Skip to content

Latest commit

 

History

History
29 lines (21 loc) · 699 Bytes

File metadata and controls

29 lines (21 loc) · 699 Bytes

Retrieval Augmented Generation with Graph Knowledge Store

Introduction

  • Used Neo4j Graph to store knowledge graph
  • Customized Memory Stream to store entities of personal knowledge

Installation

Using conda env on Notebook

conda env create -n rag python=3.11
conda activate rag
pip install -r requirements.txt
conda install ipykernel
python -m ipykernel install --user --name=rag
jupyter notebook

Run Streamlit Application

README for Streamlit

Update