Explore the intricacies of deploying machine learning models in production through this comprehensive 53-minute lecture. Delve into various deployment strategies including batch prediction, model-in-service, and model-as-service approaches. Gain insights on implementing REST APIs, managing dependencies, and optimizing performance on single machines. Learn about horizontal scaling techniques, model deployment best practices, and managed options for streamlined operations. Discover the potential of edge prediction for real-time applications. Master the essential skills needed to transition ML models from development to production environments effectively.
Deploying ML Models - Full Stack Deep Learning - Spring 2021
Overview
Syllabus
- Introduction
- Batch Prediction
- Model-In-Service
- Model-as-Service
- REST APIs
- Dependency Management
- Performance Optimization Single Machine
- Horizontal Scaling
- Model Deployment
- Managed Options
- Edge Prediction
Taught by
The Full Stack
