Live 61: How to add ML to your web app with ml5js

In today's live stream—Live 61—I gave an overview on ml5js. What it is, where it comes from, and a hands-on coding session seeing a basic image classification example. (See the code on GitHub).

I did a few Live Q&As, including discussing the Differences between ml5js and TensorFlow JS, What's next after building an ML model?, and What camera do you use for live streaming?

ml5js is an extremely easy way to add machine learning to your web app, and I was surprised to see how fast the MobileNet image classification is running in Google Chrome on my Apple Silicon M1 MacBook Pro, even while live streaming, getting my Sony Alpha 6500's camera feed in real time.

If this is something that interests you, please let me know on Twitter or, even better, on the Discord community.

Thanks for watching.

See you next week!

• 01:00 Introduction
• 02:33 ml5js
• 15:17 Q&A - Differences between ml5js and TensorFlow
• 18:27 The ml5js library
• 20:30 Getting started
• 22:30 CDNs: Content delivery networks
• 24:36 MobileNet image classifier
• 33:41 Classifying Images with ml5js - Part 1
• 43:55 Q&A - What's next after building an ML model?
• 47:15 Classifying Images with ml5js - Part 2
• 01:13:30 DoodleNet
• 01:26:50 Real-time classification
• 01:29:30 ml5js in Next.js
• 01:34:45 Q&A - What camera do you use for live streaming?
• 01:36:14 Outro