Deploying ML Models to Production with Amazon SageMaker
summary
Covers how to deploy machine learning models to production using Amazon SageMaker, discussing options for online (real-time) and batch predictions. Explains that SageMaker can serve models even if they were trained outside SageMaker and compares the different deployment alternatives available.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 4:10Why optimize deployment?
- 7:53Using Docker containers with SageMaker
- 14:42Batch inference
- 18:00Real-time inference
- 22:30Considering costs
- 23:10Multi-model endpoint
- 28:00Inference with SageMaker Pipelines
- 30:05Asynchronous inference
- 32:30Serverless inference
- 35:28SageMaker inference options
- 39:05How to choose your inference strategy
- 42:25In summary... what we covered today
related talks
34mBuild, train and deploy Machine Learning models with Amazon SageMaker
39mDevelop your ML project with Amazon SageMaker
32mBuild, Train, and Deploy Machine Learning Models on AWS
31mBuild, train, and deploy ML models using Amazon SageMaker
33mBuild, train, and deploy ML models using Amazon SageMaker
41mDeploy ML models for inference at high performance & low cost
19mUsing Amazon SageMaker to develop and deploy ML models
48mAmazon SageMaker — Survey of Ways to (Re)deploy Models on AWS