I have a csv file with two colums like a FAQs, data is different information for different banks , with embeding form also of one coloum , Right now , i am performing full RAG process
I want to use VectorDB , but i dont understand the concept of Index in Vectordb , what this will do , how to setup index , on what basis , i have a two colum in a csv text data
Indices just help you find stuff faster. They’re not really necessary if you have a small corpus (<1000 items), but become more relevant when you move more data (>10000 items).
Sometimes people just use the word “Index” as a general term for what it is that vector embedding (i.e. RAG approach) is doing. I wouldn’t get too hung up on the specific word “index”, if you just want to understand RAG and vector embeddings. The only concept you need to know is what a “vector embedding” is, and you can find lots of good videos on Youtube that explain it.