Nodes

At a glance

The community member has a question about constructing a property graph index. They note that the documentation only shows how to create the index from a set of documents, but they want to know if there is an analogous way to create the nodes first per document, similar to how it can be done with a vector store index. In the comments, another community member suggests that the community member can use the kg extractors in an ingestion pipeline, but they should set the kg_extractors in the index to an empty list to avoid running the extractors again. There is no explicitly marked answer in the post or comments.

jjayjoshy

Hi, question about construction of a property graph index. Right now, the docs mention only this:

Plain Text

index = PropertyGraphIndex.from_documents(
    documents,
    kg_extractors=[extractor1, extractor2, ...],
)

With the normal vector store index I can generate the nodes first (ie. per document)

Plain Text

nodes_A = pipeline.run(documents=[DocumentOne])
nodes_B = pipeline.run(documents=[DocumentTwo])
index = VectorStoreIndex(nodes_A + nodes_B)

Is there an analogous way to do the same thing with the property graph index? where you can do it document by document before constructing the full index?

1 comment

LLogan M

You can use the kg extractors in an ingestion pipeline

But if you do this, I think you'll want to set kg_extractors in the index to an empty list, otherwise it will run them again

Add a reply

Find answers from the community

Nodes