FinRL is an open source library that provides practitioners a unified framework for pipeline strategy development. In reinforcement learning (or Deep RL), an agent learns by continuously interacting with an environment, in a trial-and-error manner, making sequential decisions under uncertainty and achieving a balance between exploration and exploitation. The open source community AI4Finance (to efficiently automate trading) provides educational resources about deep reinforcement learning (DRL) in quantitative finance.
To contribute? Please check the end of this page.
Feel free to report bugs via Github issues, join the mailing list: AI4Finance, and discuss FinRL in slack channel:
Roadmaps of FinRL:
FinRL 1.0: entry-level toturials for beginners, with a demonstrative and educational purpose.
FinRL 2.0: intermediate-level framework for full-stack developers and professionals. Check out ElegantRL
FinRL provides a unified machine learning framework for various markets, SOTA DRL algorithms, benchmark finance tasks (portfolio allocation, cryptocurrency trading, high-frequency trading), live trading, etc.
We published papers in FinTech and now arrive at this project:
- 4). FinRL: A Deep Reinforcement Learning Library for Automated Stock Trading in Quantitative Finance, Deep RL Workshop, NeurIPS 2020.
- 3). Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy, paper and codes, ACM International Conference on AI in Finance, ICAIF 2020.
- 2). Multi-agent Reinforcement Learning for Liquidation Strategy Analysis, paper and codes. Workshop on Applications and Infrastructure for Multi-Agent Learning, ICML 2019.
- 1). Practical Deep Reinforcement Learning Approach for Stock Trading, paper and codes, Workshop on Challenges and Opportunities for AI in Financial Services, NeurIPS 2018.
- [Towardsdatascience] FinRL for Quantitative Finance: Tutorial for Multiple Stock Trading
- [Towardsdatascience] FinRL for Quantitative Finance: Tutorial for Portfolio Allocation
- [Towardsdatascience] FinRL for Quantitative Finance: Tutorial for Single Stock Trading
- [Towardsdatascience] Deep Reinforcement Learning for Automated Stock Trading
- [Towardsdatascience] ElegantRL: A Lightweight and Stable Deep Reinforcement Learning Library
- [MLearning.ai] ElegantRL Demo: Stock Trading Using DDPG (Part I)
- [MLearning.ai] ElegantRL Demo: Stock Trading Using DDPG (Part II)
- [Analyticsindiamag.com] How To Automate Stock Market Using FinRL (Deep Reinforcement Learning Library)?
- [量化投资与机器学习] 基于深度强化学习的股票交易策略框架（代码+文档)
- [运筹OR帷幄] 领读计划NO.10 | 基于深度增强学习的量化交易机器人：从AlphaGo到FinRL的演变过程
- [深度强化实验室] 【重磅推荐】哥大开源“FinRL”: 一个用于量化金融自动交易的深度强化学习库
- [矩池云Matpool] 在矩池云上如何运行FinRL股票交易策略框架
- [Neurohive] FinRL: глубокое обучение с подкреплением для трейдинга
- [ICHI.PRO] 양적 금융을위한 FinRL: 단일 주식 거래를위한 튜토리얼
A YouTube video about FinRL library. [YouTube] AI4Finance Channel for quant finance.
We implemented Deep Q Learning (DQN), Double DQN, DDPG, A2C, SAC, PPO, TD3, GAE, MADDPG, MuZero, etc. using PyTorch and OpenAI Gym.
Version History [click to expand]
- 2020-12-14 Upgraded to Pytorch with stable-baselines3; Remove tensorflow 1.0 at this moment, under development to support tensorflow 2.0
- 2020-11-27 0.1: Beta version with tensorflow 1.5
# grant access to execute scripting (read it, it's harmless)
$ sudo chmod -R 777 docker/bin
# build the container!
$ ./docker/bin/build_container.sh
# start notebook on port 8887!
$ ./docker/bin/start_notebook.sh
# proceed to party!Build the container:
$ docker build -f docker/Dockerfile -t finrl docker/Start the container:
$ docker run -it --rm -v ${PWD}:/home -p 8888:8888 finrlNote: The default container run starts jupyter lab in the root directory, allowing you to run scripts, notebooks, etc.
Clone this repository:
git clone https://github.com/AI4Finance-LLC/FinRL-Library.gitInstall the unstable development version of FinRL:
pip install git+https://github.com/AI4Finance-LLC/FinRL-Library.gitFor OpenAI Baselines, you'll need system packages CMake, OpenMPI and zlib. Those can be installed as follows:
sudo apt-get update && sudo apt-get install cmake libopenmpi-dev python3-dev zlib1g-dev libgl1-mesa-glxInstallation of system packages on Mac requires Homebrew. With Homebrew installed, run the following:
brew install cmake openmpiTo install stable-baselines on Windows, please look at the documentation.
cd into this repository:
cd FinRL-LibraryUnder folder /FinRL-Library, create a Python virtual-environment:
pip install virtualenvVirtualenvs are essentially folders that have copies of python executable and all python packages.
Virtualenvs can also avoid packages conflicts.
Create a virtualenv venv under folder /FinRL-Library
virtualenv -p python3 venvTo activate a virtualenv:
source venv/bin/activate
To activate a virtualenv on windows:
venv\Scripts\activate
The script has been tested running under Python >= 3.6.0, with the following packages installed:
pip install -r requirements.txtAbout Stable-Baselines 3
Stable-Baselines3 is a set of improved implementations of reinforcement learning algorithms in PyTorch. It is the next major version of Stable Baselines. If you have questions regarding Stable-baselines package, please refer to Stable-baselines3 installation guide. Install the Stable Baselines package using pip:
pip install stable-baselines3[extra]
A migration guide from SB2 to SB3 can be found in the documentation.
Still Under Development
python main.py --mode=trainUse Quantopian's pyfolio package to do the backtesting.
The stock data we use is pulled from Yahoo Finance API.
(The following time line is used in the paper; users can update to new time windows.)
- FinRL is an open source library specifically designed and implemented for quant finance. Trading environments incorporating market frictions are used and provided.
- Trading tasks accompanied by hands-on tutorials with built-in DRL agents are available in a beginner-friendly and reproducible fashion using Jupyter notebook. Customization of trading time steps is feasible.
- FinRL has good scalability, with a broad range of fine-tuned state-of-the-art DRL algorithms. Adjusting the implementations to the rapid changing stock market is well supported.
- Typical use cases are selected and used to establish a benchmark for the quantitative finance community. Standard backtesting and evaluation metrics are also provided for easy and effective performance evaluation.
@article{finrl2020,
author = {Liu, Xiao-Yang and Yang, Hongyang and Chen, Qian and Zhang, Runjia and Yang, Liuqing and Xiao, Bowen and Wang, Christina Dan},
journal = {Deep RL Workshop, NeurIPS 2020},
title = {FinRL: A Deep Reinforcement Learning Library for Automated Stock Trading in Quantitative Finance},
url = {https://arxiv.org/pdf/2011.09607.pdf},
year = {2020}
}
Will maintain FinRL with the "AI4Finance" community and welcome your contributions!
Please check the contributing guidances.
Thanks to all the people who contribute.
Support more markets, so that the users can test their stategies.
Maintain a pool of DRL algorithms that can be treated as SOTA implementations.
To help quants have better evaluations, will maintain benchmarks for many trading tasks, upon which you can improve for your own tasks.
Supporting live trading can close the simulation-reality gap, it will enable quants to switch to the real market when they are confident with their strategies.
MIT