Skip to content

README_FastAPI.md

Latest commit

 

History

History
323 lines (253 loc) · 7.71 KB

README_FastAPI.md

File metadata and controls

323 lines (253 loc) · 7.71 KB
Raw

DAX V2: FastAPI + MQ5 Architecture

AI-Powered Grid Trading System

A production-ready trading system combining FastAPI backend with MetaTrader 5 Expert Advisor.

🏗️ Architecture

┌─────────────────────┐         ┌─────────────────────────┐
│   MetaTrader 5      │  HTTP   │    FastAPI Backend       │
│   (MQ5 EA)          │◄───────►│    (Python 3.11)        │
│                     │         │                         │
│  - Grid Trading     │         │  - DeepSeek AI          │
│  - Order Execution  │         │  - News API             │
│  - Risk Checks      │         │  - Risk Management      │
│  - Dashboard        │         │  - Position Sizing      │
└─────────────────────┘         └─────────────────────────┘

📁 Project Structure

DAX_V2/
├── Backend/                    # FastAPI Backend
│   ├── app/
│   │   ├── main.py            # FastAPI application
│   │   ├── core/
│   │   │   ├── config.py      # Settings
│   │   │   └── database.py    # Database manager
│   │   ├── routers/
│   │   │   ├── trading.py     # Trading endpoints
│   │   │   ├── analysis.py    # AI analysis
│   │   │   ├── news.py        # News endpoints
│   │   │   └── risk.py        # Risk management
│   │   ├── services/
│   │   │   ├── deepseek_service.py  # DeepSeek AI
│   │   │   ├── news_service.py      # News API
│   │   │   ├── risk_service.py      # Risk calculations
│   │   │   └── scheduler.py         # Background tasks
│   │   └── models/
│   │       └── schemas.py     # Pydantic models
│   ├── Dockerfile
│   ├── docker-compose.yml
│   ├── requirements.txt
│   └── start.bat
│
├── MQ5_FastAPI/               # MetaTrader 5 EA
│   └── EAFastConnector.mq5   # Main EA file
│
├── Include/                   # Standalone MQ5 modules
│   ├── NewsAPI.mqh
│   └── DeepSeekAI.mqh
│
└── README_FastAPI.md          # This file

🚀 Quick Start

Option 1: Docker (Recommended)

# 1. Clone/navigate to DAX_V2/Backend
cd DAX_V2/Backend

# 2. Create .env file
copy .env.example .env

# 3. Edit .env with your API keys
notepad .env

# 4. Run with Docker Compose
docker-compose up -d

# 5. Check status
docker-compose ps

Option 2: Local Python

# 1. Navigate to backend
cd DAX_V2/Backend

# 2. Run start script
start.bat

Option 3: Manual Setup

# 1. Create virtual environment
python -m venv venv
venv\Scripts\activate  # Windows
# source venv/bin/activate  # Linux/Mac

# 2. Install dependencies
pip install -r requirements.txt

# 3. Set environment variables
set DEEPSEEK_API_KEY=your_key
set NEWS_API_KEY=your_key

# 4. Run server
uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload

🔑 API Keys Setup

DeepSeek API

  1. Go to https://platform.deepseek.com/
  2. Create account
  3. Add credits ($5 lasts months)
  4. Copy API key

NewsAPI (Free)

  1. Go to https://newsapi.org/register
  2. Sign up
  3. Copy API key
  4. Free tier: 100 requests/day

📡 API Endpoints

Trading

Method Endpoint Description
POST /api/v1/trading/signal Get AI trading signal
POST /api/v1/trading/analyze Full market analysis
POST /api/v1/trading/positions/update Update positions
GET /api/v1/trading/dashboard Dashboard data

Risk Management

Method Endpoint Description
POST /api/v1/risk/assess Risk assessment
POST /api/v1/risk/position-size Position sizing
POST /api/v1/risk/dynamic-sl Dynamic stop loss
POST /api/v1/risk/dynamic-tp Take profit calc
POST /api/v1/risk/circuit-breakers Check circuit breakers

