ModelOps(Model Operations) — Learn free

Self-check

Questions 1–10 of 10

1. How is ModelOps positioned relative to MLOps?
Question 1A. As an unrelated discipline that replaces MLOps entirelyB. As the next stage of MLOps, focused on automating, retraining and maturing models on their ownC. As a subset of MLOps concerned only with model monitoringD. As a manual review process that precedes MLOpsE. As a synonym for DevOps applied to databases
2. Which set of concerns is the MAIN focus of MLOps, as contrasted with ModelOps?
Question 2A. AI transparency and business KPIsB. Decision models and governanceC. Sharing AI performance with stakeholdersD. Model management, model monitoring and CI/CDE. Life cycle management of decision models
3. ModelOps aims to share AI performance with stakeholders without needing the involvement of which team members?
Question 3A. Data Science and Machine Learning team membersB. Business usersC. Executive stakeholdersD. IT infrastructure administratorsE. External auditors
4. MLOps focuses on operationalising ML models. What does ModelOps focus on operationalising instead?
Question 4A. Data pipelines and ETL jobsB. Network and server infrastructureC. Decision models and AID. Business intelligence dashboardsE. Relational database schemas
5. Whereas MLOps focuses on a continuous cycle of development, implementation and monitoring of models, ModelOps focuses on which of the following?
Question 5A. Feature engineering and data cleansingB. Hyperparameter tuning of individual modelsC. Writing model code for AI-enabled applicationsD. Provisioning cloud compute for trainingE. Life cycle management and governance of models
6. In ModelOps, AI transparency and the usage of KPIs are aimed at which audience?
Question 6A. Data engineersB. Business usersC. Machine learning researchersD. Database administratorsE. Software testers
7. Which of the following is NOT listed as something models do on their own under ModelOps?
Question 7A. Run on their ownB. Retrain on their ownC. Deploy on their ownD. Write their own source code on their ownE. Extract business KPIs on their own
8. As ModelOps matures, what happens to the involvement of Data Science in running models?
Question 8A. It increases because more models need supervisionB. It stays exactly the sameC. It becomes less and less as models run on their ownD. It shifts entirely to external consultantsE. It is replaced by database administration
9. Which of the following is the focus of MLOps rather than ModelOps?
Question 9A. Governance of decision modelsB. AI transparency for business usersC. Creating models for AI-enabled applicationsD. Sharing AI performance with stakeholdersE. Models retraining themselves
10. Which statement best summarises the relationship between ModelOps and MLOps?
Question 10A. They are entirely separate fields with no overlapB. MLOps is a later evolution of ModelOpsC. ModelOps only covers CI/CD pipelines for modelsD. ModelOps applies only to deep learning modelsE. There is a very thin line between them, with ModelOps building on MLOps towards self-managing models