MDPS - Multi-Dimensional Prediction System

🚀 Enhanced Trading System with Advanced PyQt Interface

MDPS v2.0 is a comprehensive trading system that combines advanced machine learning models, real-time market analysis, and sophisticated risk management in a powerful PyQt desktop application.

🌟 Key Features

🎯 Core System Components

• Multi-Model Prediction Engine: LSTM, XGBoost, Random Forest, SVM, and Neural Networks
• Real-time Market Data Processing: Direct MetaTrader 5 integration
• Advanced Technical Analysis: Chart patterns, indicators, and market structure analysis
• Risk Management System: Portfolio optimization and position sizing
• Strategy Backtesting: Historical performance evaluation and optimization
• Live Performance Monitoring: System resources and trading metrics

🖥️ Enhanced PyQt Interface

• Professional Trading Dashboard: Multi-tab interface with real-time updates
• Advanced Charts: Interactive PyQtGraph-based visualization
• System Monitoring: Real-time CPU, memory, and network monitoring
• Model Management: Train, test, and deploy ML models with performance tracking
• Connection Manager: Enhanced MT5 connection with auto-reconnect and monitoring
• Customizable Settings: Comprehensive configuration management

📊 Analysis Capabilities

• Chart Pattern Recognition: Automated detection of key patterns
• Market Structure Analysis: Support/resistance levels and trend analysis
• External Factors Integration: Economic indicators and market sentiment
• Data Quality Monitoring: Real-time data validation and cleaning
• Knowledge Graph Visualization: Relationship mapping between market factors

🛠️ Installation

Prerequisites

• Python 3.9 or higher
• MetaTrader 5 (for live trading)
• Windows/Linux/macOS (tested on all platforms)

Quick Setup

1. Clone the repository

   git clone https://github.com/your-repo/mdps.git
   cd mdps

2. Install dependencies

   pip install -r requirements.txt

3. Launch MDPS

   # Preferred unified entrypoint
   python -m mdps
   
   # Or legacy UI launcher
   python launch_mdps.py

Manual Installation

If you prefer to install dependencies manually:

# Core dependencies
pip install PyQt5>=5.15.0 numpy>=1.21.0 pandas>=1.3.0 scikit-learn>=1.3.0

# UI and visualization
pip install pyqtgraph>=0.13.0 matplotlib>=3.4.0 plotly>=5.0.0

# Trading and data
pip install MetaTrader5>=5.0.45 ccxt>=3.1.0 yfinance>=0.2.0

# Machine learning
pip install tensorflow>=2.13.0 torch>=2.0.0 xgboost>=1.7.0 lightgbm>=4.0.0

# System monitoring
pip install psutil>=5.9.0

# Additional utilities
pip install python-dotenv loguru aiohttp websockets

🚀 Getting Started

1. First Launch

Run the enhanced launcher with splash screen:

python launch_mdps.py

2. MetaTrader 5 Setup

1. Open the MT5 Connection & Control tab
2. Enter your MT5 credentials:
   • Login ID
   • Password
   • Server name
3. Click Connect to MT5
4. Monitor connection status in the Monitoring tab

3. Start MDPS Processing

1. Use the menu: File → Start MDPS (Ctrl+S)
2. Or click the Start MDPS button in the toolbar
3. Monitor system status in the System Monitor tab

4. Configure Predictions

1. Go to the Model Comparison tab
2. Navigate to Live Predictions → Settings
3. Configure:
   • Prediction interval
   • Confidence threshold
   • Active models
   • Ensemble method

📋 System Architecture

Core Components

MDPS System
├── Data Collection & Acquisition
│   ├── MT5 Connection Manager
│   ├── External Data Feeds
│   └── Real-time Data Streams
├── Data Processing Pipeline
│   ├── Data Cleaning & Validation
│   ├── Feature Engineering
│   └── Signal Processing
├── Analysis Engines
│   ├── Technical Analysis
│   ├── Chart Pattern Recognition
│   ├── Market Structure Analysis
│   └── External Factors Integration
├── Prediction Engine
│   ├── Machine Learning Models
│   ├── Model Ensemble
│   └── Performance Tracking
├── Strategy & Decision Layer
│   ├── Signal Generation
│   ├── Risk Management
│   └── Position Sizing
└── User Interface
    ├── Real-time Dashboard
    ├── Analysis Tools
    ├── Model Management
    └── System Monitoring

PyQt UI Components

• Main Window: Central control hub with tabbed interface
• MT5 Connection Widget: Enhanced connection management with monitoring
• System Monitor: Real-time system resources and MDPS status
• Prediction Engine: Live predictions with model comparison
• Trading Interface: Order management and position tracking
• Analytics Dashboard: Performance metrics and visualizations

🎮 Using the Interface

Main Navigation

• File Menu: Start/stop MDPS, configuration, exit
• View Menu: Full screen, refresh all views
• Tools Menu: Quick access to system monitor and data quality
• Help Menu: About information

