meridianalgo 0.2.0

A Python library for algorithmic trading and financial analysis with AI/ML capabilities

MeridianAlgo - Enhanced Trading Library with AI/ML

A comprehensive Python library for algorithmic trading and financial analysis, now enhanced with advanced AI and machine learning capabilities.

🚀 New Features in v0.2.0

• ML Predictor: Advanced machine learning models for stock price prediction
• AI Analyzer: AI-powered market sentiment and regime analysis
• Ensemble Models: Multiple ML models working together for better accuracy
• GPU Support: Automatic detection and optimization for NVIDIA, AMD, Intel, and Apple GPUs
• Enhanced Technical Analysis: 17+ technical indicators with ML integration

📦 Installation

Basic Installation

pip install meridianalgo

With ML Support

pip install meridianalgo[ml]

Full Installation (with visualization)

pip install meridianalgo[ml,visualization]

🔧 Quick Start

Traditional Trading Features

from meridianalgo import TradingEngine, BacktestEngine, Indicators

# Initialize trading engine
engine = TradingEngine(paper_trading=True)
engine.connect()

# Place orders
order = engine.place_order("AAPL", "buy", 10, "market")
print(f"Order placed: {order}")

# Calculate technical indicators
prices = [100, 102, 101, 103, 105, 104, 106]
sma = Indicators.sma(prices, period=5)
rsi = Indicators.rsi(prices, period=14)

New ML Features

from meridianalgo import MLPredictor, AIAnalyzer, EnsembleModels
import yfinance as yf

# ML-based price prediction
predictor = MLPredictor()

# Simple prediction (no ML dependencies required)
result = predictor.predict_simple("AAPL", days=60, forecast_days=5)
print(f"Predictions: {result['predictions']}")
print(f"Confidence: {result['confidence']}%")

# Advanced ML prediction (requires PyTorch)
ml_result = predictor.predict_ml("AAPL", days=60, epochs=20)
print(f"ML Predictions: {ml_result['predictions']}")

# AI-powered market analysis
analyzer = AIAnalyzer()
data = yf.Ticker("AAPL").history(period="3mo")

analysis = analyzer.comprehensive_analysis(data, "AAPL")
print(f"Market Sentiment: {analysis['sentiment_analysis']['sentiment']}")
print(f"Market Regime: {analysis['market_regime']['regime']}")
print(f"Overall Score: {analysis['overall_score']}")

# Ensemble models for enhanced accuracy
ensemble = EnsembleModels()
data_with_indicators = predictor.calculate_technical_indicators(data)
X, y = ensemble.prepare_ensemble_data(data_with_indicators)

# Train ensemble
training_results = ensemble.train_ensemble(X, y, epochs=15, verbose=True)
print(f"Training completed: {training_results}")

# Make ensemble predictions
predictions = ensemble.predict_ensemble(X, forecast_days=5)
print(f"Ensemble predictions: {predictions['ensemble_predictions']}")
print(f"Confidence: {predictions['confidence']}%")

🧠 AI/ML Features

MLPredictor

• Simple Prediction: Statistical ensemble methods (no ML dependencies)
• Advanced ML: Neural networks with PyTorch
• Technical Indicators: 14+ indicators automatically calculated
• Multi-GPU Support: CUDA, MPS, DirectML, XPU

AIAnalyzer

• Market Sentiment: Bullish/bearish/neutral analysis
• Market Regime: Trending, ranging, volatile detection
• Support/Resistance: Automatic level identification
• Volume Analysis: Volume profile and patterns
• AI Insights: Optional integration with Gemini AI

EnsembleModels

• Multiple Models: Linear Regression, Random Forest, Neural Networks
• Automatic Weighting: Performance-based model combination
• Confidence Scoring: Prediction reliability assessment
• Hardware Optimization: GPU acceleration when available

📊 Traditional Features

TradingEngine

• Paper and live trading support
• Position management
• Order execution
• P&L calculation

BacktestEngine

• Strategy backtesting
• Performance metrics
• Equity curve analysis
• Trade history

Indicators

• 15+ technical indicators
• Moving averages (SMA, EMA)
• Oscillators (RSI, Stochastic, Williams %R)
• Trend indicators (MACD, Bollinger Bands)
• Volume indicators

TradeUtils

• Position sizing
• Risk management
• Performance calculations
• Trade validation

🖥️ Hardware Acceleration

Automatic detection and optimization for:

• NVIDIA GPUs: CUDA support
• AMD GPUs: ROCm (Linux) / DirectML (Windows)
• Intel Arc GPUs: XPU support
• Apple Silicon: MPS acceleration
• CPU: Multi-threaded fallback

📈 Example: Complete ML Trading Workflow

import yfinance as yf
from meridianalgo import MLPredictor, AIAnalyzer, BacktestEngine

# 1. Fetch data
symbol = "AAPL"
data = yf.Ticker(symbol).history(period="1y")

# 2. AI Analysis
analyzer = AIAnalyzer()
analysis = analyzer.comprehensive_analysis(data, symbol)
print(f"Market Analysis: {analysis['sentiment_analysis']['sentiment']}")

# 3. ML Prediction
predictor = MLPredictor()
predictions = predictor.predict_ml(symbol, days=90, epochs=25)
print(f"5-day forecast: {predictions['predictions']}")

# 4. Create trading strategy based on ML predictions
def ml_strategy(row, positions, capital, predictions_data):
    # Simple strategy: buy if prediction is higher than current price
    current_price = row['close']
    if predictions_data and len(predictions_data) > 0:
        predicted_price = predictions_data[0]  # Next day prediction
        
        if predicted_price > current_price * 1.02:  # 2% threshold
            return {'symbol': 'AAPL', 'action': 'buy', 'quantity': 10}
        elif predicted_price < current_price * 0.98:  # -2% threshold
            return {'symbol': 'AAPL', 'action': 'sell', 'quantity': 10}
    
    return None

# 5. Backtest the strategy
backtest = BacktestEngine(initial_capital=10000)
backtest.load_data(data.reset_index().rename(columns={
    'Date': 'timestamp', 'Open': 'open', 'High': 'high', 
    'Low': 'low', 'Close': 'close', 'Volume': 'volume'
}))

results = backtest.run_backtest(ml_strategy, predictions_data=predictions['predictions'])
print(f"Backtest Results: {results}")

🔧 Configuration

Environment Variables

# Optional: Enable AI insights
export GEMINI_API_KEY="your_gemini_api_key"

GPU Setup

The library automatically detects available hardware. For optimal performance:

• NVIDIA: Install CUDA toolkit
• AMD: Install ROCm (Linux) or DirectML (Windows)
• Intel: Install Intel Extension for PyTorch
• Apple: macOS 12.3+ with Apple Silicon

📚 Documentation

• API Reference
• ML Guide
• Examples
• GPU Setup

🤝 Contributing

Contributions are welcome! Please read our Contributing Guide for details.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🆕 What's New in v0.2.0

• Added ML-based price prediction with ensemble models
• Integrated AI-powered market analysis
• GPU acceleration support for major vendors
• Enhanced technical indicators with ML integration
• Comprehensive backtesting with ML strategies
• Optional AI insights via Gemini API
• Improved performance and reliability

🔮 Roadmap

• Deep learning models (LSTM, Transformer)
• Reinforcement learning for strategy optimization
• Real-time data streaming
• Advanced portfolio optimization
• Options and derivatives support
• Multi-asset backtesting

---

Made with ❤️ by MeridianAlgo