Hi all, I have a question regarding the custom VectorStore Postgres table.
- How do I add other columns? From PGVectorStore, it creates id, text, metadata, node_id, and embedding only. How can I add another column like user_id?
- When I generate indexes, I want to generate vector index, keyword index, and list index. How can I store these three indexes in one table? The table column names like id, text, metadata, node_id, embedding, vector_index, keyword_index, list_index
- how to store keyword index in Postgres? From the tutorial keyword_index is generated by GPTKeywordTableIndex. and I modify the code as below:
vector_store = PGVectorStore.from_params(... )
storage_context = StorageContext.from_defaults(vector_store=vector_store)
storage_context.docstore.add_documents(nodes)
keyword_index=GPTKeywordTableIndex(nodes, storage_context=storage_context)
I am not sure if this is working or not because I cannot find any data from the table that was assigned or other tables in the database. no error messages.