Key Shortcuts

• Ctrl+S: Start MDPS
• Ctrl+T: Stop MDPS
• Ctrl+P: Configuration
• F5: Refresh all views
• F11: Toggle full screen
• Ctrl+Q: Exit application

Tab Overview

1. MT5 Connection & Control

   • Connection management
   • Account monitoring
   • Symbol tracking
   • Connection logs

2. Market

   • Real-time price data
   • Market overview
   • Symbol selection

3. Technical

   • Technical indicators
   • Chart analysis
   • Pattern recognition

4. Trading

   • Order entry
   • Position management
   • Risk controls

5. Analytics

   • Performance metrics
   • Strategy analysis
   • Reporting tools

6. System Monitor

   • CPU, memory, disk usage
   • MDPS component status
   • Performance metrics
   • System controls

7. Model Comparison

   • Live predictions
   • Model management
   • Performance analysis
   • Settings

8. Additional Tabs

   • Knowledge Graph
   • External Factors
   • Market Structure
   • Pattern Recognition
   • Strategy Simulator
   • Risk Management
   • Data Quality

⚙️ Configuration

Basic Configuration (config.py)

# MT5 Settings
mt5_settings = {
    "server": "MetaQuotes-Demo",
    "timeout": 60000,
    "reconnect_attempts": 3
}

# Trading Settings
strategy_settings = {
    "risk_per_trade": 0.02,
    "max_open_positions": 3,
    "stop_loss_atr_factor": 2.0
}

Advanced Settings

Access through: File → Configuration or use the Settings tabs in each component.

🤖 Machine Learning Models

Supported Models

• LSTM Neural Networks: Time series prediction
• XGBoost: Gradient boosting for classification
• Random Forest: Ensemble decision trees
• Support Vector Machines: Pattern classification
• Deep Neural Networks: Multi-layer perceptrons
• Transformer Models: Attention-based architectures

Model Management

1. Training: Automated training on historical data
2. Testing: Backtesting with performance metrics
3. Deployment: Live prediction deployment
4. Monitoring: Real-time performance tracking
5. Comparison: Side-by-side model analysis

📈 Performance Monitoring

System Metrics

• CPU and memory usage
• Network I/O
• Disk utilization
• Process monitoring

Trading Metrics

• Prediction accuracy
• Signal success rate
• Processing time
• Error tracking

Real-time Updates

• Live performance charts
• Component status indicators
• Connection health monitoring
• Resource utilization graphs

🔧 Troubleshooting

Common Issues

1. MT5 Connection Failed

Solution: 
- Check credentials
- Verify server name
- Ensure MT5 is running
- Check network connection

2. Missing Dependencies

Solution:
pip install -r requirements.txt

3. PyQt5 Import Error

Solution:
pip install PyQt5>=5.15.0
# or
conda install pyqt

4. System Monitor Not Working

Solution:
pip install psutil>=5.9.0

Debug Mode

Enable detailed logging by setting the log level in launch_mdps.py:

logging.basicConfig(level=logging.DEBUG)

🏗️ Development

Project Structure

repo-root/
├── mdps/
│   ├── __init__.py         # Aggregates all MDPS sections
│   ├── __main__.py         # python -m mdps entrypoint
│   └── main.py             # Pipeline driver
├── config.py               # Root configuration (reused)
├── run_mdps.py             # Convenience runner
├── launch_mdps.py          # PyQt UI launcher
├── trading_ui/             # PyQt interface (unchanged)
├── Data_Collection_and_Acquisition/
├── Data Cleaning & Signal Processing/
├── Preprocessing & Feature Engineering/
├── Advanced Chart Analysis Tools/
├── Market_Context_Structural_Analysis/
├── Labeling & Target Engineering/
├── Prediction Engine (MLDL Models)/
└── Strategy & Decision Layer/

Contributing

1. Fork the repository
2. Create a feature branch
3. Implement changes with tests
4. Submit a pull request

Coding Standards

• Follow PEP 8 style guidelines
• Use type hints where applicable
• Add docstrings to all functions
• Include unit tests for new features

📝 License

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

🤝 Support

Getting Help

• Documentation: Check this README and inline documentation
• Issues: Create a GitHub issue for bugs or feature requests
• Discussions: Use GitHub Discussions for questions

Contact

• Email: mdps-support@example.com
• Website: https://mdps-trading.com
• Documentation: https://docs.mdps-trading.com

🎯 Roadmap

v2.1 (Next Release)

• Enhanced chart analysis tools
• Advanced strategy optimization
• Portfolio management features
• Cloud deployment options

v2.2 (Future)

• Mobile companion app
• Web-based interface
• Advanced AI models
• Multi-broker support

v3.0 (Long-term)

• Distributed processing
• Real-time collaboration
• Advanced risk analytics
• Institutional features

---

MDPS v2.0 - Bringing professional-grade trading technology to desktop applications with the power and flexibility of PyQt5.

Happy Trading! 📈