Learn how to request access to the solution, import the notebooks into your Databricks workspace, and prepare your environment for running Reltio Entity Resolution.

1. Request access to the solution in Databricks Marketplace
   In the Reltio Entity Resolution in Databricks listing, click Get Access. Authenticate with your Databricks workspace and complete the request form.

   We'll review your request and contact you to proceed.

2. Import the notebook into your Databricks workspace
   Once your request is approved, we'll provide you access to the repository containing ready-to-run notebooks. Use the Databricks Workspace UI to clone the repository and import the notebooks into your own workspace.
3. Attach or create a cluster with Unity Catalog access
   Ensure that your notebook is attached to a compute cluster that:

   • Runs Databricks Runtime 16.4 or later
   • Is configured in Shared or Single-user mode (not legacy High Concurrency)
   • Has access to Unity Catalog resources

   We recommend using a GPU cluster (such as A10G) for better performance with embedding-based matching models.
4. Verify Unity Catalog permissions
   Confirm that your user or service principal has the following privileges:

   Object Type	Required Privilege
   Catalog	USE CATALOG
   Schema	USE SCHEMA
   Delta Table	SELECT

   If your cluster doesn't have access, contact your workspace administrator.
5. Run the Unity Catalog validation cell
   In the notebook, locate and run the cell labeled Step 1: Cluster Configuration. This command checks whether your cluster is correctly configured. If successful, you'll see a confirmation message stating "Cluster Check Successful".
6. Create the catalog containing machine learning models shared by us via Delta Sharing
   In your Databricks workspace, navigate to Catalog > Delta Sharing > Shared with me. Look for reltio as a provider. Clicking on it will showcase the data asset reltio_entity_resolution_models shared with you. Create a new Catalog using this asset. You will need this catalog while running step 3: Machine Learning Models Catalog Selection of the notebook.