Answer-first summary for fast verification
Answer: Use the features and the feature attributions for monitoring. Set a prediction-sampling-rate value that is closer to 0 than 1.
The correct answer is D. Given that you expect a large volume of prediction requests and aim to minimize costs, it is optimal to both use the features and the feature attributions for monitoring, and set a prediction-sampling-rate value that is closer to 0 than 1. Monitoring both features and feature attributions helps identify if the contributions of features to the model's predictions are changing, which can indicate drift. A lower prediction sampling rate reduces the number of predictions analyzed, thereby lowering monitoring costs, making it a cost-effective approach while still maintaining the ability to detect drift.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have developed a custom machine learning model using Vertex AI to forecast the sales of your company’s products based on historical transactional data. Given the dynamic nature of your business, you anticipate changes in the feature distributions and the correlations between the features in the near future. Additionally, you expect to receive a large volume of prediction requests. To maintain the accuracy and reliability of your model's predictions, you plan to utilize Vertex AI Model Monitoring for drift detection. Considering your objective to minimize costs, what configuration should you choose for monitoring?
A
Use the features for monitoring. Set a monitoring-frequency value that is higher than the default.
B
Use the features for monitoring. Set a prediction-sampling-rate value that is closer to 1 than 0.
C
Use the features and the feature attributions for monitoring. Set a monitoring-frequency value that is lower than the default.
D
Use the features and the feature attributions for monitoring. Set a prediction-sampling-rate value that is closer to 0 than 1.
No comments yet.