The community member is trying to set up a system that can store data in a vector store and perform both Q&A and summarization on the stored data. They have come across a resource (https://docs.llamaindex.ai/en/stable/examples/query_engine/JointQASummary.html) but are unsure how to implement it with a vector store.
In the comments, another community member suggests that using only a vector store may not be suitable for summarization, as most vector stores do not have a "get all" option, and setting a high top k can destroy the order of nodes, which is helpful for summarization.
I'm trying to set up a system that can store data in a vector store and can subsequently perform both Q&A and summarization on the stored data. Could someone point me to any documentation, please?
For summarization, you can't really use only a vector store a) most vector stores don't have a "get all" option. Instead, you'd have to set a really high top k... b) .. which also destroys the order of nodes, which is also helpful for summarization
