Basic Information
| Field | Details |
|---|---|
| English name | Amazon SageMaker |
| Full name | Amazon SageMaker |
| Chinese description | 端到端机器学习平台 |
| Japanese description | 機械学習プラットフォーム |
| Exam frequency | ⭐⭐⭐ |
| Often confused with | Bedrock |
In one sentence
A managed platform to prepare data and build, train, deploy, and monitor custom machine-learning models.
Key points
- It covers data preparation, training, tuning, experiments, endpoint deployment, batch inference, and monitoring.
- Training learns from historical data; inference applies the trained model to new data.
- It fits predictions, classification, recommendations, and anomaly detection using proprietary data.
Exam focus
- Build, train, and deploy your own model points to SageMaker.
- Use AWS AI Services for pretrained APIs and Bedrock for managed foundation-model APIs.
Common pitfalls
- Do not choose by product name alone; confirm data type, latency, control, operations, and cost constraints.
Remember
Your data + custom model + full ML lifecycle = SageMaker.