If I do training on Sagemaker (for example), can I still use the MLflow Tracking Server on Databricks instead of hosting my own server?

MoJaMa
Databricks Employee
Databricks Employee
 

MoJaMa
Databricks Employee
Databricks Employee

Yes!

You will have to

pip install mlflow

in your environment as a first step.

For more details, see: https://docs.databricks.com/applications/mlflow/access-hosted-tracking-server.html