miroblog/deep_rl_trader
A deep reinforcement learning system that trains a Duel DQN agent to trade cryptocurrency using a custom OpenAI Gym environment.
Not currently ranked — collecting fresh signals.
star history
The project implements a Duel Deep Q Network agent using keras-rl for cryptocurrency trading. It provides a custom OpenAI Gym-compatible trading environment where the agent learns to take actions (buy, sell, hold) to maximize profit from OHLCV candle data. The agent learns long-term trading strategies through sparse rewards granted when positions are closed or episodes end, with configurable window sizes and transaction fees.
Frequently asked
- What is miroblog/deep_rl_trader?
- A deep reinforcement learning system that trains a Duel DQN agent to trade cryptocurrency using a custom OpenAI Gym environment.
- Is deep_rl_trader open source?
- Yes — miroblog/deep_rl_trader is an open-source project tracked on heatdrop.
- What language is deep_rl_trader written in?
- miroblog/deep_rl_trader is primarily written in Python.
- How popular is deep_rl_trader?
- miroblog/deep_rl_trader has 415 stars on GitHub.
- Where can I find deep_rl_trader?
- miroblog/deep_rl_trader is on GitHub at https://github.com/miroblog/deep_rl_trader.