Adventures in iOS machine learning, part 2: Let’s build our own Core ML models!

• What we'll do
In our last meetup, we built a simple app that could analyze a photo and say if it was a picture of healthy or unhealthy food. It used a CoreML model that someone trained by giving it hundreds of photos of food and classifying each as healthy (mostly photos of fruit) and unhealthy (photos of pastries, cookies, and other sugary foods).

Someone asked “How do make our own models?” at the end of that meetup. In this meetup, we’ll answer that question by going through the steps to build our own model using Python and Turi Create, and then incorporating that model into a Core ML app.

If you’ve always wanted to get into machine learning, but found the material on it intimidating, you’ll want to join us at this friendly meetup, where you’ll get than hang of machine learning, iOS-style!

• What to bring
This meetup will be a “code along with me” exercise. That means you’ll need to bring a Mac laptop with Xcode 10 installed. You’ll probably also want to bring an iOS device along, although this exercise should run just fine in the Simulator.

You’ll also want to make sure that Python is installed on your machine. I’ll email additional instructions as we get closer to the event date.

• Important to know
Nobody learns well on an empty stomach, so we’ll provide pizza, soft drinks, and water, courtesy of my employer, our venue-provider and sponsor: Sourcetoad!