Answer-first summary for fast verification
Answer: Enable request-response logging for the Vertex AI endpoint, and set up alerts using Cloud Logging. Review the feature attributions in the Google Cloud console when an alert is received.
The question requires monitoring feature attribution drift with continuous evaluation metrics using Vertex AI Model Monitoring. Option D is correct because it enables request-response logging for the Vertex AI endpoint to capture prediction data, sets up alerts using Cloud Logging to notify when drift is detected, and leverages the built-in Google Cloud console to review feature attributions, providing a fully managed solution. Option A is incorrect because it relies solely on Cloud Logging alerts without specifying how feature attribution data is captured or analyzed systematically. Option B introduces unnecessary complexity with Looker Studio dashboards when Vertex AI's console already provides visualization capabilities. Option C is invalid because request-response logging cannot be enabled when Model Monitoring is already configured, as confirmed by the community discussion reference to Google's documentation. The community discussion shows split opinions between A and D, but D aligns with Vertex AI's integrated monitoring features for continuous evaluation.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have deployed an ML model to a Vertex AI endpoint and set up a Vertex AI Model Monitoring job. To continuously evaluate the model by monitoring for feature attribution drift, what steps should you take?
A
Set up alerts using Cloud Logging, and use the Vertex AI console to review feature attributions.
B
Set up alerts using Cloud Logging, and use Looker Studio to create a dashboard that visualizes feature attribution drift. Review the dashboard periodically.
C
Enable request-response logging for the Vertex AI endpoint, and set up alerts using Pub/Sub. Create a Cloud Run function to run TensorFlow Data Validation on your dataset.
D
Enable request-response logging for the Vertex AI endpoint, and set up alerts using Cloud Logging. Review the feature attributions in the Google Cloud console when an alert is received.
No comments yet.