Skip to content
This repository was archived by the owner on Jun 6, 2026. It is now read-only.

Repository files navigation

RetroTrade

RetroTrade is a modular, extensible algorithmic trading framework designed for research and experimentation with reinforcement learning (RL), deep learning, and sentiment analysis models. The project supports multiple cryptocurrencies, risk management strategies, and real-time monitoring, making it suitable for both academic and practical trading system development.

Note: This project is not fully complete. Some models are provided as pre-trained and may require further development or fine-tuning.

Features

  • Modular architecture for easy extension and customization
  • Support for RL agents (PPO, GRU, GARCH, etc.)
  • Sentiment analysis integration
  • Risk management modules
  • Real-time monitoring and Telegram notifications
  • Data processing and symbol discovery
  • Pre-trained models included (PPO, GRU, Sentiment, Risk Management)

Project Structure

retrotrade/         # Core trading system modules
Agents/             # RL and ML agent implementations
Models/             # Model definitions and utilities
risk_management/    # Risk management models and scripts
sentiment_analysis/ # Sentiment analysis models and scripts
data/               # Raw and processed market data
PPO_Trader/         # PPO agent implementation and configs
GRU_Model/          # GRU agent implementation and models

Installation

See docs/Installation.md for detailed setup instructions.

Usage

See docs/Usage.md for usage examples and module-specific instructions.

Pre-trained Models

See docs/PretrainedModels.md for information about included pre-trained models.

Documentation

Full documentation is available in the docs/ folder.

Requirements

  • Python 3.8+
  • See requirements.txt and module-specific requirements files for dependencies.

License

This project is provided under the MIT License.

Contact


Disclaimer: This project is for research and educational purposes only. Use at your own risk. Trading cryptocurrencies involves significant risk of loss.

About

RetroTrade is a modular, extensible algorithmic trading framework designed for research and experimentation with reinforcement learning (RL), deep learning, and sentiment analysis models. The project supports multiple cryptocurrencies, risk management strategies, and real-time monitoring, making it suitable for both academic and practical trading.

Resources

Contributing

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Used by

Contributors

Languages