AWS Certified Machine Learning – SpecialtyMachine Learning Implementation and OperationsMedium
An e-commerce company uses an ML model for real-time product recommendations. The model processes millions of inference requests per hour. Due to the high volume and the need for auditing and debugging, the company requires that all inference requests and responses, including the input and output data, be captured and stored for future analysis. Which SageMaker feature enables this capability?
- ASageMaker Data Capture
- BSageMaker Debugger
- CSageMaker Clarify
- DSageMaker Model Monitor
Show answer & explanationAnswer & explanation
Correct answer: A. SageMaker Data Capture
SageMaker Data Capture allows you to configure a real-time inference endpoint to automatically capture and save input and output data, as well as model predictions, to an Amazon S3 bucket. This data can then be used for model monitoring, debugging, auditing, and retraining purposes.
Why the other options are wrong
- B. Debugger is for analyzing model training jobs, not for capturing live inference data.
- C. Clarify is for bias detection and explainability, not for capturing raw inference data.
- D. Model Monitor uses captured data to detect drift and quality issues, but Data Capture is the feature that performs the actual capturing.
SageMaker Data Capture
A SageMaker feature that automatically captures and stores inference requests, responses, and predictions from real-time endpoints to Amazon S3.
- Crucial for auditing, debugging, and compliance.
- Provides data for model monitoring and retraining.
- Captures both input and output payloads.
Memory trick: Capture Data, See All, SageMaker's Audit Call.