Blog

Python and IoT: From Chips and Bits to Data Science

Ever want to know what is behind the "Internet of Things" hype? Back in Feburary, I wanted to as well, so I embarked on a side project to learn more. This talk is the story of my journey, using, of course, my favorite programming langauge, Python.

In this talk, I will take you through my project, a lighting replay system. I captured light sensor data (using Micropython and the ESP8266) in three rooms of my house. I then analyzed the data using Jupyter notebooks, Numpy, Pandas, and Scikit-learn. My goal was to replay realistic light usage when my family and I are out of town. After exploring several machine learning approaches, I settled on Hidden Markov Models (using Hmmlearn). I now have a simple application that runs on my Raspberry Pi and controls Phillips Hue lights based on the learned model. Along the way, I played around with hardware for the first time since college and co-developed an open source data filtering framework, called AntEvents (joint work with Rupak Majumdar of the Max Planck Institute for Software Systems).