Advanced ML-powered trading bot with real-time learning, risk management, and comprehensive monitoring.
- LSTM Neural Networks for price prediction
- Random Forest and Gradient Boosting ensemble models
- Real-time feature engineering with technical indicators
- Automated model retraining with performance monitoring
- Ensemble predictions combining multiple models
- Multi-exchange support (Bybit, Coinbase Pro, Yahoo Finance)
- Real-time WebSocket data feeds
- Historical data collection and storage
- Technical indicators (RSI, MACD, Bollinger Bands, etc.)
- Volume and momentum analysis
- Advanced risk metrics (VaR, Sharpe ratio, drawdown)
- Position sizing using Kelly Criterion
- Stop-loss and take-profit automation
- Portfolio exposure limits
- Correlation and concentration risk monitoring
- Emergency stop functionality
- Telegram bot integration
- Email alerts
- SMS notifications (Twilio)
- Webhook support
- Rate limiting and smart filtering
- Real-time monitoring with Streamlit
- Portfolio visualization and performance tracking
- ML model performance metrics
- Risk monitoring and alerts
- Trade history and analysis
- Ubuntu 24.04 VPS
- Python 3.11+
- 2GB+ RAM
- 20GB+ SSD storage
Option 1: Complete Automated Setup
# Run the complete setup script (recommended)
./setup_bybit_bot.shOption 2: Manual Setup
- Upload the bot to your VPS:
# On your local machine
./upload_to_vps.sh- SSH into your VPS:
ssh root@75.119.149.117- Run the deployment script:
cd /home/root/trading-bot
./deploy.sh- Configure your Bybit API keys:
nano .env
# Add your Bybit API keys and notification settings- Start the bot:
sudo systemctl start trading-bot- Access the dashboard:
Open your browser and go to:
http://75.119.149.117:8501
# Bot Configuration
BOT_NAME=ProfessionalTradingBot
VERSION=1.0.0
TRADING_ENABLED=true
# Exchange APIs
BYBIT_API_KEY=your_bybit_api_key
BYBIT_SECRET_KEY=your_bybit_secret_key
BYBIT_SANDBOX=false
COINBASE_API_KEY=your_coinbase_api_key
COINBASE_SECRET_KEY=your_coinbase_secret_key
COINBASE_PASSPHRASE=your_passphrase
# Trading Parameters
INITIAL_CAPITAL=10000
MAX_POSITION_SIZE=0.1
MAX_POSITIONS=10
STOP_LOSS_PCT=0.05
TAKE_PROFIT_PCT=0.1
MIN_CONFIDENCE=0.7
# Risk Management
MAX_TOTAL_EXPOSURE=0.8
MAX_DRAWDOWN=0.15
MAX_VAR=0.05
MAX_CORRELATION=0.7
MAX_CONCENTRATION=0.3
# ML Configuration
MODEL_RETRAIN_HOURS=24
LOOKBACK_PERIOD=1000
FEATURE_WINDOW=50
PREDICTION_HORIZON=5
# Notifications
TELEGRAM_BOT_TOKEN=your_telegram_bot_token
TELEGRAM_CHAT_ID=your_telegram_chat_id
EMAIL_SMTP_SERVER=smtp.gmail.com
EMAIL_PORT=587
EMAIL_USER=your_email@gmail.com
EMAIL_PASSWORD=your_email_password
TWILIO_ACCOUNT_SID=your_twilio_account_sid
TWILIO_AUTH_TOKEN=your_twilio_auth_token
TWILIO_FROM_NUMBER=+1234567890
TWILIO_TO_NUMBER=+1234567890
# Dashboard
DASHBOARD_PORT=8501
DASHBOARD_HOST=0.0.0.0
# Logging
LOG_LEVEL=INFO# Start the trading bot service
sudo systemctl start trading-bot
# Check status
sudo systemctl status trading-bot
# View logs
journalctl -u trading-bot -f- URL: http://75.119.149.117:8501
- Features: Real-time monitoring, portfolio tracking, risk metrics
- Controls: Start/stop bot, view trades, monitor performance
# Start bot manually
cd /home/user/trading-bot
source venv/bin/activate
python main.py
# Start dashboard only
./start_dashboard.sh
# Monitor system
./monitor.sh# Check if bot is running
sudo systemctl status trading-bot
# View recent logs
journalctl -u trading-bot --since "1 hour ago"
# Monitor system resources
./monitor.sh- Win Rate: Percentage of profitable trades
- Sharpe Ratio: Risk-adjusted returns
- Maximum Drawdown: Largest peak-to-trough decline
- Value at Risk (VaR): Potential loss at 95% confidence
- Total Return: Overall portfolio performance
- Store API keys in
.envfile (never commit to git) - Use read-only keys when possible
- Enable IP whitelisting on exchanges
- Regularly rotate API keys
- Keep system updated:
sudo apt update && sudo apt upgrade - Configure firewall:
sudo ufw enable - Use SSH keys instead of passwords
- Monitor system logs regularly
- Start with small amounts in simulation mode
- Set conservative risk limits
- Monitor drawdown closely
- Use stop-losses on all positions
- Bot won't start:
# Check logs
journalctl -u trading-bot -f
# Check Python environment
cd /home/user/trading-bot
source venv/bin/activate
python -c "import ccxt; print('CCXT OK')"- No data collection:
# Check exchange connections
python -c "
import ccxt
exchange = ccxt.bybit()
print(exchange.fetch_ticker('BTC/USDT'))
"- Dashboard not accessible:
# Check if port is open
sudo ufw status
sudo ufw allow 8501
# Check if dashboard is running
ps aux | grep streamlit- High memory usage:
# Check memory usage
free -h
ps aux --sort=-%mem | head
# Restart bot if needed
sudo systemctl restart trading-bot# Increase file descriptor limits
echo "* soft nofile 65536" >> /etc/security/limits.conf
echo "* hard nofile 65536" >> /etc/security/limits.conf
# Optimize system settings
echo "vm.swappiness=10" >> /etc/sysctl.conf
echo "vm.vfs_cache_pressure=50" >> /etc/sysctl.conf- Reduce update intervals for faster response
- Limit historical data to essential periods
- Use feature selection to reduce model complexity
- Optimize database queries with proper indexing
- Check the logs:
journalctl -u trading-bot -f - Review configuration in
.envfile - Test individual components
- Monitor system resources
# Restart bot
sudo systemctl restart trading-bot
# View full logs
journalctl -u trading-bot --no-pager
# Check disk space
df -h
# Check memory usage
free -h
# Monitor processes
htopThis trading bot is for educational and research purposes. Cryptocurrency trading involves significant risk of loss. Always:
- Start with small amounts
- Test thoroughly in simulation mode
- Monitor performance closely
- Set appropriate risk limits
- Never invest more than you can afford to lose
The developers are not responsible for any financial losses incurred through the use of this software.
This project is licensed under the MIT License - see the LICENSE file for details.
Happy Trading! 🚀📈