Skip to main content
Version: Current

Databricks Unity Catalog ML models registry integration

Enterprise

Description​

Nussknacker ML Runtime can serve models registered in the Databricks Unity Catalog ML models registry. Once the integration is configured, Nussknacker will generate an ML Enricher component, which can be used in the scenario to infer ML models.

This document explains how to configure Databricks Unity Catalog integration.

Integration parameters​

  • Databricks instance URL:
    Databricks instance URL.
    Example: https://dbc-a1b2345c-d6e7.cloud.databricks.com

  • Client ID:
    The Service Principal client ID.

  • Client Secret:
    The Service Principal secret.

Refer to the Databricks documentation for instructions on on configuring Service Principals and authorizing access to your instance. Ensure the Service Principal is assigned the correct roles (Can Read), on experiments and models.