You retrain an existing model. You need to register the new version of a model while keeping the current version of the model in the registry. What should you do?

The correct answer is B. Register the model with the same name as the existing model. Registering a model with the SAME name as an existing model is the built-in Azure ML versioning mechanism. The registry automatically assigns an incremented version number (v1, v2, v3, etc.) while retaining all prior versions. This is the correct approach to keep both old and…

Train and deploy models

Question

You retrain an existing model. You need to register the new version of a model while keeping the current version of the model in the registry. What should you do?

Options

ARegister a model with a different name from the existing model and a custom property named
BRegister the model with the same name as the existing model.
CSave the new model in the default datastore with the same name as the existing model. Do not
DDelete the existing model and register the new one with the same name.

How the community answered

(49 responses)

A
4% (2)
B
86% (42)
C
8% (4)
D
2% (1)

Explanation

Registering a model with the SAME name as an existing model is the built-in Azure ML versioning mechanism. The registry automatically assigns an incremented version number (v1, v2, v3, etc.) while retaining all prior versions. This is the correct approach to keep both old and new versions accessible. Option A (different name) creates a completely separate model entry, not a new version. Option C (saving to datastore without registering) makes the model inaccessible via the registry. Option D (deleting the existing model first) destroys the version history, violating the requirement to keep the current version.

Topics

#Model Registry#Model Versioning#Azure Machine Learning

Community Discussion

No community discussion yet for this question.

Full DP-100 Practice