How to create personalized digital & mobile experiences with machine learning
summary
A working prototype called Boot Finder uses Turi Create to train an image similarity Core ML model, ships that artifact into an iOS app via Xcode and Vision, and returns a ranked set — oddly reporting five hundred five thousand seven hundred eighty-six similar boots for one query. A Python SDK uploads new model versions and a Swift framework reloads them on device so inventory-driven models update without republishing the app.
AI-written summary · about
chapters (12)
- 0:00Start
- 0:58Outline
- 1:19Why Machine Learning?
- 4:11Example Machine Learning Use Case
- 8:41What do I need?
- 9:16Why boot finder?
- 9:53User Experience: Basic app or Embedded Feature
- 10:15Workflow
- 13:12A Problem!
- 14:24Decoupling App & Model Delivery
- 15:32Machine Learning Code
- 18:11Wrap-up
related talks
22mGetting Started with Apple ML
50mBuilding Intelligent Mobile Apps using Machine Learning
40mNavigating the Machine Learning Landscape for Apple Devices
39mCustomize Your App with MLKit
01h 03mTraining dynamic ML models on iOS 15
10mMachine Learning And React Native
50mBuilding Intelligent Mobile Apps using Machine Learning
15mEmbed Your Machine Learning Model in Your App?