Noga AI - Unified API

API untuk analisis nutrisi makanan menggunakan OCR + Google Gemini AI. Service ini menggabungkan OCR ingredients, nutrition facts, AI summary, dan validation dalam satu API.

🚀 Quick Start

1. Setup Environment

# Copy dan edit .env file
cp .env.example .env

# Install dependencies
pip install -r requirements.txt

2. Jalankan Server

# Development
python main.py

# Production
uvicorn main:app --host 0.0.0.0 --port 8080

Note: Database table akan otomatis dibuat saat pertama kali server dijalankan.

3. Akses API Docs

• Swagger UI: http://localhost:8080/docs
• ReDoc: http://localhost:8080/redoc

🐳 Docker

docker build -t noga-ai .
docker run -p 8080:8080 --env-file .env noga-ai

Note: Docker dan deployment otomatis setup database via init_db.py saat startup.

📡 API Endpoints

Endpoint	Method	Deskripsi
/ocr-ingredients	POST	Extract ingredients dari foto label makanan
/ocr-nutrition-info	POST	Extract nutrition facts dari foto label
/summary_foods	POST	Generate AI summary & health analysis
/validate-fix-typo	POST	Validasi dan perbaiki typo otomatis

📝 Contoh Request

Extract Ingredients

POST /ocr-ingredients
{
  "session_id": "user123",
  "composition": "https://example.com/image.jpg"
}

Extract Nutrition Info

POST /ocr-nutrition-info
{
  "session_id": "user123",
  "nutrition_info": "https://example.com/nutrition.jpg"
}

Generate Summary

POST /summary_foods
{
  "ingredients": [
    {"nama": "Sugar", "status": "bad", "detail": "High in empty calories"}
  ],
  "nutrition_info": [
    {"nama": "Calories", "nilai": 250, "type": "kilocalorie", "status": "neutral"}
  ]
}

⚙️ Environment Variables

DB_HOST=your_mysql_host
DB_PORT=3306
DB_USER=your_username
DB_PASSWORD=your_password
DB_NAME=your_database
GOOGLE_API_KEY=your_gemini_api_key
PORT=8080

🏗️ Tech Stack

• FastAPI - Web framework
• Google Gemini AI (2.0-flash) - OCR & AI processing
• MySQL - Database untuk menyimpan hasil OCR
• Uvicorn - ASGI server

📦 Struktur Proyek

Noga-AI/
├── main.py              # FastAPI app & routing
├── handlers/            # Business logic per endpoint
├── functions/           # OCR & AI utilities
├── schemas/             # Pydantic models
├── db/                  # Database connection
└── models/              # Health classification data