Databricks Certified Machine Learning - Associate
Get started today
Ultimate access to all questions.
Ultimate access to all questions.
A machine learning engineer is enhancing a project to automatically refresh the model each time the project runs. The project is connected to an existing model named model_name in the MLflow Model Registry. The following code snippet is part of their strategy:
mlflow.sklearn.log_model(
sk_model=model,
artifact_path="model",
registered_model_name=model_name
)
Given that model_name is already present in the MLflow Model Registry, what does the parameter registered_model_name=model_name signify?
Explanation:
When mlflow.sklearn.log_model is called with registered_model_name specified, MLflow not only logs the model but also registers it with the Model Registry. If the model name already exists in the registry, a new version of the model is created under that name. Therefore, the parameter registered_model_name=model_name in this context means that a new version of the model named model_name will be registered in the MLflow Model Registry.