Professional Data EngineerOperationalizing machine learning modelsEasy
A telecommunications company is developing an ML model to predict network congestion. The model is trained on streaming network telemetry data. To ensure efficient resource utilization and rapid iteration, they need to manage and version multiple trained models, allowing data scientists to easily discover, share, and deploy specific model versions to different environments (e.g., staging, production). Which Vertex AI service is designed to be the central repository for these trained models?
- AVertex AI Experiments
- BVertex AI Model Registry
- CVertex AI Model Monitoring
- DVertex AI Pipelines
Show answer & explanationAnswer & explanation
Correct answer: B. Vertex AI Model Registry
Vertex AI Model Registry serves as a centralized repository for managing the lifecycle of ML models, including versioning, metadata, and deployment status. It allows easy discovery, sharing, and deployment of specific model versions.
Why the other options are wrong
- A. Experiments are for tracking training runs and their results, not for storing and managing trained model artifacts for deployment.
- C. Model Monitoring tracks model performance and data drift after deployment, not model storage and versioning.
- D. Pipelines orchestrate ML workflows, but the Model Registry specifically manages the trained model artifacts.
Vertex AI Model Registry
A centralized repository within Vertex AI for managing the lifecycle of machine learning models, including versioning, metadata, and deployment status.
- Enables model versioning and artifact management.
- Facilitates model discovery and sharing across teams.
- Supports deployment to endpoints and monitoring.
Memory trick: The Model Registry is like a 'library' for all your trained models.