QUICKSTART.md

Document Intelligence System - Quick Start Guide

🚀 Getting Started in 5 Minutes

Option 1: Local Installation (Using Python)

Step 1: Install Dependencies

# Install Python dependencies
pip install -r requirements.txt

# Install Tesseract OCR
# Windows: Download from https://github.com/UB-Mannheim/tesseract/wiki
# Linux: sudo apt-get install tesseract-ocr
# macOS: brew install tesseract

Step 2: Run the Application

# Start the API server
python main.py

Step 3: Access Dashboard

Open your browser and go to:

http://localhost:8000/dashboard

---

Option 2: Docker Installation

Step 1: Build and Run

# Using Docker Compose (recommended)
docker-compose up -d

# Or using Docker directly
docker build -t doc-intelligence .
docker run -p 8000:8000 doc-intelligence

Step 2: Access Dashboard

http://localhost:8000/dashboard

---

📖 Usage Examples

1. Extract Data from Text

Via Dashboard

1. Go to the "Extract from Text" card
2. Paste your document text
3. Click "Extract Data"
4. View results

Via API

curl -X POST "http://localhost:8000/extract" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Invoice #123 Amount Due: $500"
  }'

Via Python

import asyncio
from app.pipeline import DocumentProcessingPipeline

async def main():
    pipeline = DocumentProcessingPipeline()
    result = await pipeline.process_document(
        document_id="doc_001",
        text="Your document text here"
    )
    print(result.classification.document_type)
    print(result.extraction.extracted_fields)

asyncio.run(main())

2. Upload Document

Via Dashboard

1. Click "Upload Document" card
2. Select a file (image, PDF, or text)
3. System processes automatically
4. View results in "Results" tab

Via API

curl -X POST "http://localhost:8000/upload" \
  -F "file=@document.pdf"

3. Batch Processing

Via API

curl -X POST "http://localhost:8000/batch" \
  -H "Content-Type: application/json" \
  -d '{
    "documents": [
      {"text": "Document 1 text"},
      {"text": "Document 2 text"},
      {"text": "Document 3 text"}
    ]
  }'

Via Python

documents = [
    {"id": "doc_1", "text": "Invoice text..."},
    {"id": "doc_2", "text": "Receipt text..."}
]
results = asyncio.run(pipeline.process_batch(documents))
stats = pipeline.get_statistics(results)
print(stats)

---

🔍 Interactive Dashboard Features

Features Available

• ✅ Upload Documents - Upload images, PDFs, or text files
• ✅ Extract Data - Extract structured data from text
• ✅ View Results - See extraction and validation results in real-time
• ✅ Monitor Jobs - Track processing jobs and their status
• ✅ View Statistics - Monitor system performance
• ✅ API Documentation - Browse available endpoints

Navigation

Tab	Purpose
Results	View processing results
Jobs	Monitor active and completed jobs
API Docs	View API endpoint documentation

---

📊 How It Works

Processing Pipeline

Document Input
    ↓
[OCR] Extract text from images
    ↓
[Classify] Determine document type
    ↓
[Extract] Pull out structured data
    ↓
[Validate] Check data quality
    ↓
Results & Insights

Example: Processing an Invoice

1. Input: Invoice image or text
2. Classification: "invoice" (95% confidence)
3. Extraction:
   • Invoice Number: INV-2024-001
   • Date: 01/15/2024
   • Amount: $500.00
   • Vendor: Acme Corp
4. Validation:
   • All required fields present ✓
   • Data format correct ✓
   • Quality score: 92%
5. Output: Structured JSON with confidence scores

---

🎯 Supported Document Types

1. Invoice - Bill, sales invoice, receipt
2. Receipt - Purchase receipt, transaction record
3. Contract - Legal agreement, terms & conditions
4. Report - Business report, analysis
5. Email - Email message, correspondence
6. Form - Application, questionnaire
7. Letter - Business letter, notification

