This repo contains a set of my works about learning and active learning on distributions. The code has the generality to work for both input and distributional space examples. Since I have written the code for my works without the intention to release it, it doesn't have documentation, and some parts may not be in good shape for someone unfamiliar with it. In the meantime, I hope that I find the time to reshape the code and improve it.