MLflow and Model Lifecycle Management MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   When is promoting a model preferred over promoting code?

   • AWhen the model must be retrained on a schedule
   • BWhen the model is already trained and only needs to be deployed for inference
   • CWhen the training data changes frequently
   • DWhen the training code is complex and needs versioning
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   A user wants to see the model versions associated with a registered model in the MLflow UI. Which tab or section should they navigate to?

   • AThe Experiments tab
   • BThe Models tab
   • CThe Artifacts tab
   • DThe Runs tab
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   After using MlflowClient().search_runs() to retrieve runs from an experiment, which attribute of a returned Run object contains the metrics that can be used to identify the best run?

   • Arun.info.metrics
   • Brun.tags
   • Crun.data.metrics
   • Drun.params
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   In which scenario is promoting code preferred over promoting a model?

   • AWhen the model is a one-time artifact and will not be retrained
   • BWhen the training code must be reproducible and the model will be retrained frequently
   • CWhen the inference latency must be minimal
   • DWhen the model is small and can be stored in a registry
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   What is a key benefit of registering models in the Unity Catalog registry compared to the workspace registry?

   • ANo need to log models with MLflow
   • BAutomatic model retraining
   • CCentralized governance and lineage tracking across the organization
   • DFaster model inference at runtime
   Reveal Solution Discussion Select an answer first