Embed Your Machine Learning Model in Your App?
summary
Pl@ntNet describes converting a server ML model into a mobile-optimized runtime and shipping a 210 MB model plus metadata and example images to phones to cover 40,000 species. The app uses a React Native bridge to call native Java/Swift code, resizes input photos to ~184 px, runs inference on CPU, evaluates top-1/top-5 against a downloaded test set, and implements download/update management plus handling format regressions and cross-platform preprocessing differences.
AI-written summary · about
chapters (5)
- 0:00Pl@ntNet app context
- 2:50Embed ML in mobile app
- 7:29Download local data
- 10:17Process results
- 12:47Production feedback after 1 year
related talks
10mMachine Learning And React Native
07mMachine Learning on the Edge
29mPushing Machine Learning into your app When TensorFlow and MLKit rule the world
17mDeploy a custom Machine Learning model to mobile
25mGetting Started with Machine Learning in your iOS apps
26mBuild your Machine Learning Model on Edge with React Native
39mMachine Learning on Your Device: The Options
53mBuilding Intelligent .NET MAUI Apps with ML.NET