Summarize.ai

Title: PDF to Summary Documentation

Project Overview

Project Name: PDF to Summary

URL: https://summarizer-aigurukul.vercel.app/

PDF to Summary is a web application that allows users to upload PDF files and receive a concise summary of their content. This documentation provides a detailed guide on the project's structure, tech stack, deployment process, and usage instructions.

New Feature Added

Quiz Generator : You can now generate quiz based on the uploaded pdf file

• Just upload the pdf file and click on generate quiz button and you will get the quiz
• You can also solve the quiz and get the marks

Tech Stack

Backend

• Language: Python
• Framework: FastAPI
• SDK: Gorq SDK
• Modal: Meta Llama 3 70B Model

Frontend

• Framework: Next.js (based on ReactJS)
• Libraries:
  • Shadcn (UI library)
  • Tailwind CSS

Deployment

• Backend: DigitalOcean (previously on Render)
• Frontend: Vercel

Project Structure

Backend (FastAPI)

1. Main Application File: main.py
   • Contains the FastAPI app initialization, route definitions, and Gorq SDK integration.
2. Routes:
   • /health: Endpoint to check the health of the Application.
   • /summarize: Endpoint to return the summary of the uploaded PDF.
   • /chat: Endpoint can chat with the model.
   • /ytsummarize: Endpoint to return the summary based on URL of Youtube Video
   • /ytchat : Endpoint can chat with the transcripted data from Youtube Video.
   • /ytquiz : Endpoint can generate the quiz with the transcripted data from YouTube.
3. Dependencies:
   • requirements.txt: Lists all Python dependencies.
   • render.yaml: Configuration file for deploying on Render.

Frontend (Next.js)

1. Pages:
   • page.js: Main landing page where users can log In to the Application.
   • dashboard/page.js: Page to display the main interface of the application.
2. Components:
   • UI: Component for shadcn UI elements.
3. Styles:
   • Tailwind CSS for styling.

Deployment Guide

Backend Deployment (Render)

1. Create Render Account: Sign up at Render.
2. Add Project:
   • Connect your GitHub repository containing the backend code.
3. Configuration:
   • Ensure requirements.txt is present for dependencies.
   • Add render.yaml for deployment configuration.
4. Deploy:
   • Follow the prompts on Render to deploy the backend server.

Frontend Deployment (Vercel)

1. Create Vercel Account: Sign up at Vercel.
2. Add Project:
   • Connect your GitHub repository containing the frontend code.
3. Configuration:
   • Ensure next.config.js is properly set up.
   • Deploy directly from the Vercel dashboard.
4. Deploy:
   • Follow the prompts on Vercel to deploy the front end.

Usage Instructions

1. Access the Application:
   • Go to https://summarizer-aigurukul.vercel.app/.
2. Upload a PDF:
   • On the main page, use the upload form to select and upload a PDF file.
3. Get the Summary:
   • After uploading, the application processes the PDF and displays a summary on the page.

Troubleshooting

Common Issues and Solutions

1. Backend Server Not Responding:
   • Ensure the backend server on Render is running.
   • Check network requests in the browser's developer tools for errors.
2. CORS Issues:
   • Verify CORS settings in FastAPI.
   • Ensure the frontend is correctly configured to make requests to the backend.
3. Deployment Errors:
   • Check deployment logs on Render and Vercel for error messages.
   • Ensure all dependencies are listed in requirements.txt (backend) and package.json (frontend).

Additional Resources

• FastAPI Documentation
• Next.js Documentation
• Gorq SDK Documentation
• Render Deployment Guide
• Vercel Deployment Guide

Conclusion

PDF to Summary is a robust application designed to simplify the process of extracting summaries from PDF documents. By leveraging FastAPI for the backend and Next.js for the front end, the application provides a seamless user experience. Deployment on Render and Vercel ensures that the application is scalable and easy to manage. This documentation serves as a comprehensive guide to understanding, deploying, and using the application effectively.