MLA-C01 Question 87
Select 3You are a Machine Learning Engineer tasked with building a machine learning model to predict customer churn for a subscription-based streaming platform. The dataset contains millions of records with various features like demographics, subscription history, and engagement metrics. To efficiently train the model on this large dataset, which AWS services and features should you leverage?
- A
Amazon SageMaker Pipe mode to stream data directly from Amazon S3 into the training instance
- B
Amazon SageMaker Feature Store to preprocess and store the features for consistency across training and inference
- C
Amazon S3 Multipart Upload to split the dataset into smaller parts and upload it to S3
- D
AWS Glue to perform ETL (Extract, Transform, Load) operations on the dataset before training
- E
Amazon Rekognition to extract visual features from customer profile pictures
Show answer and explanation
Correct answers: A, B, D
Explanation
When working with large datasets for machine learning, it is important to leverage AWS services designed for scalability, feature management, and efficient data processing. Amazon SageMaker Pipe mode enables streaming data directly from S3, eliminating the need to load the entire dataset into memory, which is particularly useful for large datasets. Amazon SageMaker Feature Store ensures consistent feature engineering across training and inference pipelines. AWS Glue provides the ability to preprocess and transform data at scale, which is an essential step in preparing the dataset for training. These services collectively address key challenges in large-scale machine learning workflows.
- A. Correct.
Correct: Pipe mode streams data directly from Amazon S3 into the training instance without loading the entire dataset into memory, making it suitable for large datasets.
- B. Correct.
Correct: Amazon SageMaker Feature Store ensures consistent feature engineering for both training and inference, which is crucial for reproducibility and performance.
- C. Incorrect.
Incorrect: Amazon S3 Multipart Upload is used to upload large files to S3 but does not directly address the challenges of training on large datasets.
- D. Correct.
Correct: AWS Glue can preprocess and transform large datasets efficiently, making it a valuable tool for preparing data before machine learning training.
- E. Incorrect.
Incorrect: Amazon Rekognition is used for image and video analysis, which is irrelevant for a customer churn prediction task based on demographics and engagement metrics.