Nodes

At a glance

The community member is trying to view the nodes or chunks after creating a vector query index. They are facing an issue where they get proper results from similarity search when passing a chunk_size of 512 in the service context, but not when they don't pass anything (assuming the default chunk_size of 1024). They would like to see what the chunks look like to understand what's wrong.

In the comments, another community member suggests checking the nodes by accessing the docstore like nodes = index.docstore.docs. However, the original poster mentions that they were told to call get_nodes, which doesn't exist. Another community member agrees that there is no such method in the VectorStoreIndex class yet, and suggests that the Mendable docs need better prompt instructions.

There is no explicitly marked answer in the provided information.

cchantlong

How can I view the nodes or I guess the result of the chunks after creating a vector query index?
I'm running into an issue where in the service context if I pass in a chunk_size of 512 I am able to get proper results from similarity search. But when I don't pass anything into the service context, which I assume the chunk_size is 1024 by default from llama index, I am getting no results back so I'd like to see what the chunks end up looking like to see whats wrong.

Plain Text

vector_query_engine_index = VectorStoreIndex.from_documents(documents, use_async=True,
service_context=service_context
)

3 comments

WWhiteFang_Jr

You can check all the nodes in the index by accessing the docstore like this
nodes = index.docstore.docs

cchantlong

thanks! Was asking the mendable docs and it told me to call get_nodes which didn't exist.

WWhiteFang_Jr

Yeah, there is no method I think in the vectorStoreIndex class yet.

Mendable needs better prompt instructio I think 😆. Lol

Add a reply

Find answers from the community

Nodes