In previous sections we already learnt what a Vector database is and why this is so important. In the RAG pattern Vector database plays the crucial component to store the Embeddings data as well as providing semantic search capability.
PostGres (pgvector) in AWS:
PostGreSQL database has the community built pgVector extension which if enabled can store embeddings from machine learning (ML) models into the database as well as supports semantic search to return similar results based on the distance between two vectors by applying Cosine algorithm.
With pgVector extension it is easy to use the known common PostGres database in the RAG pattern implementation to build ML capabilities into your QnA solutions, enterprise search, recommendation engine into e-commerce, media, finance or health applications.
...