Amazon SageMaker
Amazon SageMaker is a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly.
Key Features
- Build: Provides hosted Jupyter notebooks to explore and visualize data.
- Train: One-click training of models with built-in algorithms or custom code.
- Deploy: Automatic deployment of models to hosting services with auto-scaling.
- SageMaker Ground Truth: A fully managed data labeling service (Human-in-the-loop).
- SageMaker Autopilot: Automatically builds, trains, and tunes the best ML models based on your data (AutoML).
- SageMaker Model Monitor: Continuously monitors the quality of your ML models in production (detects drift).
Exam Tips
- "Build, Train, Deploy": If you see these three words together, the answer is SageMaker.
- "Notebooks": SageMaker provides managed Jupyter Notebooks for data science.
- "Labeling Data": Use SageMaker Ground Truth.
- "AutoML" / "No Code ML": Use SageMaker Autopilot or SageMaker Canvas.
- "Edge Devices": Use SageMaker Edge Manager.
Common Use Cases
- Fraud Detection: Training models to detect fraudulent transactions.
- Recommendation Engines: Building systems to recommend products to users.
- Predictive Maintenance: Predicting when machinery will fail.