MLOps CI/CD Quiz

What is the primary reason for using CI/CD in MLOps?

• To manually deploy models

• To automate integration, testing, and deployment processes

• To prevent model updates

• To manually track data changes

In MLOps, which tool is typically used for version control of code and models?

• Docker

• Kubernetes

• Git

• Jenkins

What is the role of artifact storage in the CI/CD pipeline?

• To delete model files after deployment

• To store Docker images or model files for later use

• To build machine learning models

• To handle manual deployment tasks

Which of the following best describes Continuous Delivery (CD)?

• Automated testing and manual deployment to production

• Manual integration and testing of new models

• Automated model integration without deployment

• Fully manual integration and deployment

What is the main benefit of continuous learning in MLOps?

• Models remain static and unchanged

• Models are updated regularly with new data

• Models require manual updates

• Models are only tested once

In which environment are models deployed for testing before going into production?

• Production environment

• Development environment

• Staging environment

• Local environment

What is the key difference between Continuous Delivery and Continuous Deployment?

• Continuous Delivery requires manual approval for production deployment

• Continuous Delivery is fully automated

• Continuous Deployment requires manual approval for testing

• Continuous Deployment does not include any testing

What is a primary benefit of using microservices in machine learning?

• Reduced performance

• Simplifies debugging by breaking down components

• Creates a monolithic architecture

• Prevents modular development

Which of the following can trigger a CI/CD pipeline execution in MLOps?

• Git commits

• Model failure

• Manual intervention only

• Downtime

What is the purpose of REST APIs in MLOps?

• To remove models from production

• To facilitate communication between services and applications

• To test models manually

• To perform batch processing of models