MLA-C01 Question 322
Select 2You are working as a Machine Learning Engineer for a company that wants to automate the deployment of their ML models to production using CI/CD pipelines. The models are trained in Amazon SageMaker, and the deployment involves building, testing, and deploying the models to an endpoint. Which of the following steps are required to set up an automated CI/CD pipeline for this use case?
- A
Use AWS CodePipeline to define stages for model building, testing, and deployment.
- B
Use Amazon SageMaker Model Monitor to automate endpoint monitoring.
- C
Use AWS CodeBuild to execute build and test jobs for the ML model artifacts.
- D
Configure AWS Lambda to trigger model retraining based on endpoint monitoring results.
- E
Use Amazon SageMaker Pipelines to orchestrate the CI/CD process and manage workflows.
Show answer and explanation
Correct answers: A, C
Explanation
To set up an automated CI/CD pipeline for deploying ML models to production, AWS CodePipeline is used to define the pipeline stages, and AWS CodeBuild executes the build and test jobs. Amazon SageMaker Model Monitor, AWS Lambda, and SageMaker Pipelines are valuable for other aspects of the ML lifecycle but are not directly involved in the CI/CD pipeline setup.
- A. Correct.
Correct. AWS CodePipeline is a CI/CD orchestration service used to define pipelines for building, testing, and deploying machine learning models.
- B. Incorrect.
Incorrect. While Amazon SageMaker Model Monitor is useful for monitoring endpoints, it is not directly involved in the CI/CD pipeline setup.
- C. Correct.
Correct. AWS CodeBuild is used to execute build and test jobs as part of the CI/CD process.
- D. Incorrect.
Incorrect. Although AWS Lambda can automate certain tasks, it is not required for setting up a CI/CD pipeline for this use case.
- E. Incorrect.
Incorrect. Amazon SageMaker Pipelines is used for managing ML workflows, but it is not a CI/CD orchestration tool like AWS CodePipeline.