Data partition for Comma and Udacity

Hello, I can't find the information about the data partition of Comma datasets, where can I find it? Also did you separately test on 11 videos and them averaged the results?

Also which Udacity dataset did you use, on the official github page there are challenge 2 and 3.  