Find answers from the community

Updated 3 months ago

I used the ingestion pipeline to store

I used the ingestion pipeline to store embeddings (with Postgres) and when I try to load in the index using VectorStoreIndex.from_vector_store, the nodes are empty. Does anyone know how to resolve this?

Im doing index=VectorStoreIndex.from_vector_store(vector_store=vector_store)

7 comments

LLogan M

what do you mean the nodes are empty?

mmewtoo

@Logan M

nodes=index.docstore.docs.values()
print(“nodes:”,nodes)

mmewtoo

https://discord.com/channels/1059199217496772688/1223362664458092685/1242279961084297348

mmewtoo

seems like I might have to set the nodes within the index from my database?

LLogan M

Yea, Docstore isn't used by default with vector db integrations

mmewtoo

@Logan M thank you. so I’d go through all my docs and add them to db via ingestion pipeline to Postgres.

And then have to load all the docs into the docstore of the VectorStoreIndex from the db?

I’m just trying to use the chat engine for a chatbot. Is this the correct way?

LLogan M

Unless you specifically need the docstore, I wouldn't worry about it 👀

Add a reply