Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Comments
Loading comments...
Ultimate access to all questions.
As a Machine Learning Engineer at a gaming company developing MMO games, you've developed a TensorFlow model to predict if players will spend over $10 in in-app purchases within the next two weeks. The model aims to personalize the gaming experience, with user data stored in BigQuery. The company prioritizes cost efficiency, minimal latency for user experience, and ease of model management. Given these constraints, which of the following solutions is the BEST to serve this model? Choose one correct option.
A
Deploy the model to Vertex AI Prediction for real-time predictions, using Cloud Bigtable for low-latency data access, and store the prediction results in Cloud SQL for easy access and management.
B
Embed the model directly in the mobile app to generate predictions on-device after each in-app purchase event, using Pub/Sub for event notification, and store the data in Cloud SQL.
C
Integrate the model with BigQuery ML to perform batch predictions directly on the data stored in BigQuery, and save the outcomes to Cloud SQL for reporting and further analysis.
D
Use a streaming Dataflow pipeline to process each in-app purchase event in real-time, trigger predictions using the model deployed on Vertex AI, and store the results in Cloud SQL.