Question 16
Domain 2: ML Model DevelopmentA machine learning team wants to compare training results from multiple model runs in one place so they can track metrics across those runs over time. Which SageMaker feature is designed for this purpose?
Correct answer: A
Explanation
Use SageMaker Experiments when you need to organize and compare metrics across multiple training runs under a common tracking structure. — SageMaker Experiments for tracking metrics across runs.
Why each option is right or wrong
A. SageMaker Experiments
The stated need is to track and compare metrics across multiple runs, and the source identifies SageMaker Experiments for tracking metrics across runs.
B. SageMaker endpoints
Endpoints are for deploying models for inference, not for tracking metrics across runs.
C. SageMaker notebooks
Notebooks provide a development environment, not a feature specifically for tracking metrics across runs.
D. SageMaker ground truth
Ground Truth is used for data labeling, not for comparing run metrics.