Mastering Retrieval-Augmented Generation with the GraphRAG Python Package
summary
Shows how to build Retrieval-Augmented Generation (RAG) apps in Python using the neo4j-genai-python (GraphRAG) package to connect Neo4j graph databases with generative AI models. Covers integrating Neo4j with LLMs to power graph-aware AI solutions, with links to the workshop repo, docs, and a GraphAcademy course.
AI-written summary · about
chapters (19)
- 0:00Introduction and setup
- 3:20Environment configuration
- 8:57Environment test
- 12:04RAG concepts
- 16:00Semantic search challenge
- 20:06Generating embeddings
- 24:31Vector index for images
- 30:13Full-text search
- 37:51Knowledge graphs
- 42:27Data chunking and embeddings
- 49:37Graph building pipeline
- 56:50Graph visualization
- 1:08:16RAG retriever overview
- 1:10:04Vector retriever
- 1:19:29Vector Cypher retriever
- 1:27:40Hybrid retriever
- 1:35:14Text-to-Cypher retriever
- 1:38:29Multimodal application
- 1:55:22Final Q&A and wrap-up
related talks
38mRetrieval-Augmented Generation (RAG) in NLP
20mRetrieval Augmented Generation for AI Apps
16mEnhancing Retrieval-Augmented Generation with GraphRAG Patterns in Neo4j
01mtalk about RAG
51mDeveloping a Custom RAG System
38mBuilding an RAG in PHP: demystifying AI
50mRetrieval Augmented Generation Beyond Toy Demos
39mEnhancing RAG with Neo4j Knowledge Graph