QuestionQ35

VerifiedML Model Development

A medical company receives streams of data from devices that monitor patients’ vital signs. The company uses Amazon SageMaker and intends to prepare ML models that predict adverse patient events. The dataset is large and has thousands of features.

An ML engineer must run several hundred training iterations using different feature sets, different algorithms, and numerous possible parameters. The ML engineer must implement a solution that logs the characteristics and results of every training iteration.

Which solution meets these requirements with the LEAST implementation effort?

Explanation

Amazon SageMaker Experiments is purpose-built to record and organize ML training iterations. It tracks run inputs, configurations, parameters, metrics, artifacts, and results, enabling comparison across experiments without building custom logging and query infrastructure.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!