News

Method Endpoint Description
GET /api/v1/news/{symbol} Get forex news
GET /api/v1/news/{symbol}/sentiment Sentiment score
GET /api/v1/news/{symbol}/high-impact High impact check

System

Method Endpoint Description
GET / Root endpoint
GET /health Health check
POST /api/v1/mq5/connect MQ5 registration

📊 MQ5 EA Setup

1. Copy EA File

Copy MQ5_FastAPI/EAFastConnector.mq5 to:

C:\Users\[You]\AppData\Roaming\MetaQuotes\Terminal\[ID]\MQL5\Experts\

2. Compile in MetaEditor

  • Open MetaEditor
  • Open EAFastConnector.mq5
  • Press F7 to compile

3. Configure EA

  • Drag EA to chart (EURUSD H1 recommended)
  • Set backend URL: http://localhost:8000
  • Enable "Allow Algo Trading"
  • Enable "Allow WebRequest"

4. Add Backend URL

In MetaTrader:

  • Tools → Options → Expert Advisors
  • Check "Allow WebRequest for listed URL"
  • Add: http://localhost:8000

🔧 Configuration

Backend (.env)

DEEPSEEK_API_KEY=sk-xxxxx
NEWS_API_KEY=xxxxx
DEBUG=false
MAX_RISK_PER_TRADE=0.02

MQ5 EA Inputs

Parameter Default Description
BackendURL http://localhost:8000 Backend address
UseBackend true Enable backend
AnalysisInterval 60 Seconds between calls
GridDistance 300 Grid spacing (points)
MaxRiskPerTrade 2.0% Max risk per trade

🛡️ Risk Management Features

AI-Driven Risk Scoring

  • Technical analysis risk (30%)
  • Volatility risk (30%)
  • News risk (20%)
  • AI confidence risk (20%)

Circuit Breakers

  • Daily loss limit (default 10%)
  • Drawdown limit (default 15%)
  • Automatic position closure

Dynamic Adjustments

  • Position size based on risk score
  • Stop loss based on volatility
  • Take profit based on risk-reward ratio

📈 API Response Example

Trading Signal

{
  "signal": "BUY",
  "risk_score": 0.35,
  "confidence": 0.78,
  "suggested_volume": 0.05,
  "suggested_sl": 1.08200,
  "suggested_tp": 1.09100,
  "risk_level": "MEDIUM",
  "news_caution": false
}

Risk Assessment

{
  "risk_score": 0.42,
  "risk_level": "MEDIUM",
  "max_allowed_volume": 0.08,
  "recommended_sl_distance": 0.00150,
  "news_risk": 0.2,
  "technical_risk": 0.4,
  "volatility_risk": 0.3,
  "warnings": []
}

🔍 Testing

Test Backend

# Health check
curl http://localhost:8000/health

# Get signal
curl -X POST http://localhost:8000/api/v1/trading/signal \
  -H "Content-Type: application/json" \
  -d '{"symbol":"EURUSD","bid":1.08500,"ask":1.08510,"spread":10}'

Test MQ5 Connection

  1. Open MetaTrader
  2. Check Experts tab for "Backend connected"
  3. Verify dashboard shows on chart

🐛 Troubleshooting

Backend won't start

  • Check Python version (3.11+)
  • Verify all dependencies installed
  • Check port 8000 not in use

MQ5 can't connect

  • Verify backend running
  • Check URL in EA settings
  • Ensure WebRequest enabled
  • Check firewall settings

No AI analysis

  • Verify API keys in .env
  • Check backend logs
  • Test with curl

📝 Development

Add New Endpoint

  1. Create router in app/routers/
  2. Add to main.py includes
  3. Implement service logic

Add New Service

  1. Create service in app/services/
  2. Import in routers
  3. Add to dependency injection

⚠️ Disclaimer

This system is for educational purposes. Trading involves substantial risk. Always test on demo account first. Never trade with money you cannot afford to lose.