This is a Streamlit web app that uses OpenAI's GPT-3.5 Turbo model and llm to provide a PDF text-based question-answering system. Users can upload PDF files, ask questions about the content, and receive answers generated by the GPT-3.5 Turbo model. This README provides an overview of the app's functionality and how to use it.
PDF Upload: Users can upload PDF files containing text documents.
Question Asking: Users can ask questions related to the uploaded PDF content.
Answer Generation: The app uses OpenAI's GPT-3.5 Turbo model to generate answers to user questions.
Summary Generation: The app can generate a summary of the uploaded PDF document.
Message History: The app keeps a history of user questions and model-generated responses.
The following packages are required to run this app:
- streamlit
- streamlit-chat
- PyPDF2
- langchain
- faiss
- openai
- tiktoken
- requests
To install these packages, you can run:
pip install -r requirements.txt
To use this app, follow these steps:
- Upload your data file.
- Ask your question about the data file.
The app will then search for the answer to your question in the uploaded PDF file using the OpenAI API. The answer will be displayed in the "Past questions and answers" section.
The app may require an OpenAI API key, and API usage may be subject to OpenAI's pricing policies. This app currently supports PDF files, CSV files. File formats like markdown, txts, and also support the open-source model of llm that is llama-2.