RAG

RAG Chatbot Application🤖

Introduction

This project implements a Context-Awarew Retrieval-Augmented Generation (RAG) chatbot using Streamlit.The chatbot is powered by the Mistral-7B-Instruct-v0.3 language model integrated with ChromaDB as vector database.

Table of Contents

• Installation
• Usage
• Features
• Dependencies
• Configuration
• License

Installation

To install and set up the project, follow these steps:

1. Clone the repository.

    git clone https://github.com/todap/RAG.git
   

2. Navigate to the project directory.

    cd RAG
   

3. Install the required dependencies.

    pip install -r requirements.txt
   

4. Set your Hugging Face token in the app.py file:

   HF_TOKEN = st.secrets["HF_TOKEN"]
   

Usage

1. Run the main application:

    streamlit run app.py
   

   ### OR -deployed on streamlit:https://team-qubits.streamlit.app/

2. Interact with the chatbot via the web interface.
3. Upload documents using the Document Management section and process them for use within the chatbot.

Features

1. Contextual Responses: The chatbot retrieves relevant documents from a knowledge base and uses them to provide contextual responses to user queries.
2. Conversational History: The chatbot maintains a conversation history, allowing it to reference and build upon previous interactions.
3. Document Management: The application provides a document management interface, allowing users to upload and store new documents in the knowledge base.
4. Feedback Mechanism: Users can provide feedback on the chatbot’s responses, which is used to improve the quality of future responses.

Dependencies

The project relies on the following major dependencies:

• streamlit
• huggingface_hub
• langchain
• chromadb

Configuration

• Store your Hugging Face token in the Streamlit secrets file as HF_TOKEN.
• Additional configuration options may be found within the app.py file.

Contributor