Machine Learning in Mobile Applications

Learn how to apply the power of machine learning to mobile app development, using platforms such as IBM Watson, Microsoft Azure Cognitive Services, and Apple Core ML.

Machine learning is reaching the mainstream. With the new tools available to developers, it's now possible to implement machine learning features—voice, face, and image recognition; personalized recommendations; and more—in a mobile context. This course explores how to apply the power of machine learning to mobile app development, using platforms such as IBM Watson, Microsoft Azure Cognitive Services, and Apple Core ML. Instructor Kevin Ford demos each product, reviewing the different features and approaches to machine learning. He shows how to train and deploy models for natural language and visual recognition and how to generate statistical models for use in a Xamarin application. In chapter five, he compares client-side and server-side models and explains when a developer might choose one platform over another.

Introduction

• Machine learning in mobile apps
• What you should know
• Using the exercise files

1. Introduction to Machine Learning

• What is machine learning?
• Required concepts
• Why does this matter for my app?
• Training a model
• Machine learning vs. deep learning
• What can I do with machine learning?
• Server-side vs. client-side ML
• ML frameworks

2. Server Models: IBM Watson

• Overview of Watson
• Natural Language Understanding: Set up
• Natural Language Understanding: Train the model
• Visual Recognition: Set up
• Visual Recognition: Train the model
• Create a custom model
• Train and deploy a custom model
• Install client SDK package
• Client tie to Natural Language
• Client tie to Visual Recognition call setup
• Client tie to Visual Recognition response
• Client tie to custom model: Get an access token
• Client tie to call custom model service
• Client tie to get custom model response
• Run the client app

3. Server Models: Azure Machine Learning

• Azure Machine Learning overview
• Language Understanding: Set up
• Language Understanding: Intents
• Language Understanding: Utterances
• Custom Vision: Set up
• Machine Learning Studio: Set up
• Machine Learning Studio: Create model
• Machine Learning Studio: Publish model
• Install client SDK package
• Client tie to LUIS
• Client tie to Custom Vision model
• Client tie to custom model
• Client tie to custom model: Set up request
• Client tie to custom model: Make the call
• Run the clent app

4. Client Models: Core ML

• Core ML overview
• Core ML: Create Natural Language model
• Core ML: Create Visual Recognition model
• Client tie to Natural Language model
• Client tie to Visual Recognition model
• Client tie to Visual Recognition: Converting model
• Run the client app

5. Understanding the Offerings

• Different philosopies of the vendors
• Why client-side model vs. server-side
• When to use one or the other of these solutions

Conclusion

• Next steps