README.md

metadata

title: Cyclone Prediction & Resource Optimization API
emoji: 🌪️
colorFrom: blue
colorTo: red
sdk: docker
pinned: false
license: mit

🌪️ Cyclone Prediction & Resource Optimization API

Live ML-based cyclone intensity prediction with real-time IMD data integration and intelligent resource allocation for disaster management.

Features

• 🔮 ML Prediction: XGBoost model predicts cyclone intensity 24 hours ahead
• 🌐 Live IMD Integration: Fetches real-time data from India Meteorological Department
• 📊 Resource Optimization: Intelligent allocation of shelters, medical teams, and equipment
• 🚁 Field Team Routing: Automated deployment planning with route optimization
• 🗺️ District-Level Analysis: Risk assessment for all Odisha districts

API Endpoints

Live Data Endpoints (Recommended)

• GET /live/full-analysis - Complete live workflow (fetch → predict → optimize)
• GET /live/cyclones - Fetch active cyclone alerts from IMD
• GET /live/predict - Predict using live IMD bulletin
• GET /live/optimize - Resource optimization from live data

Manual Input Endpoints (Testing)

• POST /predict/manual - Predict from manual cyclone parameters
• POST /optimize/manual - Optimize resources from manual input
• POST /analyze/manual - Complete analysis from manual input

Data Endpoints

• GET /data/districts - List all Odisha districts with data
• GET /allocation/summary - Current allocation system status

Quick Start

Using the API

Visit /docs for interactive API documentation (Swagger UI).

Example: Get live analysis

curl https://your-space-name.hf.space/live/full-analysis

Example: Manual prediction

curl -X POST https://your-space-name.hf.space/predict/manual \
  -H "Content-Type: application/json" \
  -d '{
    "LAT": 15.5,
    "LON": 85.3,
    "MAX_WIND": 65,
    "MIN_PRESSURE": 990
  }'

Technology Stack

• Framework: FastAPI
• ML Model: XGBoost (trained on historical cyclone data)
• Data Source: IMD RSMC Live Bulletins
• Optimization: Greedy allocation + 2-opt routing

Model Performance

• MAE: ~8-10 knots
• RMSE: ~12-15 knots
• R²: ~0.85

Use Cases

• Emergency disaster management
• Resource pre-positioning
• Evacuation planning
• Real-time threat assessment

Credits

Built for disaster preparedness and response optimization in Odisha, India.