AWS Certified Machine Learning Engineer Associate · Free Practice Question Medium
Question 31
A financial services company deployed a machine learning model using Amazon SageMaker Asynchronous Inference in the past with successful performance. Now, the company needs to deploy a new ML model that detects fraudulent credit card transactions in real-time within their banking application. However, when using SageMaker Asynchronous Inference for this new model, the performance is poor and does not meet the real-time requirements. Additionally, the company wants to receive notifications whenever there is a deviation in the model's quality.
As an AWS Certified Machine Learning Engineer Associate, what do you recommend?
-
A
Retain SageMaker Asynchronous Inference and enable autoscaling for the endpoint to improve performance, while using Amazon SQS for notifications
-
B
Use SageMaker Batch Transform for inference and set up a monitoring system with SageMaker Model Monitor to send notifications for quality deviations
-
C
Switch to SageMaker Real-Time Inference for the deployment and use SageMaker Model Monitor for model quality deviations
-
D
Increase the instance size and response timeout settings for SageMaker Asynchronous Inference and use Amazon SNS for quality deviation notifications
Reveal correct answer
Correct answer: C
Explanation
Correct option:
Switch to SageMaker Real-Time Inference for the deployment and use SageMaker Model Monitor for model quality deviations
SageMaker Real-Time Inference is designed for low-latency applications, making it ideal for real-time fraud detection. Real-time inference is ideal for inference workloads where you have real-time, interactive, low latency requirements. You can deploy your model to SageMaker AI hosting services and get an endpoint that can be used for inference. These endpoints are fully managed and support autoscaling.
Model quality monitoring jobs monitor the performance of a model by comparing the predictions that the model makes with the actual Ground Truth labels that the model attempts to predict. To do this, model quality monitoring merges data that is captured from real-time or batch inference with actual labels that you store in an Amazon S3 bucket, and then compares the predictions with the actual labels. If you set the value of the enable_cloudwatch_metrics to True when you create the monitoring schedule, model quality monitoring jobs can send all metrics to CloudWatch, which can further be used to send notifications.
Incorrect options:
Increase the instance size and response timeout settings for SageMaker Asynchronous Inference and use Amazon SNS for quality deviation notifications - While increasing the instance size and timeout may improve the performance, Asynchronous Inference is not suitable for real-time inference use cases.
Use SageMaker Batch Transform for inference and set up a monitoring system with SageMaker Model Monitor to send notifications for quality deviations - Batch Transform is intended for offline, large-scale batch processing and is not suitable for real-time fraud detection applications.
Retain SageMaker Asynchronous Inference and enable autoscaling for the endpoint to improve performance, while using Amazon SQS for notifications - While autoscaling can help handle higher loads, Asynchronous Inference is inherently not designed for real-time inference applications, making it a poor fit for this use case.
References:
https://docs.aws.amazon.com/sagemaker/latest/dg/deploy-model.html
https://docs.aws.amazon.com/sagemaker/latest/dg/realtime-endpoints.html
https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-model-quality-metrics.html
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
