AIF-C01 Question 49
Select 3You are tasked with building an end-to-end machine learning pipeline on AWS for a retail company. The pipeline needs to include data preprocessing, feature engineering, model training, and monitoring deployed models for data drift. Which combination of AWS services will best suit this pipeline?
- A
Amazon SageMaker Data Wrangler for data preprocessing
- B
Amazon SageMaker Feature Store for storing and managing features
- C
Amazon SageMaker Model Monitor for monitoring data drift in production
- D
Amazon Rekognition for model training
- E
Amazon SageMaker Studio for training and monitoring models
Show answer and explanation
Correct answers: A, B, C
Explanation
The correct answer combines AWS services that align with each stage of the ML pipeline. Amazon SageMaker Data Wrangler is used for data preprocessing and feature engineering. Amazon SageMaker Feature Store is used to store and manage engineered features. Amazon SageMaker Model Monitor is used to monitor models in production for data drift. Amazon Rekognition and Amazon SageMaker Studio are not directly relevant to the specified pipeline requirements.
- A. Correct.
Amazon SageMaker Data Wrangler is designed for data preprocessing and feature engineering, making it an ideal choice for this stage of the pipeline.
- B. Correct.
Amazon SageMaker Feature Store is a purpose-built service to store and manage features, enabling consistent and reusable feature engineering.
- C. Correct.
Amazon SageMaker Model Monitor is designed to monitor deployed models for concept or data drift, making it the correct choice for the monitoring stage of the pipeline.
- D. Incorrect.
Amazon Rekognition is an AI service for image and video analysis, and it is not used for model training within a custom ML pipeline.
- E. Incorrect.
Amazon SageMaker Studio is a development environment for building, training, and debugging models, but it is not directly responsible for data preprocessing or monitoring.