Optimizing Vector Similarity Searches at Scale

Hi,

The provided example demonstrates using pre-computed vector data stored in a CSV file. While this works, I'm interested in a more permanent solution. Can we directly store vector data within a Fabric Warehouse table, in a dedicated column, instead of relying on a notebook to load it each time? My goal is to persistently store the vector data in a database, making it readily available for future use without requiring manual loading through notebooks.