https://github.com/headless-start/pdf-summary-app

https://github.com/headless-start/pdf-summary-app

langchain pdf

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/headless-start/pdf-summary-app
• Owner: headless-start
• License: mit
• Created: 2025-02-09T10:57:07.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2025-02-10T13:58:40.000Z (3 months ago)
• Last Synced: 2025-02-10T14:50:05.604Z (3 months ago)
• Topics: langchain, pdf
• Language: Python
• Homepage:
• Size: 1000 Bytes
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# 📄 PDF Summary Tool  

## 📌 Project Overview  

This project demonstrates the **loading, preprocessing, and summarization** of text from **PDF files** using **Streamlit** and **LLM APIs like (Deepseek r1,o3-mini)**. The application allows users to upload a PDF, extract its text, and generate a concise summary. The summary can be downloaded as a `.txt` file for offline use.  

---

## 🚀 Key Features  

1. **PDF Summarization**  

   - Users can upload a PDF file (up to 200MB) and get the summary of the document.  

---

## 🔍 How It Works  

1. **Upload a PDF**:  

   - Users upload a PDF file using the file uploader in the app.  

2. **Extract Text**:  

   - The app extracts text from the PDF using the `PyPDF2` library.  

3. **Generate Summary**:  

   - The extracted text is sent to the LLM API (Deepseek r1, o3-mini), which generates a summary.  

4. **Display and Download**:  

   - The summary is displayed on the app, and users can download it as a `.txt` file.  

**Check Demo Here**:  

[![Open in Streamlit](https://static.streamlit.io/badges/streamlit_badge_black_white.svg)](https://pdfdeepv1.streamlit.app/)  

---

## 🛠 System Requirements  

### Dependencies  

- Python 3.8+  

- Libraries: `streamlit`, `PyPDF2`, `openai`  

- Hardware: CPU (GPU not required)  

---

## 📄 License  

This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.