AWS Certified Machine Learning Engineer Associate · Free Practice Question Medium
Question 26
A retail company uses Amazon SageMaker to train machine learning models for product recommendation. The team stores raw customer behavior data, such as clicks and purchases, in an Amazon S3 bucket. They want to preprocess this data to remove duplicates, normalize numerical features, and encode categorical variables before training their model. Additionally, the preprocessing needs to scale to handle increasing data volumes.
Which solution will most effectively meet their requirements?
-
A
Write a preprocessing script in Python and run it on Amazon EC2 instances with auto-scaling enabled.
-
B
Use Amazon SageMaker Processing Jobs with a pre-built container for data preprocessing and store the processed data back in S3.
-
C
Use Amazon Athena to query the data and perform all preprocessing using SQL, then export the processed data back to Amazon S3.
-
D
Leverage AWS Glue for ETL to preprocess the data, and then load the transformed data into Amazon SageMaker.
Reveal correct answer
Correct answer: B
A.
Running custom scripts on EC2 instances introduces complexity and lacks the managed, scalable environment of SageMaker Processing Jobs.
B.
Amazon SageMaker Processing Jobs provide a managed environment to preprocess data at scale, leveraging pre-built containers for tasks like normalization and encoding. This approach is scalable, integrated with SageMaker, and ideal for ML pipelines.
C.
Athena is primarily used for querying data, not preprocessing tasks like feature scaling or encoding.
D.
AWS Glue is powerful for ETL tasks but less tailored for ML-specific preprocessing needs compared to SageMaker Processing Jobs.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
