Skip to content

Latest commit

 

History

History
136 lines (91 loc) · 4.48 KB

README.md

File metadata and controls

136 lines (91 loc) · 4.48 KB
Raw

OpenAI RAG Application

This repository provides a full-stack application for document-based question answering using a Retrieval-Augmented Generation (RAG) system. The application consists of a frontend built with Next.js and a backend built with FastAPI, leveraging OpenAI's language models and Elasticsearch for document retrieval.

Table of Contents

Features

Frontend

  • Multiple Chats: Users can start and manage multiple chats, each with its own context and history.
  • Real-time Communication: Communicates with the RAG backend in real-time to provide prompt responses.

Backend

  • Document Retrieval: Uses Elasticsearch to connect with the vector database.
  • Custom Wrappers: Simplifies interactions with OpenAI and Elasticsearch.
  • FastAPI: Provides a production-ready API for document-based question answering.

Architecture

The application is composed of two main services:

  1. Backend Service: Handles document retrieval and interaction with OpenAI's models.
  2. Frontend Service: Provides a user interface for interacting with the backend.

Both services are orchestrated using Docker Compose.

Setup

To set up the application, ensure that you have Docker and Docker Compose installed. The following steps will guide you through the process:

  1. Clone the Repository:

    git clone https://github.com/Soumadip-Saha/openai-RAG-Application
cd openai-RAG-Application

  2. Configure Environment Variables:

    Modify the docker-compose.yml file to set the environment variables needed for your setup.

  3. Start the Application:

    docker-compose up -d

Usage

Frontend

To view the frontend application, open your browser and navigate to http://localhost:3000.

Backend

The backend service is available at http://localhost:5400.

Endpoints

The backend provides the following endpoints:

/build_context

  • Request:
    • userId: The unique ID for each user.
    • query: The query asked by the user.
    • chats: A list of dictionaries containing the previous question and answers.
  • Response:
    • query, stand_alone_query, docs, context, userId, references

/stream

  • Request:
    • userId: The unique ID for each user.
    • context: The context to generate the answer.
    • query: The query to be answered.
  • Response: SSE (Server-Sent Events) stream.

/evaluate_response

  • Request:
    • userId: The unique ID for each user.
    • answer: The answer to be evaluated.
    • stand_alone_query: The query used to generate the answer.
  • Response:
    • similar_queries, response_score, userId

/generate_ans

  • Request:
    • userId: The unique ID for each user.
    • query: The query asked by the user.
    • chats: A list of dictionaries containing the previous question and answers.
    • developer_details: Boolean indicating whether to return developer details.
  • Response:
    • answer, query, references, userId, stand_alone_query, docs, confidence_score, response_score, token_usage

Future Features

Frontend

  • Authentication: Secure user authentication to manage access and user-specific data.
  • Database Integration: Persistent storage of chat history and user data for a seamless experience across sessions.
  • Enhanced UI/UX: Improvements to the user interface and experience based on user feedback.

Backend

  • Logger Handling: Improved logging for better traceability and debugging.
  • Rate-Limit Handling: Managing API call rates to ensure service stability.

Contributing

We welcome contributions to enhance the functionality and usability of the application. To contribute:

  1. Fork the repository.
  2. Create a new branch (git checkout -b feature-branch).
  3. Commit your changes (git commit -am 'Add new feature').
  4. Push to the branch (git push origin feature-branch).
  5. Create a new Pull Request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

By providing a user-friendly interface and a robust backend, the OpenAI RAG Application aims to facilitate efficient document-based question answering, with a focus on scalability and future enhancements for a better user experience.