This is the documentation of my talk on fairness and transparency in machine learning that I gave at PyData Berlin in 2016. Please bear with me while I clean up the code and data, if you have questions please don't hesitate to contact me (contact info here: https://7scientists.com/contact).
For the talk, I used a publicly available data set from the NYC Police Department, which can be downloaded here:
http://www1.nyc.gov/site/nypd/stats/reports-analysis/stopfrisk.page