Managing Projects and Access in the H2O Feature Store
- Predictive AI
- H2O Feature Store
This lesson covers how projects and access work in the H2O Feature Store. It shows browsing public projects to see what other teams are working on, such as a telecommunications project aimed at reducing churn, alongside a user's own projects. It covers marking a project secret or locked to keep it visible only to a specific group, and using the access list to grant different consumers across the organization access to a feature set. It closes by opening a feature set to see details like median value and unique value counts, and by showing how versioning lets you pull the latest data or see what it looked like at a past point in time.
Key takeaways
- How marking a feature store project secret or locked limits its visibility to a specific group instead of the whole organization
- What the access list lets you do: grant different consumers across the organization access to a specific feature set
- What feature level details are visible in a feature set, such as a feature's median value and number of unique values
- Why feature set versioning lets you pull the latest data or see what it looked like at a specific point in time
Start owning your AI