QuestionQ56

VerifiedDeployment and Orchestration of ML Workflows

A company has created a computer vision model and must deploy it to production on Amazon SageMaker AI. The company has not previously hosted a model on SageMaker AI.

An ML engineer must implement a solution to track model versions. The solution must also recommend which Amazon EC2 instance types to use for hosting the model.

Which solution satisfies these requirements?

Explanation

SageMaker Model Registry registers models as versioned model packages within model groups, providing model-version tracking. SageMaker Inference Recommender runs recommendation jobs and load tests to identify instance types that best fit an inference deployment’s cost, latency, and throughput requirements.

Learn more

Community Discussion

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