📄 DOC ANALYZER

Doc Analyzer is an AI-powered document processing backend that extracts text from PDF files using OCR, generates summaries, and enables question-answering based on extracted content. The application is built with FastAPI, packaged using Docker, and deployed on AWS EC2 using a production-grade, image-based architecture that pulls versioned Docker images instead of code.

---

🚀 Features

• Upload PDF documents via API
• OCR text extraction using Tesseract
• AI-based summarization and Q&A
• FastAPI asynchronous architecture
• Fully containerized using Docker
• Image-based deployment with GitHub Actions
• Runs on AWS EC2 with Docker Compose
• Persistent document storage support

---

⚙️ Local Setup

1. Clone the repository
   git clone https://github.com/codicecustode/doc-analyzer.git
cd doc-analyzer

2. Create and activate virtual environment
   Windows:
   python -m venv ai-doc-analyzer-venv
ai-doc-analyzer-venv\Scripts\activate
   Linux / Mac:
   python3 -m venv ai-doc-analyzer-venv
source ai-doc-analyzer-venv/bin/activate

3. Install dependencies
   pip install -r requirements.txt

4. Run FastAPI server
   uvicorn app.main:app --reload

5. Access URLs
   Application: http://localhost:8000/api/v1
   Swagger Docs: http://localhost:8000/docs

---

🐳 Docker Setup (Local)

1. Build image
   docker build -t doc-analyzer .

2. Run container
   docker run -p 8000:8000 doc-analyzer

3. Access application
   http://localhost:8000/api/v1

---

📦 Docker Hub Integration

1. Login
   docker login

2. Tag image
   docker tag doc-analyzer <username>/doc-analyzer:<tag>

3. Push image
   docker push <username>/doc-analyzer:<tag>

4. Pull anywhere
   docker pull <username>/doc-analyzer:<tag>

Images are automatically built and pushed via GitHub Actions on each push to main.

---

☁️ AWS EC2 Deployment (Production)

This application uses image-based deployments — EC2 never pulls source code, only Docker images.

1. Launch EC2 Instance
   OS → Ubuntu 22.04
   Open Port 80 in inbound rules

2. Connect to EC2
   ssh ubuntu@<EC2_PUBLIC_DNS>

3. Install Docker & Docker Compose
   sudo apt update && sudo apt install -y docker.io docker-compose
sudo usermod -aG docker ubuntu
   (logout and login again to apply permissions)

4. Create deployment directory
   mkdir -p /home/ubuntu/doc-analyzer
cd /home/ubuntu/doc-analyzer

5. Create .env file and add required environment variables

   .env

   IMAGE_TAG=latest SECRET_KEY=<your_secret> GEMINI_API_KEY=<your_secret> PINECONE_API_KEY=<your_secret> OPENAI_API_KEY=<your_secret> GOOGLE_API_KEY=<your_secret> MONGO_DB_URI=<your_url> MONGO_DB_NAME=<your_db_name>

6. Create docker-compose.yml in the SAME directory

   docker-compose.yml

   version: '3.8' services: doc-analyzer: image: /doc-analyzer:${IMAGE_TAG} restart: unless-stopped ports: - "80:8000" env_file: - .env volumes: - uploads_data:/usr/python/app/doc-analyzer/app/uploads

   volumes: uploads_data:

7. Start the application
   docker-compose pull
docker-compose up -d

8. Verify running services
   docker-compose ps

9. Access the application in browser
   http://<EC2_PUBLIC_DNS>