Answer-first summary for fast verification
Answer: Decrease the sample_rate parameter in the RandomSampleConfig of the monitoring job
The correct answer is C. Decreasing the sample_rate parameter in the RandomSampleConfig of the monitoring job will reduce the model monitoring cost by instructing Vertex AI Model Monitoring to analyze a smaller percentage of incoming requests. This approach directly reduces the billing cost associated with monitoring while still providing enough data to capture significant data drift. Though other options like replacing the monitoring job with a DataFlow pipeline (A) or a custom SQL script (B) could be considered, they involve additional development, management overhead, and may not be as efficient or maintainable. Increasing the monitor_interval parameter (D) would reduce the frequency of monitoring checks, potentially delaying drift detection, which is not desirable if data drifts frequently.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have recently deployed a machine learning model to a Vertex AI endpoint, and due to the frequent data drifts, you have enabled both request-response logging and created a Vertex AI Model Monitoring job to ensure the model's performance. However, you've noticed that the model is receiving higher-than-expected traffic, leading to increased costs. Your goal is to reduce these monitoring costs while still being able to quickly detect any data drift. How should you adjust your current setup to achieve this?
A
Replace the monitoring job with a DataFlow pipeline that uses TensorFlow Data Validation (TFDV)
B
Replace the monitoring job with a custom SQL script to calculate statistics on the features and predictions in BigQuery
C
Decrease the sample_rate parameter in the RandomSampleConfig of the monitoring job
D
Increase the monitor_interval parameter in the ScheduleConfig of the monitoring job
No comments yet.