masachika-kamada / streamlit-rag-chatbot

Streamlit RAG Chatbot

Overview

This is a web application that implements a question-answering system using the Retrieval-Augmented Generation (RAG) model. It combines GPT models with information retrieval techniques to generate appropriate responses to user queries.

Environment Setup

Create a container from VSCode's Dev Containers.

Setup Instructions

1. Use the .env.sample file as a reference to create a .env file and set your OpenAI API key and vector store type.

2. Change the URL in the process_webpage(url) function in retrievers.py.

3. Run python retrievers.py to create indexes in the vectorstore directory.

4. Execute the following command to start the application:

   streamlit run app.py
   

How to Use

• After launching the application, enter a question on the web interface.
• The RAG model generates a response to your question and displays it on the interface.