Week 7

At this point we have successfully extracted a reasonable set of features, incorporating all of the data sets we have access to. Up to this point I have only been running the relatively simple LinearRegression on our test models. This week I experimented with a bunch of other regression algorithms provided by Python's machine learning library sklearn. Using different algorithms I was able to improve upon our predictive accuracy (as measured by metrics such as MAE, RMSE, and R2 score) both for predicting T3 and final grades (see table below). Peter and I also made a lot of updates to our paper write up including filling out the lit review section.

Our next task is to start working on our video and poster presentations, both of which are due by the end of next week.