Google Professional Machine Learning Engineer Question 419
Select 3Google Cloud PlatformYou are working as a Machine Learning Engineer at a company that deploys models using a CI/CD pipeline in Google Cloud. You need to ensure that your model updates are automatically validated, tested, and deployed to production with minimal manual intervention. Which combination of tools and practices should you use to achieve this?
- A
Use Cloud Build to automate the build and test steps for your ML pipeline.
- B
Use AI Platform Prediction to deploy the model and manage versioning.
- C
Manually trigger deployment steps to ensure accuracy in production.
- D
Implement Cloud Functions to automatically trigger retraining jobs based on new data availability.
- E
Use Monitoring tools like Cloud Monitoring and AI Platform's built-in monitoring to track model performance post-deployment.
Show answer and explanation
Correct answers: A, B, E
Explanation
To implement a successful CI/CD pipeline for model deployment in Google Cloud, you need to automate the build, test, and deployment steps using tools like Cloud Build and AI Platform Prediction. Additionally, monitoring tools are essential for tracking performance post-deployment. Manual intervention and unrelated automation (e.g., Cloud Functions for retraining) are not aligned with the goals of CI/CD pipelines.
- A. Correct.
Cloud Build is a fully managed CI/CD platform in Google Cloud that can automate the build and test processes, ensuring models are validated before deployment.
- B. Correct.
AI Platform Prediction is specifically designed for deploying ML models and managing multiple versions, making it a key component for the deployment process.
- C. Incorrect.
Manually triggering deployments contradicts the goal of a CI/CD pipeline, which aims to automate processes with minimal manual intervention.
- D. Incorrect.
While Cloud Functions can automate some tasks, such as triggering retraining, it is not directly related to CI/CD deployment pipelines.
- E. Correct.
Monitoring tools like Cloud Monitoring and AI Platform's built-in monitoring are crucial for tracking the performance of your model in production and ensuring it continues to meet business goals.