When I started learning web dev, I dove into building my first Rails app. I didn’t know how it all worked, but after hours of hacking I had a blogging app running. I imagine many share a similar learning experience. Similarly, you don’t need a PhD to do deep learning, you can get started with Python skills and open-source frameworks. It can be fun and rewarding, and inspire you to dive deeper.
You know how to code in Python, and you want to learn more about deep learning. This talk will put the tools in your hands to complete your first deep learning project, and put you on the path to growing your knowledge of this exciting field.
Along the way, we’ll discuss the learning process, and how you can leverage what you already know in order to jump start your progress tackling something new. I’ll even take a detour to talk (briefly) about how I got started in web development with Rails, and the benefits you can get from open-source frameworks with high levels of abstraction.
I’ll walk through the pieces you can put together to build a powerful image classifier with deep learning: hardware (GPUs), training data, model architectures and pretrained weights, and software libraries for running training and prediction with your models. Since all of these things are becoming increasingly available for free on the internet, it’s never been a better time to give deep learning a try.
In the end, you don't need a PhD to get started doing deep learning. You can have fun learning enough to do a simple project, like a cats vs. dogs image classifier. From there, you can go wherever your curiosity takes you!
William Horton is a Backend Engineer at Compass, where he works on systems for ingesting, processing, and serving millions of real estate listings. He’s been studying deep learning in his spare time for the past year, most recently taking the fast.ai course “Cutting Edge Deep Learning For Coders” as an International Fellow. He has written about his recent work in deep learning on Medium. When he’s not doing tech things, he enjoys powerlifting and singing a cappella.