Skip to content

Latest commit

 

History

History
51 lines (32 loc) · 1.36 KB

README.md

File metadata and controls

51 lines (32 loc) · 1.36 KB

GastroBot

This is the official codebase for GastroBot.

Introduction

GastroBot is a RAG framework Chinese gastric disease chatbot. It uses 25 gastric disease-related guideline documents as a corpus, fine-tuned the embedding model, and is developed based on llama-index. An overview of GastroBot architecture can be seen below:

GastroBot architecture diagram

Getting Started

Step 1 Installation

Set up conda environment and clone the github repo

# create a new environment
$ conda create --name rag python=3.10.13 -y
$ conda activate rag
# install requirements
$ pip install -r requirements.txt

Step 2 Create index

Executing the code in the build_index.ipynb file will build the index locally. The built index is saved in the vector_index folder.

Step 3 Run app.py

$ streamlit run app.py

Step 4 Navigate to browser

Open the browser and enter the address http://127.0.0.1:port/ Start Q&A with it!

GastroBot Q&A

Acknowledgement