So, let's recap over what we have covered. We looked at what GPUs are in depth as well as how we could utilize them for more general purpose tasks. We covered some of the more realistic scenarios that data scientists would typically encounter and why these are ideal scenarios for us to leverage these GPU wrapper libraries.

We then looked at some of the major libraries that exist today that allow us to leverage the full power of our graphics processing hardware. You should now have some idea as to how to get started writing your own GPU- as well as APU-based applications, whether this be for data science purposes or otherwise.

In the final chapter of this book, we'll take a look back at the different techniques we covered within this book and summarize some of the key places to use them.