A machine learning engineer is converting a Hyperopt-based hyperparameter tuning process from manual MLflow logging to MLflow Autologging. They are trying to determine how to manage nested Hyperopt runs with MLflow Autologging.
Which of the following approaches will create a single parent run for the process and a child run for each unique combination of hyperparameter values when using Hyperopt and MLflow Autologging?
To create a single parent run for the process and a child run for each unique combination of hyperparameter values when using Hyperopt and MLflow Autologging, the correct approach is to start a manual parent run before calling fmin. This allows MLflow to properly nest the runs by ensuring that the entire hyperparameter tuning process is encapsulated within the parent run, and each evaluation of a hyperparameter combination is tracked as a child run.
A the notebook linked here https://learn.microsoft.com/en-us/azure/databricks/machine-learning/automl-hyperparam-tuning/hyperopt-spark-mlflow-integration#parallelize-hyperparameter-tuning-with-automated-mlflow-tracking-notebook shoes that a parent run is started before calling fmin
A Correct. The approach that will create a single parent run for the process and a child run for each unique combination of hyperparameter values when using Hyperopt and MLflow Autologging is starting a manual parent run before calling fmin1.