/QnAChatBot

A chatbot that parses your PDF files and answers your questions around that file using GPT

Primary LanguageJupyter Notebook

QnAChatBot

Live Demo

QnAChatBot allows you to upload text heavy pdf documents, and chat with the bot based on the document.

Use Cases

  1. Company financials, and earnings call analyzer.
  2. Ask questions about a Terms and Services document to uncover what you're legally signing up for.
  3. Write cover letters based on resumes.
  4. Ask questions on course work, lab materials, or any educational document

Prior Work

This work is inspired from my HackNYU Hackathon project, GPT Tutor. We built a Discord Bot that is able to synthesize course material uploaded by a professor, and serve as a 24/7 available teacher's assistant. The success of the work enables us to expand beyond just coursework, and extend it to a wider usecase.

To Reproduce (On MacOS)

  1. Clone this repository
git clone https://github.com/farhan0167/QnAChatBot
cd QnAChatBot
  1. Configure OpenAI API key and secrets folder
mkdir .streamlit
cd .streamlit
nano secrets.toml

and paste the following:

OPENAI_API_KEY = "your-openAI-api-key"

Press Control+X and Y to exit the editor. Go back to project directory:

  1. Install dependencies
pip install -r requirements.txt
  1. Run the Application
streamlit run app.py