This project was similar to the weather prediction algorithm, however, instead of a autocorrelation model this used a random forest algorithm. Also unlike the autocorrelation model this model predicted the temperature of the next day, as opposed to three days, and had a mean absolute error of 4.44. You can read more in the blog section.

A comparison of my random forest model’s predictions, my autocorrelation model’s predictions, and the real temprature.