rag chatbot

Chat with multiple PDFs locally

Setup

1. Kaggle (Recommended)

• Import notebooks/kaggle.ipynb to Kaggle

• Replace with your tokens.

2. Local

2.1. Clone project

git clone https://github.com/datvodinh/rag-chatbot.gitcd rag-chatbot

2.2 Install

2.2.1 Docker

docker compose up --build

2.2.2 Using script (Ollama, Ngrok, python package)

source ./scripts/install_extra.sh

2.2.3 Install manually

1. Ollama

• MacOS, Window: Download

• Linux

curl -fsSL https://ollama.com/install.sh | sh

2. Ngrok

• Macos

brew install ngrok/ngrok/ngrok

• Linux

curl -s https://ngrok-agent.s3.amazonaws.com/ngrok.asc 
| sudo tee /etc/apt/trusted.gpg.d/ngrok.asc >/dev/null 
&& echo "deb https://ngrok-agent.s3.amazonaws.com buster main" | sudo tee /etc/apt/sources.list.d/ngrok.list 
&& sudo apt update 
&& sudo apt install ngrok

3. Install rag_chatbot Package

source ./scripts/install.sh

2.3 Run

source ./scripts/run.sh

or

python -m rag_chatbot --host localhost

• Using Ngrok

source ./scripts/run.sh --ngrok

3. Go to: http://0.0.0.0:7860/ or Ngrok link after setup completed

Todo

• Add evaluation.

• Better Document Processing.

• Support better Embedding Model for Vietnamese and other languages.

• ReAct Agent.

• Document mangement (Qrdant, MongoDB,...)