---

⚙️ Configuration

Environment Variables

Create a .env file in the root directory:

# API Configuration
API_HOST=0.0.0.0
API_PORT=8000
DEBUG=false

# Database
DATABASE_URL=sqlite:///./documents.db

# OCR Settings
OCR_LANG=eng
OCR_PSM=3

# File Upload
UPLOAD_DIR=./uploads
MAX_FILE_SIZE=52428800

# Logging
LOG_LEVEL=INFO

For Production

# Use PostgreSQL instead of SQLite
DATABASE_URL=postgresql://user:password@localhost/doc_intelligence

# Use Redis for caching
REDIS_URL=redis://localhost:6379

# Enable CORS
ENABLE_CORS=true

# Disable debug mode
DEBUG=false

---

🧪 Testing

Run Examples

python examples.py

This will:

• ✓ Process a single invoice document
• ✓ Batch process multiple documents
• ✓ Demonstrate custom field extraction
• ✓ Show validation results

Check Health

curl http://localhost:8000/health

---

📈 API Endpoints

Method	Endpoint	Purpose
GET	/	API information
GET	/health	Health check
POST	/upload	Upload document file
POST	/extract	Extract from text
POST	/batch	Batch process documents
GET	/jobs	List all jobs
GET	/jobs/{job_id}	Get job status
GET	/stats	System statistics
GET	/dashboard	Web dashboard

---

🐳 Docker-Compose Services

The included docker-compose.yml sets up:

1. App - Main FastAPI application (port 8000)
2. PostgreSQL - Database (port 5432)
3. Redis - Cache/Queue (port 6379)

Start All Services

docker-compose up -d

View Logs

docker-compose logs -f app

Stop Services

docker-compose down

---

📁 Project Structure

Agentic-Doc-Intelligence/
├── agents/
│   ├── classifier.py      # Document classification
│   ├── extractor.py       # Data extraction
│   └── validator.py       # Data validation
├── tools/
│   ├── ocr_engine.py      # OCR processing
│   └── table_parser.py    # Table extraction
├── app/
│   ├── pipeline.py        # Main pipeline
│   ├── main.py            # FastAPI app
│   └── README.md          # Documentation
├── database.py            # Database models
├── settings.py            # Configuration
├── utils.py               # Utilities
├── examples.py            # Usage examples
├── requirements.txt       # Dependencies
├── Dockerfile             # Docker image
├── docker-compose.yml     # Docker services
└── .env.example           # Environment template

---

🚨 Troubleshooting

"Tesseract not found" Error

Solution: Install Tesseract OCR

• Windows: Download installer from GitHub
• Linux: sudo apt-get install tesseract-ocr
• macOS: brew install tesseract

"Port 8000 already in use" Error

Solution: Change port in settings or stop the service using port 8000

Low OCR Accuracy

Tips:

• Ensure image resolution is at least 150 DPI
• Keep documents upright and well-lit
• Use PNG or TIFF formats
• Remove noise/shadows if possible

Database Errors

Solution: Reset the database

rm documents.db
python database.py

---

📚 Additional Resources

• FastAPI Documentation
• Tesseract OCR
• SQLAlchemy ORM
• OpenCV Documentation

---

🎓 Learning Path

1. Start with the Dashboard to familiarize yourself
2. Run examples.py to see different capabilities
3. Try the API using curl or Python
4. Explore the source code to understand the architecture
5. Customize for your use case with custom fields and schemas

---

💡 Tips & Tricks

• Batch Processing: Process multiple documents at once for efficiency
• Custom Fields: Define extraction patterns for your specific needs
• Validation Schemas: Create custom validation rules
• Caching: Enable Redis for faster repeated processing
• Database: Upgrade to PostgreSQL for production use

---

📞 Support

For issues or questions:

1. Check the detailed README
2. Review examples.py
3. Check API documentation at /docs

---

Ready to process documents? 🚀

Start the application and go to: http://localhost:8000/dashboard