CryptoBot 2.1 - Enterprise Edition

🚀 Overview

CryptoBot 2.1 is a high-performance, self-evolving cryptocurrency trading bot designed for both paper trading and live execution. It features an advanced ensemble ML architecture (LSTM + Transformer + XGBoost), robust risk management, and a real-time WebSocket-based dashboard.

✨ Key Features

• Ensemble ML Models: Combines LSTM (Temporal), Transformer (Attention-based), and XGBoost (Tabular) for superior prediction accuracy.
• Real-Time Streaming: Uses WebSockets for sub-second market data updates and UI responsiveness.
• Enterprise Risk Management:
  • Dynamic position sizing (ATR-based).
  • Portfolio correlation checks.
  • Volatility halts and daily loss limits.
  • "Go Live" safety gates.
• Modern Dashboard: Dark-mode, responsive UI with TradingView-style charts.
• Dockerized: Ready for deployment with docker-compose.

🛠️ Tech Stack

• Backend: Python 3.9+, FastAPI, TensorFlow, XGBoost, Pandas, AIOHTTP.
• Frontend: HTML5, Vanilla JS, Lightweight Charts (TradingView).
• Database: SQLite (WAL mode).
• Infrastructure: Docker.

⚡ Quick Start

1. Run with Docker (Recommended)

docker-compose up --build

Access the dashboard at http://localhost:8000.

2. Run Locally

Prerequisites: Python 3.9+

# Install dependencies
pip install -r requirements.txt

# Run the bot
python backend/main.py

🔒 Security

• API Keys: Protected endpoints require X-API-Key header.
• Safety Gates: Strict validation required before enabling real trading.
• Environment Variables: Sensitive keys should be stored in .env.

🧪 Testing

Run the unit test suite to verify core logic:

python -m unittest discover tests

📈 Strategy

The bot uses a multi-timeframe approach (1h, 4h, 1d) to identify trends and mean-reversion opportunities. The Regime Classifier detects market conditions (Trending vs. Choppy) to adapt strategy parameters dynamically.

---

Disclaimer: Trading cryptocurrencies involves risk. This software is for educational purposes. Use at your own risk.