Michael Hunger
Michael Hunger
I think, I fixed it, should work again: https://arrows.app
Yes makes sense to add an optional examples, section in the prompt, not sure how the other retrievers handle this in langchain4j?
Here is a quick & dirty variant. Have fun. 1. Python notebook with df.load_parquet and the neo4j driver 2. Cypher script using apoc.load.parquet (needs the hadoop dependencies) https://gist.github.com/jexp/74bd5a43305550236321eab8f0c723c0 We'll provide...
@tomasonjo did an improved version adding claims/covariates and findings. https://github.com/tomasonjo/blogs/blob/master/llm/ms_graphrag_import.ipynb We need to enable embedding generation in the index workflow, for text-units and community summaries and entities to also add...
@stevetru1 done: https://github.com/microsoft/graphrag/pull/544 Preview: https://github.com/microsoft/graphrag/blob/1f1d7b2c764db883a031b19923cb00a9db4844fe/examples_notebooks/neo4j/graphrag_import_neo4j_cypher.ipynb
Yep. I had even done a small integration with neo4j https://github.com/jexp/neo4j-3d-force-graph and neo4j.com/labs/neodash also supports 3d force graph.
see: https://github.com/microsoft/graphrag/issues/418#issuecomment-2224246436
https://github.com/microsoft/graphrag/pull/544
Which exact models did you use locally? and what did you change/configure?
I think as this is a mass processing job, it would make sense to use the underlying python code with LLMGraphTransformer in Langchain. https://python.langchain.com/v0.1/docs/use_cases/graph/constructing/#llm-graph-transformer