Public Code

Here is code I released under the open source MIT license for all to use. If you use it, please cite the corresponding paper if applicable! Citations help fund new research =).


You can pip3 one of my creations, dictsearch, which is a Python module that allows you to grab a list of all leaf nodes in a multi-level nested dictionary. The PyPi page is here and the public github repo is here.

WeBike data portal

I built the website for WeBike (Waterloo's electric bike field trial) participants to log in and view their data. The website is here (requires login) and the source code is here (prod) and here (dev). The portal has several visualization tools for viewing trip statistics and monitoring battery charge levels.

electric vehicle sentiment analysis

Here is the system I built for our submission "Is The Grass Greener?" (in my publications section). It is a complete online sentiment analysis system built in Python. It crawls designated web forums, parses and cleans the posts, classifies them, and summarizes the results on a feature-wise basis (e.g., "range", "maintenance", etc). Currently the Volt, Leaf and Tesla Model S are supported. Everything you need to start mining these three forums is included.

fun with cuda

I write a lots of NVIDIA CUDA code because my research uses lots of discrete math. CUDA can be quite overwhelming, so simply releasing the kernels alone would not help new users learn the language. Thus, when I have time I will wrap various useful kernels and release them here! Here is a blazing fast matrix multiplication kernel.