Set up for Vertex AI Experiments

Vertex AI Experiments is supported by the Agent Platform SDK for Python and Google Cloud console. Vertex AI Experiments requires and depends on Vertex ML Metadata.

Set up

1. In the Google Cloud console, on the project selector page, select or create a Google Cloud project.

   Roles required to select or create a project

   • Select a project: Selecting a project doesn't require a specific IAM role—you can select any project that you've been granted a role on.
   • Create a project: To create a project, you need the Project Creator role (roles/resourcemanager.projectCreator), which contains the resourcemanager.projects.create permission. Learn how to grant roles.
   Note: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.

   Go to project selector

2. Verify that billing is enabled for your Google Cloud project.

3. Enable the required API.

   Roles required to enable APIs

   To enable APIs, you need the serviceusage.services.enable permission. If you created the project, then you likely already have this permission through the Owner role (roles/owner). Otherwise, you can get this permission through the Service Usage Admin role (roles/serviceusage.serviceUsageAdmin). Learn how to grant roles.

   Enable the API

4. Create a Service account. See Create a service account with required permissions.
5. Install the Agent Platform SDK for Python.
6. Check for existence of the default Metadata Store in your project. (required)
   • To see if your project has the default Metadata Store, go to the Metadata page in the Google Cloud console.
   • If the default Metadata Store doesn't exist, it's created when
     • you run the first PipelineJob,
     • or, create your first experiment in the Agent Platform SDK for Python.
       Optional: To configure with CMEK, see Configure your project's metadata store.

Supported Locations

The Feature availability table lists the available locations for Vertex AI Experiments. When using Agent Platform Pipelines or Vertex AI TensorBoard, they must be in the same location as your Agent Platform experiment.

What's next

• Create an experiment

Relevant notebook tutorials

1. Compare trained and evaluated models
2. Model training with prebuilt data pre-processing code
3. Compare pipeline runs
4. Autologging