ChiPy Data SIG presents Property Testing Pandas and Risk Weighting

Chicago Python Data Special Interest Group's (SIG) September meeting will be held on September 18th. This month's meeting is graciously hosted and sponsored by Braintree!

AGENDA
6:00 - Doors open
6:30 - Talks Start
8:00 - See you all next time!

------------

Our talks:

Property Testing Pandas with Bulwark
by Zax Rosenberg

While testing has long been shown to reduce bugs and lower development costs, tests are sorely lacking in much of the data science world. Some of the reasons for this lack of tests are that data scientists/analysts often don't have the time, knowledge, or even exact data to write useful tests. Property testing, which involves checking that an object has certain properties, can allow us to still write tests even under such circumstances.

Since a lot of data work is done in Python, and Pandas is one of the go-to libraries for analyzing tabular data in Python, wouldn't it be nice to be able to add property tests to our DataFrames while we code? Enter Bulwark - a pip-installable property testing library for Pandas DataFrames that makes it easy to check that your data meets your assumptions of what it should look like at any (and every) step in your code, without making you work too hard.

This talk is for anyone who works with data in Python, and no prior testing experience is required. It will expand the definition of property testing, why it's useful, and demonstrate multiple examples of how to easily add property tests to code with Bulwark.

Speaker Bio:

Zax is currently a Senior Data Scientist at SPINS, where he develops data-driven analytic solutions for grocery retailers and CPG companies. He is also the Director of ChiPy‘s (Chicago Python User Group’s) mentorship program, co-host of ChiPy’s Project Night, and an avid tech enthusiast/blogger.

----

Risk Weighting: The Best Thing Since One Hot Encoding!
by Laurel Ruhlen

One Hot Encoding (OHE) is a common method for turning categorical variables into numbers. But what if there were a better way? This talk introduces a risk-weighting technique for wrangling categorical variables into submission. Also included: sklearn pipeline compatible risk weight transformer!

Speaker Bio:

Laurel Ruhlen is a data scientist at Braintree. She is a recovering physicist and a pool noodle enthusiast.

----

Braintree is hiring for the following positions:
https://www.braintreepayments.com/careers#positions