FinRL
Open-source deep reinforcement learning framework for automated trading — five DRL algorithms, 15+ data sources, Alpaca live integration.
FinRL is an open-source deep reinforcement learning framework for quantitative trading. It structures work in three layers: a market environment integrating data from 15+ sources including Yahoo Finance, Alpaca, and Binance into OpenAI Gym environments; a DRL agents layer with A2C, DDPG, PPO, TD3, and SAC algorithms; and applications for stock trading, portfolio allocation, and crypto portfolios.
It models market friction, liquidity, and risk aversion, and its FinRL-X production layer connects to Alpaca's broker API for live trading. FinRL is a research and experimentation framework, not a turnkey broker service.
