Google Professional Machine Learning Engineer Question 192
Single answerGoogle Cloud PlatformYour team is developing a machine learning model for a recommendation system. The model requires extensive data preprocessing, custom model training, and hyperparameter tuning. The team wants to minimize infrastructure management while leveraging Google Cloud's managed services. Which Google Cloud environment should you choose for this development?
- A
AI Platform (Vertex AI) Workbench
- B
Google Kubernetes Engine (GKE)
- C
Cloud Functions
- D
Compute Engine
Show answer and explanation
Correct answer: A
Explanation
For machine learning development involving data preprocessing, custom model training, and hyperparameter tuning, AI Platform (Vertex AI) Workbench is the most appropriate choice. It provides a managed and integrated environment tailored for ML workflows, reducing the need for manual infrastructure management while supporting the entire ML lifecycle.
- A. Correct.
AI Platform (Vertex AI) Workbench is a fully managed environment designed for data scientists and ML engineers, offering Jupyter-based notebooks, seamless integration with Vertex AI for training and hyperparameter tuning, and minimal infrastructure management.
- B. Incorrect.
Google Kubernetes Engine (GKE) is suitable for deploying containerized applications and managing Kubernetes clusters but requires significant infrastructure management. It is not optimized for ML-specific workflows like training or hyperparameter tuning.
- C. Incorrect.
Cloud Functions is a serverless platform for running event-driven code. It is not suitable for data preprocessing, custom model training, or hyperparameter tuning due to its stateless nature and limited runtime capabilities.
- D. Incorrect.
Compute Engine provides virtual machines for custom workloads, offering flexibility but requiring significant infrastructure management. It is not ideal for a managed ML development workflow.