Hi @raghagra,

Thank you for posting your question in the Databricks community.

The reason why the code is only logging the last run is because you are using the mlflow.start_run() function inside the objective_function() function. This means that each time you call the objective_function() function, it will start a new run. The mlflow.spark.log_model() function only logs the model for the current run, so the model will only be logged for the last run.

To fix this, you can move the mlflow.start_run() function outside of the objective_function() function. This will ensure that the model is logged for every run.

Please check how it works.