 I'm going to talk about the Python-based reinforcement learning, artificial intelligence, and intelligence, and neural network library. I'm a student at a master's level at McGill. And I work in the Department of Music Technology. The reason I started working with PyBrain was we work in the laboratory of new instruments, digital instruments. And a lot of times what you need to do for digital instruments is you don't really have preset gestures. You don't know how an artist is going to play a given instrument. And so there are different ways of training your gestures on a new instrument. And one of them is using what's called implicit mapping. And we do implicit mapping with machine learning. So what I did was I had an instrument and I had about 64 signals coming from different sensors. And I used PyBrain as an implicit layer to train my gestures. So I had this kind of gesture, this kind of gesture, and different kinds of bending. And then you can train these. And then you activate your network. And it'll compute the results in real time. And then you connect those results to sound. So PyBrain is a collaborative research project. It's developed by researchers at the Dalmo Institute for Artificial Intelligence and the Technical University of Munich in Germany. There is a website, PyBrain.org. And the documentation is pretty extensive. It covers a series of algorithms, supervised learning, reinforcement learning, different kinds of artificial neural networks. There's plotting tools. You can read and write to XML. And they started building environments, specifically for gaming or for 3D. There are some environments for building toys. And it's generally a lot of fun to work with. I'm not really an artificial intelligence expert. But I found this to be really easy to get started with, both in terms of learning and in terms of, you know, learning about machine learning in general and about Python and about neural networks. So what I'm going to show you is a really quick example of how you would build an artificial neural network. And then I have two pictures of a demo of it before training and after training environment. So what you need to do, since there are so many different algorithms in PyBrain, and in general, I mean, you have to kind of know what you want, what you're going to be training. So you have to import a bunch of necessary sub libraries. Since I'm working, I decided for this demo to do a supervised training on a neural network and a back propagation training. So I imported the sub libraries. I taken supervised data set and the backprop trainer. And then I take a tool to build the network. And I also taken tools to read and write to XML. Then to create your network, you create the data set. And then you create a network. You can load your external data set either from an XML file or from a CSV file. And then you train it. The parameters for training are a bit more detailed. And then you can test it to see how well you did. So this is a demo that's on the PyBrain website. It's basically two sine waves that are superimposed on top of each other. And at the beginning of the training, they're just random numbers. And then after an hour of training, they're literally superimposed. So in general, I found this library to be really easy to use and relatively good performance. And I don't have a huge comparison base with other machine learning libraries. But for me, this worked perfectly. So do you have any questions or would you like to see the website?