry/pydata

Afterwards, point your browser to localhost:8000 and you should see a very simple web page with links to a few states.

To run the more advanced Ferry application, you'll have to clone this repository first (this isn't actually true, you just need one file in this repository):

	$ git clone https://github.com/jhorey/pydata

Then type:

	$ ferry start pydata/cassandra.yml -b pydata/

If you are running the latest version of Ferry or if you are restarting the Ferry demo, you can skip the -b pydata/ part (which tells Ferry to manually build the image). If you take a peek in the pydata/cassandra.yml file, you will see:

backend:
   - storage:
        personality: "cassandra"
        instances: 1
connectors:
   - personality: "ferry/pydata-cassandra"
     ports: ["8000:8000"]

This file just tells Ferry how to build your application. You can see that it relies on Cassandra and uses exposes port 8000 (just like the plain Docker demo).