Exploring Neural Networks for Recognition of Hand-drawn Shapes – Chris Parrish

Along with the boom in practical applications for artificial intelligence, the popularity of Neural Networks for machine learning has exploded. At the same time, the iPad Pro is the first iOS device to support a native stylus. This makes today a unique and exciting time to explore handwritten user interface methods in your apps.

We'll cover the basic concepts and then discuss some of the challenges and art in building neural networks including leveraging Apple's CoreML and CreateML. We'll use a demo iOS app to demonstrate the feasibility of using machine learning to recognize hand-drawn shapes such as circles, squares, and triangles.


About Chris Parrish:

Chris is one of the founders of Aged & Distilled, developers of the popular Mac application Napkin. Chris has been programming for Apple's platforms for over 3 decades, starting with assembly language on an Apple II+. He has contributed to dozens of high profile applications including Postage, the 2009 Apple Design Award winning iPhone app. In addition to his development work, Chris is the co-host of The Record podcast with Brent Simmons.

@twenty3 / therecord.co