MLA-C01 Question 358
Select 2You are building a machine learning pipeline using AWS services to automate the training and deployment of a model. The pipeline is triggered whenever new code is committed to the model's repository. Which combination of services would best support this scenario?
- A
AWS CodeCommit, AWS CodePipeline, AWS SageMaker
- B
AWS CodePipeline, AWS Lambda, AWS SageMaker
- C
Amazon S3, AWS CodePipeline, AWS SageMaker
- D
AWS CodeCommit, AWS CodeBuild, AWS Glue
- E
AWS CodeCommit, AWS CodePipeline, AWS CloudFormation
Show answer and explanation
Correct answers: A, B
Explanation
In this scenario, AWS CodeCommit acts as the version-controlled code repository. AWS CodePipeline is used to automate the pipeline, integrating with other services. AWS SageMaker is the service that supports building, training, and deploying ML models. AWS Lambda is an additional option to trigger specific steps in the pipeline. This combination ensures an automated, end-to-end ML workflow.
- A. Correct.
Correct. AWS CodeCommit serves as the code repository, AWS CodePipeline orchestrates the pipeline, and AWS SageMaker handles ML model training and deployment.
- B. Correct.
Correct. AWS CodePipeline orchestrates the pipeline, AWS Lambda can trigger specific steps, and AWS SageMaker is used for ML model training and deployment.
- C. Incorrect.
Incorrect. While Amazon S3 can store data, it is not used as a code repository, and this combination lacks a proper repository service like AWS CodeCommit.
- D. Incorrect.
Incorrect. Although AWS CodeCommit and AWS CodeBuild are useful for code compilation, AWS Glue is not relevant for model training and deployment.
- E. Incorrect.
Incorrect. AWS CloudFormation is used for infrastructure provisioning, not for orchestrating ML pipelines.