AML Model Deployment & MLOps Flashcards
6 cards from real AML practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 6 AML Model Deployment & MLOps flashcards as text
What is MLOps in the context of advanced machine learning?
Answer: The practice of combining ML development with operations to deploy and maintain models in production
MLOps applies DevOps principles to machine learning, automating training, deployment, monitoring, and retraining workflows.
What is model drift?
Answer: The degradation of model performance as real-world data distribution changes from training data
Model drift occurs when the statistical properties of input data or target relationships change post-deployment, reducing model accuracy.
Which tool is commonly used to containerize ML models for consistent deployment across environments?
Answer: Docker
Docker packages ML models with their dependencies into portable containers that run consistently across development and production environments.
What is a REST API in the context of machine learning model deployment?
Answer: A standardized interface that allows applications to send data to a model and receive predictions via HTTP
A REST API exposes model inference as HTTP endpoints so any downstream application can request predictions via standard web requests.
What is A/B testing in the context of ML model deployment?
Answer: Comparing two model versions by routing live traffic to each and measuring performance differences
A/B testing routes a portion of real production traffic to a challenger model while the rest serves the incumbent, enabling data-driven model selection.
What does CI/CD stand for in an MLOps pipeline?
Answer: Continuous Integration / Continuous Deployment
CI/CD automates testing and deployment pipelines so new model versions can be validated and released quickly and reliably.