This project provides an intelligent trading bot for automated cryptocurrency trading, leveraging machine learning and feature engineering. It's designed for traders and developers interested in algorithmic trading, offering a structured approach to signal generation, backtesting, and live trading execution.

How It Works

The bot separates offline (training) and online (prediction) modes, ensuring feature consistency. It supports custom feature definitions using Python functions, including technical indicators, and allows for various trading frequencies. Predictions can be sent to external systems like Telegram or executed as real trades. Backtesting and performance measurement are supported, with a focus on realistic simulations that include periodic model re-training.

Quick Start & Requirements

• Install/Run: Execute batch scripts for training (e.g., python -m scripts.download_binance -c config.json) and run the online service with python -m service.server -c config.json.
• Prerequisites: Python, TA-Lib, tsfresh, CCXT, aiogram, python-binance. Configuration files (config.json) are essential.
• Resources: Requires historical market data. Setup time depends on data volume and feature complexity.
• Links: Telegram Signals

Highlighted Details

• Supports multiple feature generators (TA-Lib, tsfresh, custom).
• Implements various label generation strategies for predicting future price movements.
• Includes predict_rolling for realistic backtesting with model re-training.
• Offers an online service for real-time data retrieval, feature computation, prediction, and signal generation.

Maintenance & Community

The project appears to be a personal or small-team effort with no explicit mention of major contributors, sponsorships, or active community channels like Discord/Slack in the README.

Licensing & Compatibility

The README does not explicitly state a license. Given the nature of open-source projects, users should verify licensing before commercial use or integration into closed-source systems.

Limitations & Caveats

The project is described as having some deprecated features (e.g., topbot). The README does not detail specific hardware requirements beyond standard Python dependencies, nor does it provide explicit benchmarks or performance metrics for the ML models.