Hello, I have a problem with neo4j PropertyGraphIndex, where it creates a contraint uniqueness for Entity and Node type nodes for the "id" property. But, it sets the id to the same value as name, which is common to other nodes and therefore clashes. Is there a way around this? Why is the id be set to a common string or at all? Or I'm missing something?
I also found that v0.3.5 has a cypher error. I had to revert to v0.3.3. This is on package llama-index-graph-stores-neo4j
Also, I didn't find release notes for this package on pip. Are they somewhere else? With you guys or neo4j people?
Hello. I'm using the DocumentSummaryIndex class to summarise pdf documents. I load the pdfs with the SimpleDirectoryReader. The reader parses each page as a doc/node (not sure the right term). As a result, I get a summary for each page. This is ok, but what I'm really looking for is a summary if the overall pdf. I've tried extracting the summary nodes and use the build_index_from_nodes method on the index, but it does the same. Any suggestion? Including a different approach 🙂
I'm using the RedisIndexStore to manage document summary indices. Is there a way I can add and remove documents to an existing index, instead of having to create a new full index from scratch every time?
Can DocumentSummaryIndex be stored in Neo4j? I already store PropertyGraphIndex in Neo4j, so it would be convenient to have all indices under the same hood.
