VecDBCompare

VecDBCompare is a straightforward Streamlit application designed to evaluate and compare three distinct vector database retrieval strategies.

Features

• Upload a PDF: Users can upload a PDF file, which will be utilized for the subsequent retrieval strategies.

• Three Retrieval Strategies: The application offers three different retrieval strategies for users to select and test:

  1. Smaller Chunks
  2. Summary
  3. Hypothetical Questions

• Real-time Feedback: Users can see the results of each strategy in real-time and compare them as needed.

Installation & Running

1. Clone this repository locally.
2. Install the required dependencies: pip install streamlit openai langchain
3. Run the application: streamlit run app.py

Note

• Ensure you have set the correct OpenAI API key.
• Due to the inclusion of API keys and other sensitive details, avoid making this code public in open repositories.

Contributing

Contributions to this project are welcome! Please fork this repository first, make your changes, and then create a pull request.

License

This project is under the MIT License.