meridianalgo-smarttrader 1.0.0

Ultra-Accurate AI Stock Analysis with Universal GPU Support (AMD • Intel • NVIDIA • Apple)

🚀 MeridianAlgo Smart Trader

Ultra-Accurate AI Stock Analysis with Universal GPU Support

Professional-grade stock analysis powered by ensemble machine learning with universal GPU acceleration. Features advanced volatility spike detection and real-time technical analysis.

✨ Key Features

• 🎯 Ultra-Accurate Predictions: Ensemble ML models (LSTM + Transformer + XGBoost)
• 🔥 Universal GPU Support: AMD • Intel • NVIDIA • Apple Silicon
• ⚡ Volatility Spike Detection: Predict market turbulence before it happens
• 📊 Real-time Analysis: Live market data with technical indicators
• 🎨 Clean Output: Simplified, essential information only
• 🚀 Easy Integration: Simple Python API and CLI

🚀 Quick Start

Installation

pip install meridianalgo-smarttrader

Command Line Usage

# Analyze Apple stock
smart-trader AAPL

# Custom parameters
smart-trader TSLA --days 90 --epochs 15

# Show GPU information
smart-trader --gpu-info

Python API Usage

from meridianalgo import SmartTrader, analyze_stock

# Simple analysis
result = analyze_stock('AAPL')
print(f"Current: ${result['current_price']:.2f}")
print(f"Tomorrow: ${result['predictions'][0]:.2f}")

# Advanced usage
trader = SmartTrader(verbose=True)
analysis = trader.analyze('TSLA', days=60, epochs=10)

# Check volatility spike risk
vol_risk = analysis['volatility_spike']['spike_probability']
if vol_risk > 60:
    print("⚠️ High volatility spike risk detected!")

📊 Sample Output

🚀 AAPL Analysis
Device: CPU (8 threads)

┌────────────┬──────────┬──────────────────┐
│ Metric     │ Value    │ Info             │
├────────────┼──────────┼──────────────────┤
│ Current    │ $213.96  │ Real-time        │
│ Day +1     │ $216.45  │ +1.2%            │
│ Day +2     │ $218.30  │ +2.0%            │
│ Day +3     │ $215.80  │ +0.9%            │
│ Confidence │ 84%      │ Model reliability│
│ Vol Risk   │ 23%      │ ✅ Low risk      │
└────────────┴──────────┴──────────────────┘

🔥 Universal GPU Support

Smart Trader automatically detects and optimizes for your GPU:

Vendor	Technology	Status
🟢 NVIDIA	CUDA	✅ Supported
🔴 AMD	ROCm/DirectML	✅ Supported
🔵 Intel	XPU	✅ Supported
🍎 Apple	MPS	✅ Supported

GPU Setup

# NVIDIA GPU
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121

# AMD GPU (Windows)
pip install torch-directml

# Intel GPU
pip install intel-extension-for-pytorch

# Apple Silicon (automatic)
pip install torch torchvision torchaudio

⚡ Volatility Spike Detection

Smart Trader's advanced algorithm analyzes historical volatility patterns to predict future market turbulence:

from meridianalgo import detect_volatility_spikes
import yfinance as yf

# Get stock data
data = yf.Ticker('AAPL').history(period='1y')

# Detect volatility spikes
spike_info = detect_volatility_spikes(data)

print(f"Spike Probability: {spike_info['spike_probability']:.1f}%")
print(f"Expected in: {spike_info['expected_spike_days']} days")
print(f"Risk Level: {spike_info['risk_level']}")

🎯 Advanced Features

Ensemble Models

• LSTM: Captures long-term dependencies
• Transformer: Attention-based pattern recognition
• XGBoost: Gradient boosting for robustness

Technical Indicators

• RSI (Relative Strength Index)
• MACD (Moving Average Convergence Divergence)
• Bollinger Bands
• Volume analysis

Risk Management

• Volatility spike prediction
• Market regime detection
• Confidence scoring
• Position sizing recommendations

📈 Performance

Metric	CPU	GPU
Training Time (10 epochs)	~2-3 seconds	~0.5-1 seconds
Batch Size	32	64+
Memory Usage	2-4 GB RAM	GPU VRAM
Accuracy	High	Higher

🛠️ Development

Local Installation

git clone https://github.com/MeridianAlgo/In-Python.git
cd In-Python
pip install -e .

Running Tests

pytest tests/

Building Package

python setup.py sdist bdist_wheel
twine upload dist/*

📚 Documentation

• GPU Setup Guide
• API Reference
• Examples
• Contributing

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

📄 License

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

🔗 Links

• PyPI: https://pypi.org/project/meridianalgo/
• GitHub: https://github.com/MeridianAlgo/In-Python
• Documentation: https://meridianalgo.github.io/In-Python/
• Issues: https://github.com/MeridianAlgo/In-Python/issues

🏆 About MeridianAlgo

MeridianAlgo specializes in advanced financial AI solutions. Our mission is to democratize professional-grade trading tools through cutting-edge machine learning and universal GPU acceleration.

---

Made with ❤️ by MeridianAlgo

Empowering traders with AI-driven insights