Old School - Functional Data Analysis

This talk will use core functionality from the PyToolz projects. Students will leave both with a set of concrete tools and with an understanding of some of the more applicable lessons from the functional style.

Presented at PyData NYC 2013.

The tutorial video is now available on Vimeo.

• Python > 2.6
• Toolz - pip install toolz

The following links will open the tutorial notebooks in IPython's online notebook viewer:

• map-filter-reduce-groupby
• composition
• curry
• wordcount
• laziness
• genomics
• fib-memoize
• groupby-github

Data for most of the notebooks resides in the data directory

The human genome data can be found here

The Github data can be found here

After the talk Jake Vanderplas showed me that indeed numpy does support accumulation on any binary operator. Try the following in ipython for a list of supported binops

import numpy
numpy.*.accumulate?