@Logan M or anyone avaliable. I am running into an issue of VERY dense neighborhoods in my Graph for instance if I have a common person that is mentioned often in the documents I am injesting they could have 100s-1000s of relationships. Are there any best practices worked out for getting the most relevant relationships to a query?

Can you create the citation Query engine in LlamaIndex.ts?

I was just asking about this. I could definitely use this feature.

How can I use Claude 3 Opus to create the property graph?

Does Query_engine would vercel AI vector stream?

IS there currently a way to combine the Citation Query Engine with the PGRetriever?

What is the best way to filter the results from index.as_retriever().retrieve()?

Why do some nodes in the property graph get labelled "Unknown"

If you wanted to do some sort of entity resoltution. What would be the best way to delete and add entities back into the graph? Would it be defining a custom retriever to retrieve nodes you want to combine then coming up with a way to delete and upsert those nodes?

how do I increase the chunk size in simply directory reader?

How do I see how many documents I have added to a Property Graph Index?

When will Llama Parse be back up? I noticed that status page is not correct?

Does the Llama Index Property graph have these properties:

Are there any production GraphRAG Apps out?

Has anyone thought about modifying the property graph code to do semantic chunking and triples extraction at the same time?

My thought process:

• Parse the document into chunks
• Recombine all the chunks into one text string but insert metadata denoting each chunk in to the combined chunks text string

so basically original text: "paragraph 1, paragraph 2, paragrah 3" => [{Chunk1 text: "paragraph 1"}, {Chunk2 text: "paragraph 2"}, {Chunk3 text: "paragraph 3"}] =>

"""
Chunk1
=========
paragraph 1
=========

Chunk2
=========
paragraph 2
=========

Chunk3
=========
paragraph 3
=========
"""

• when you are doing extraction you pass the entire document and instead pullout triples by them selves you prompt the model to pull out a triple plus the chunk ids that it is from

Entity("Paragraph") -- Relationship("HAS_NUMBER") --> Entity("1") extracted from chunk1


I would say allow for / ask for 1-MAX_CHUNKS_PER_TRIPLE to be references via chunk ids when preforming extraction. Now you get the exact location of every single triple plus you get the triples created with the context of the whole document.

@Logan M Llama Parse seems to be down but the status page is showing green?

@Logan M Hi Logan I think I found a bug I wanted to clarify if it is expected behavor with the property graph. When creating the property graph I noticed that every single entity only has one chunk mentioning them. I was curious if a chunk is supposed to only mention a entity if it is the first time the entity is created (which is what it looks like) or if it should mention a entity whenever an entity is referenced in a chunk? If this behavior is intentional why? what is the best path to fixing this?

Does anyone know about good tutorials or uses of Sub-agents? I would like to be give the model the ability to spin up a number of sub-agents for a given task, these agents I would like to be defined like tools where they are just given an input and prompt is pre-defined.

@Logan M We are having a problem where I am can't seem to the figure out why our vector retriever is getting really wonky results. (Seems like the embeddings model aren't matched up correctly) Have you seen this bug/problem before and any reccomendations on how to fix it?

@Logan M One of the members from my team added functionality for filtering on queries from for a Neo4J Database. He has an open PR what are the steps for getting that merged?

https://github.com/run-llama/llama_index/pull/14362

@kapa.ai When using a custom Retriever when are filters run? is the custom retriever only using objects that have already been filtered by the filter?

@kapa.ai How do I add filters to retrievers?

@Logan M I am trying to create a graph with Llama 3 70b from Bedrock but I see that a nessasary function acomplete isn't implemented? what is the best way to go getting the implemented, tested and PRed?

@kapa.ai what query engine does the property graph index use?