My masters thesis work: predicting the Sacramento River's natural flow using a variety of machine learning techniques.
This code makes use of my mlutilities package (which you can check out in its repo here) as well as sklearn and pandas, among others.
For a quick overview of my results, take a look at these slides.
If you want to know all the gritty details, here is a copy of my thesis.