Table of Contents
This is the project home for NWU's library chatbot. The chatbot encapsulates the natural language understanding component of the physical robotic assistant found in the library.
To get a local copy up and running follow these simple example steps.
This project was developed and tested on Ubuntu 20.04
- Install
sudo apt update sudo apt install python3
- To use the chatbot, you need to set up a Qdrant instance with the knowledge base and an OpenAI account for LLM capabilities. Additionally, the application requires a
config.json
file in theactions
folder, containing the following key-value pairs:{ "openai_api_key": "<your_openai_api_key>", "model": "<model_name>", "token_limit": <token_limit>, "temperature": <temperature_value>, "qdrant_key": "<your_qdrant_api_key>", "qdrant_url": "<your_qdrant_url>", "vector_size": <vector_size>, "collection_name": "<collection_name>"}
- Clone the repo - make sure you have the correct access
git clone https://github.com/FrancisTembo/libbot.git cd libbot
- Install poetry:
curl -sSL https://install.python-poetry.org | python3 -
- Install dependancies:
poetry install
- Activate poetry environment
poetry shell
- Train the Rasa assistant
rasa train
- Starting the bot in the terminal
rasa shell
The app.sh script can start the Chabot as a background process via nohup. A more elegant solution would be to set this up as a service or package it via a Docker container.
- Implementing current library FAQ.
- Add
Telegram
channel.
This repository belongs to the North-West University LIbrary
Francis Tembo: [email protected]