🌧️ RAIN - Real-time AI News Intelligence

AI-powered macroeconomic news analysis with industry-specific sentiment models and automated trading simulation.

🚀 Quick Start

# Install dependencies
pip install -r requirements.txt

# Run terminal application
python terminal_app.py
# or
./rain.sh

📚 Documentation

All documentation is in the docs/ folder:

• START_HERE.md - Training quickstart guide
• INDUSTRY_ANALYSIS_ROADMAP.md - 3-month implementation plan
• PROJECT_STRUCTURE.md - Complete codebase layout
• PHASE4_COMPLETE.md - Advanced features
• DATABASE_GUIDE.md - Database setup
• DOCKER_GUIDE.md - Docker deployment

⚡ Key Features

Bloomberg-Style Terminal

• Interactive menu system
• Real-time news analysis
• Trade simulation & PnL tracking
• Performance metrics dashboard

Industry-Specific Analysis (6 Sectors)

• Tech: Bullish/Bearish/Neutral
• Finance: Hawkish/Dovish/Neutral
• Medicine: Breakthrough/Setback/Neutral
• Real Estate: Expansionary/Contractionary/Neutral
• Defense: Escalation/De-escalation/Neutral
• Politics: Progressive/Conservative/Neutral

Custom FinBERT Training

• Fine-tune models on your data
• Sector-specific sentiment labels
• Interactive labeling tool
• 600+ stocks tracked (top 100 per sector)

Advanced Trading System

• Multiple strategies (Sentiment Fusion, Composite)
• Risk management (position limits, drawdown protection)
• Technical indicators (RSI, MACD, Bollinger Bands)
• Performance metrics (Sharpe, Sortino, Max Drawdown)

🎯 Make Commands

make run              # Run terminal application
make train-prepare    # Prepare training data
make train-tech       # Train Tech sector model
make build            # Build Docker image
make up               # Start with Docker
make test-strategy    # Test trading strategies
make test-risk        # Test risk management

🏗️ Project Structure

RAIN/
├── terminal_app.py          # Main application
├── train_tech_model.py      # Model training
├── config/                  # Sector configurations
├── training/                # FinBERT training system
├── scripts/                 # Data pipeline scripts
├── trading/                 # Trading simulation
├── strategies/              # Trading strategies
├── risk/                    # Risk management
├── indicators/              # Technical indicators
├── models/                  # ML models
├── utils/                   # Utilities (cache, retry, metrics)
└── docs/                    # 📚 All documentation

🔧 Setup

1. Clone & Install

   pip install -r requirements.txt
   python -m spacy download en_core_web_sm

2. Configure API Keys

   cp .env.example .env
   # Add your API keys to .env

3. Run

   python terminal_app.py

📊 Tech Stack

• NLP: FinBERT, Transformers, spaCy
• Trading: Custom strategies, risk management
• Data: NewsAPI, Alpha Vantage
• Cache: Redis
• UI: Rich (terminal)
• Container: Docker

🎓 Training Custom Models

See docs/START_HERE.md for complete training guide.

# 1. Prepare data
python scripts/prepare_training_data.py

# 2. Label interactively
python start_labeling.py

# 3. Train model
python train_tech_model.py

📈 Performance

• Redis Caching: 70-80% API call reduction
• Sentiment Accuracy: 80%+ with custom training
• Trading Strategies: Multiple signal types with ensemble voting
• Risk Management: Automated position sizing & stop loss

🐳 Docker

make build && make up
# or
docker-compose up -d

📝 License

MIT

🤝 Contributing

1. Fork the repository
2. Create feature branch
3. Make changes
4. Submit pull request

---

Documentation: See docs/ folder
Issues: Report on GitHub
Status: ✅ Production ready