Skip to content

README.md

Latest commit

 

History

History
33 lines (22 loc) · 2.37 KB

README.md

File metadata and controls

33 lines (22 loc) · 2.37 KB

Lab 4 - neo4j-graphrag Library

Run four notebooks that build a GraphRAG search pipeline using the neo4j-graphrag Python library. You start by loading pre-computed chunk embeddings into the knowledge graph from Labs 1-2, build retrieval pipelines that progress from pure vector search to graph-enriched search, then wrap both retrievers as agent tools so the model picks the right strategy per question.

What You'll Learn

  • Data Loading: Add document chunks with pre-computed embeddings to an existing knowledge graph
  • Vector Search: Use VectorRetriever to find semantically similar chunks and generate answers with GraphRAG
  • Graph-Enriched Search: Use VectorCypherRetriever to combine vector search with Cypher graph traversal for richer context
  • Agent-Driven Retrieval: Wrap both retrievers as Strands @tool functions and let the agent choose the right strategy per question

Prerequisites

Before starting this lab, make sure you have:

  • Completed Lab 1 (Aura instance created and knowledge graph loaded)
  • CONFIG.txt at the project root filled in with NEO4J_URI, NEO4J_USERNAME, NEO4J_PASSWORD, MODEL_ID, and REGION
  • A running environment (SageMaker Studio or GitHub Codespace)

Note: This lab uses the same Neo4j Aura instance from Labs 1-2. The structured graph (companies, products, risk factors, asset managers) must already be loaded.

Notebooks

Notebook Title What You Build
01_load_and_query.ipynb Load Data and Query Load chunk embeddings into the graph, create a vector index, link entities to chunks, and run test queries
02_vector_retriever.ipynb Vector Retriever Semantic search with VectorRetriever and end-to-end question answering with GraphRAG
03_vector_cypher_retriever.ipynb VectorCypher Retriever Graph-enriched retrieval that adds companies, products, and risk factors to vector search results
04_strands_graphrag_agent.ipynb Strands GraphRAG Agent A Strands agent that decides which retriever to use based on the question

Next Steps

After completing this lab, continue to Lab 5 to connect agents to Neo4j via MCP, or to Lab 6 to build the data pipeline from scratch.