Practical Machine Learning for Front End Developers
summary
Shows JavaScript prototypes that actually run: an on-page object detector using Coco SSD to flag recycle bins, a head-motion “teachable keyboard” rebuilt with MobileNet and transfer learning, and a canvas doodle classifier trained from scratch (about 200 hand drawings vs QuickDraw’s ~200k). Uses TensorFlow.js and ml5, notes model downloads (≈15 MB), training can take hours, and highlights limits — data volume, mobile cost, explainability and bias — with demos that mostly work but require careful preprocessing and parameter tuning.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:58DIFFERENCE FROM TRADITIONAL PROGRAMMING
- 3:59EXAMPLES OF ALGORITHMS
- 4:23TYPES OF MACHINE LEARNING ALGORITHMS
- 5:02SUPERVISED LEARNING
- 8:02GENERATING ALT TEXT
- 8:48FILTERING UNWANTED CONTENT
- 9:56ACCESSIBILITY
- 13:15WHAT CAN YOU DO
- 13:53TOOLS
- 15:22RECYCLING
- 17:39THE TEACHABLE KEYBOARD
- 19:47CODE SAMPLE
- 21:39THE WILLY DETECTOR
- 24:46PROCESS
- 27:51LIMITS
- 31:32INSPIRATION
- 35:17DATASETS
related talks
41mPractical Machine Learning for Front-End Developers
34mMachine learning for front-end developers
29mMachine learning for front-end developers
34mA Powerful Introduction to Machine Learning with Js
34mA Powerful Introduction to Machine Learning with Js
41mTensorFlow.js 101: ML in the Browser and Beyond
24mThe Future of Machine Learning & Javascript
35mThe Future of JavaScript & Machine Learning