There are two data sets, one of images of digits and the other of faces. The algorithms, Naive Bayes, Perceptron, and k Nearest Neighbor attempt to determine which digit the digit image is and whether the face image is a face or not.
Run .\TA_PRESENTATION.PS1 in powershell to see the basic performance of each algorithm. You may change some parameters in this file to your liking as well.
The shell for the classifications can be found at Berkeley here, http://inst.eecs.berkeley.edu/~cs188/sp11/projects/classification/