The Experiments page gives you quick access to MLflow experiments across your organization. You can track machine learning model development by logging to these experiments from Databricks notebooks and jobs.
To access this page, click Experiments in the sidebar. This icon appears only when you are in the machine learning persona.
Only experiments that you have access to in your current workspace appear on this page. From this page, you can:
View, search for, and create experiments.
Compare runs from multiple experiments.
Rename or delete an experiment.
Change experiment permissions to control who has access to the experiment.
To display an experiment page, click the name of an experiment in the table.
To search for experiments, type text in the Search field and click Search. The experiment list changes to show only those experiments that contain the search text in the Name, Location, Created by, or Notes column.
AutoML experiment. The Configure AutoML experiment page displays. See Train ML models with the Databricks AutoML UI for information about using AutoML.
Blank experiment. The Create MLflow Experiment dialog appears. Enter a name and optional artifact location in the dialog. The default artifact location is
dbfs:/databricks/mlflow-tracking/<experiment-id>. See Create workspace experiment for more details.
To log runs to this experiment, call
mlflow.set_experiment()with the experiment path. The experiment path appears at the top of the experiment page. See Example notebook for details and an example notebook.
See Compare runs from multiple experiments for instructions.
You can rename or delete an experiment, and change experiment permissions to control who has access to the experiment.
To rename or delete an experiment, click in the Actions column and select Rename or Delete.
To change the permissions for an experiment, click in the Actions column and select Permission. The Permission Settings dialog appears.