MLA-C01 Question 85
Select 3You are building a machine learning model to predict customer churn for an e-commerce platform. You decide to use Amazon SageMaker to train and deploy your model. During the feature engineering phase, you need to preprocess a large dataset stored in Amazon S3. Which of the following AWS services or features can you use to efficiently preprocess your data for training?
- A
Amazon SageMaker Processing Jobs
- B
AWS Glue
- C
Amazon SageMaker Ground Truth
- D
Amazon EMR
- E
Amazon SageMaker Debugger
Show answer and explanation
Correct answers: A, B, D
Explanation
Efficient preprocessing of large datasets in S3 is a critical step in machine learning workflows. Amazon SageMaker Processing Jobs, AWS Glue, and Amazon EMR all support this objective and integrate well with S3. SageMaker Ground Truth is not relevant to preprocessing as it focuses on data labeling, and SageMaker Debugger is used for debugging during training, not preprocessing.
- A. Correct.
Amazon SageMaker Processing Jobs are designed to preprocess and transform data at scale. They are an efficient way to manage preprocessing workflows directly within Amazon SageMaker.
- B. Correct.
AWS Glue is a fully managed ETL (Extract, Transform, Load) service that can be used to preprocess and transform large datasets in S3 before they are used for training.
- C. Incorrect.
Amazon SageMaker Ground Truth is meant for labeling datasets, not for preprocessing or transforming data for machine learning.
- D. Correct.
Amazon EMR (Elastic MapReduce) is a scalable big data platform that can process large datasets efficiently and can be used for data preprocessing.
- E. Incorrect.
Amazon SageMaker Debugger is used for monitoring and debugging during training, but it does not provide data preprocessing capabilities.