gogplot - ggplot2 with pandas from python

I figure any self-respecting data person these days has their own attempt at making ggplot2 available from Python, especially with pandas integration. So I just had to start my own...

In all seriousness though, ggplot2 is an absolutely wonderful implementation of the Grammar of Graphics and is an indispensible tool for data analysis. Similarly pandas too is a brilliant toolset for working with data in Python. And it actually is a real problem that Python does not have a proper ggplot2-like package.

Various attempts at making such a package are under way. That's great but none of them are complete or stable enough yet. So I decided to make the simplest implementation I could think of for my own usage. It might even be more feature complete than the alternatives? I guess I'll find out soon enough :)

Dependencies

For this to work you will need both Python with pandas and R with ggplot2 installed. How to do this has been documented in many other places.

Quick start

$ git clone [email protected]:leondutoit/gogplot.git
$ cd gogplot
$ python setup.py install
$ ipython

import pandas as pd
import gogplot as gp

# a simple scatterplot
df = pd.DataFrame({'x': [1,2,3,4,5], 'y': [33,7,60,5,43]})
gp.gogplot(df, 'ggplot(df, aes(x, y)) + geom_point()')