Congratulations on coming to the end of this first course, and well, you've come a long way. Yeah, I mean, we've looked at with the students like you looked at how to use TensorFlow, all the way from the beginning with doing something like fitting a straight line, to then like recognizing fashion items. It's really cool. It's like three lines of code to do a really complicated task like that. Isn't it amazing, computer vision in three lines like that? Then, we improve that by adding convolutions. We call it an improvement, but that means now four lines of code to define two conv layers and two maxpool layers, and then a two like TensorFlow take cares of a lot of the rest. Then, we looked at going beyond the simple images to bigger and colored images, and we didn't even add any lines of code for that, right? It was just changing the shape of the data coming in. Yeah. So by learning a framework like this, you can write very complicated programs with just a few lines of code, and this helps people save a lot of time. But there's still a lot more to learn. Oh there is, like some of the things that we'd love to explore are things such as, when you have very small data sets that can lead to an error called over-fitting. So we'll explore some techniques and tools that we can use to avoid that. Transfer learning. If you can download someone else's say TensorFlow model, and use that for your own problem, even though I was trained on a totally different data set, TensorFlow has tools that let you do that efficiently as well. Absolutely, and one of the things that a lot of AI practitioners love to be involved in is like this online competitions, where places like Kango provide a data set, and they ask you to build a classifier around that data set. We're going to explore some of the skills that you can use to take part in those competitions. So congratulations again on coming to the end of the first course and there's still all these amazing things to learn about TensorFlow. So please go on to the next course.