E-RAG: Advancing Retrieval-Augmented Generation for Entity-oriented Question Answering
Author: Pingwy Wu, Jing Tang, Huimin Chen
git clone https://github.com/wpydcr/E-RAG.git
cd E-RAGconda create -n erag python=3.10 -y
conda activate erag
pip install -r requirements.txtyou can download dataset with this link, and place it in data/.
Note: If it's the first time running and gpt-3.5-turbo is available, it will automatically generate the corresponding embedding file for the data. If you don't want to regenerate it, you can also download the embedding file through this link and place it in data/.
You need to configure gpt-3.5-turbo properly in config/base.json.
run python webui.py to start.