From the course: MLOps with Databricks
Join today to access over 25,600 courses taught by industry experts.
Register models in Unity Catalog - Databricks Tutorial
From the course: MLOps with Databricks
Register models in Unity Catalog
- [Instructor] Databricks recommends registering models in Unity Catalog and the Workspace Model Registry is marked as legacy. In this course, I only focus on registering models in Unity Catalog, which comes with some advantages. The biggest advantage is the ability to share models across different workspaces such as workspaces in different environments, or even workspaces that belong to different teams and products. Unity Catalog also simplifies lineage and governance of machine learning models. There are couple of things to pay attention to when registering models in Unity Catalog. As mentioned earlier, you cannot register a model if model signature is not provided. Search functionality is limited compared to the workspace model registry. For example, you cannot search registered models by tag, which makes it harder to find registered models that belong to a certain realm. Before we deep dive into the code, please make sure that the MLflow registry URI is set to UnityCatalog by…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.