Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

.NET Fest 2018. Михаил Галушко. Искусственный интеллект на платформе Xamarin

14 views

Published on

Искусственный интеллект продолжает быть глобальным трендом. В данный момент он позволяет нам создавать приложения "с интеллектом", что открывает дополнительные возможности для разработчиков. Xamarin не имеет собственных AI или Machine Learning возможностей, но является важным инструментом, который позволяет использовать различные интеллектуальные системы, собирать данные и отображать их. В докладе мы рассмотрим некоторые сценарии использования AI/ML на платформе Xamarin.

Published in: Education
  • Be the first to comment

  • Be the first to like this

.NET Fest 2018. Михаил Галушко. Искусственный интеллект на платформе Xamarin

  1. 1. Xamarin + AI
  2. 2. Agenda 1. Xamarin overview 2. AI for mobile platform 3. Cognitive Services overview 4. Demo
  3. 3. Anything you can do in Objective-C, Swift, Java or Kotlin can be done in C# and Visual Studio with Xamarin.
  4. 4. Native Performance
  5. 5. Always up-to-date Same day support for new versions Support for Apple Watch, KindleFire, Android Wear etc
  6. 6. Xamarin vs * iOS/Android vs Hybrid
  7. 7. Xamarin.Forms Native UI for each platform Platforms: iOS, Android, UWP, WPF, MacOS, Tizen, GTK#
  8. 8. Xamarin + Xamarin.Forms Traditional Xamarin Approach With Xamarin.Forms: More code-sharing, all native iOS C# UI Windows C# UIAndroid C# UI Shared C# Backend Shared UI Code Shared C# Backend
  9. 9. Why you should use Xamarin • You are C# developer • Reuse existing code • Cross-platform • Continue in Microsoft ecosystem
  10. 10. Why you should use Xamarin.Forms • Fast POC • Minimize costs for development for multiple platforms:  common interactions across device platforms  code sharing is more important than custom UI  little platform-specific features • You are focusing on LOB apps • You are not writing games • You would like to partially reuse code from WPF/UWP
  11. 11. AI for mobile platform 1. Consuming AI: Cognitive Services, ML Kit for Firebase 2. AI on devices • ONNX: open neural network exchange format Cross-platform: Caffe2, TensorFlow, Core ML • iOS: Core ML • Android: TensorFlow
  12. 12. Microsoft AI
  13. 13. Pre-built RESTful APIs Azure Cognitive Services 1. Search 2. Speech 3. Language 4. Knowledge 5. Vision
  14. 14. ONNX Open neural network exchange format Platforms: Caffe2, TensorFlow, Core ML etc Create model and use on multiple platforms – no vendor lock
  15. 15. iOS: Core ML • Domains: Vision, NLP • Built on Accelerate and Metal • Public model format • Conversion tools
  16. 16. Android: TensorFlow • Supports Android/iOS/Raspberry PI • Lite version for mobile
  17. 17. Android: ML Kit for Firebase • Supports: iOS/ Android • Integrates with Google Cloud Services • Text recognition • Face detection • Barcode scanning • Image labeling • Landmark recognition
  18. 18. Demo
  19. 19. Links • https://customvision.ai • https://docs.microsoft.com/en-us/azure/cognitive- services/custom-vision-service/home • https://github.com/jimbobbennett/Xam.Plugins.OnDeviceC ustomVision • https://blogs.technet.microsoft.com/machinelearning/2018 /04/03/intelligent-edge-building-a-skin-cancer-prediction- app-with-azure-machine-learning-coreml-xamarin/ • https://github.com/Microsoft/Mobile-Chest-X-Ray-Analysis
  20. 20. Q&A mykhail.galushko@devrain.com

×