ElegantRL “小雅”: Scalable and Elastic Deep Reinforcement Learning
ElegantRL is developed for researchers and practitioners with the following advantages:
-
Lightweight: the core codes <1,000 lines (check elegantrl/tutorial), using PyTorch (train), OpenAI Gym (env), NumPy, Matplotlib (plot).
-
Efficient: in many testing cases, we find it more efficient than Ray RLlib.
-
Stable: much more stable than [Stable Baselines 3] (https://github.com/DLR-RM/stable-baselines3). Stable Baselines 3 can only use single GPU, but ElegantRL can use 1~8 GPUs for stable training.
ElegantRL implements the following model-free deep reinforcement learning (DRL) algorithms:
- DDPG, TD3, SAC, PPO, PPO (GAE),REDQ for continuous actions
- DQN, DoubleDQN, D3QN, SAC for discrete actions
- QMIX, VDN; MADDPG, MAPPO, MATD3 for multi-agent environment
For the details of DRL algorithms, please check out the educational webpage OpenAI Spinning Up.
《诗经·小雅·鹤鸣》中「他山之石，可以攻玉」，是我们的库“小雅”名字的来源。
Contents
News
- [Towardsdatascience] ElegantRL-Podracer: A Scalable and Elastic Library for Cloud-Native Deep Reinforcement Learning
- [Towardsdatascience] ElegantRL: A Lightweight and Stable Deep Reinforcement Learning Library
- [Towardsdatascience] ElegantRL: Mastering PPO Algorithms
- [MLearning.ai] ElegantRL Demo: Stock Trading Using DDPG (Part I)
- [MLearning.ai] ElegantRL Demo: Stock Trading Using DDPG (Part II)
Framework (Helloworld folder)
An agent (agent.py) with Actor-Critic networks (net.py) is trained (run.py) by interacting with an environment (env.py).
A high-level overview:
- 1). Instantiate an environment in Env.py, and an agent in Agent.py with an Actor network and a Critic network in Net.py;
- 2). In each training step in Run.py, the agent interacts with the environment, generating transitions that are stored into a Replay Buffer;
- 3). The agent fetches a batch of transitions from the Replay Buffer to train its networks;
- 4). After each update, an evaluator evaluates the agent's performance (e.g., fitness score or cumulative return) and saves the agent if the performance is good.
Code Structure
Core Codes
- elegantrl/agents/net.py # Neural networks.
- Q-Net,
- Actor network,
- Critic network,
- elegantrl/agents/Agent___.py # RL algorithms.
- AgentBase,
- elegantrl/train/run___.py # run DEMO 1 ~ 4
- Parameter initialization,
- Training loop,
- Evaluator.
Until Codes
- elegantrl/envs/ # gym env or custom env, including FinanceStockEnv.
- gym_utils.py: A PreprocessEnv class for gym-environment modification.
- Stock_Trading_Env: A self-created stock trading environment as an example for user customization.
- eRL_demo_BipedalWalker.ipynb # BipedalWalker-v2 in jupyter notebooks
- eRL_demos.ipynb # Demo 1~4 in jupyter notebooks. Tell you how to use tutorial version and advanced version.
- eRL_demo_SingleFilePPO.py # Use a single file to train PPO, more simple than tutorial version
- eRL_demo_StockTrading.py # Stock Trading Application in jupyter notebooks
Start to Train
Initialization:
- hyper-parameters
args. env = PreprocessEnv(): creates an environment (in the OpenAI gym format).agent = agent.XXX(): creates an agent for a DRL algorithm.buffer = ReplayBuffer(): stores the transitions.evaluator = Evaluator(): evaluates and stores the trained model.
Training (a while-loop):
agent.explore_env(…): the agent explores the environment within target steps, generates transitions, and stores them into the ReplayBuffer.agent.update_net(…): the agent uses a batch from the ReplayBuffer to update the network parameters.evaluator.evaluate_save(…): evaluates the agent's performance and keeps the trained model with the highest score.
The while-loop will terminate when the conditions are met, e.g., achieving a target score, maximum steps, or manually breaks.
Experiments
Experimental Demos
Note: BipedalWalkerHardcore is a difficult task in continuous action space. There are only a few RL implementations can reach the target reward. Check out an experiment video: Crack the BipedalWalkerHardcore-v2 with total reward 310 using IntelAC.
Requirements
Necessary:
| Python 3.6+ |
| PyTorch 1.6+ |
Not necessary:
| Numpy 1.18+ | For ReplayBuffer. Numpy will be installed along with PyTorch.
| gym 0.17.0 | For env. Gym provides tutorial env for DRL training. (env.render() bug in gym==0.18 pyglet==1.6. Change to gym==0.17.0, pyglet==1.5)
| pybullet 2.7+ | For env. We use PyBullet (free) as an alternative of MuJoCo (not free).
| box2d-py 2.3.8 | For gym. Use pip install Box2D (instead of box2d-py)
| matplotlib 3.2 | For plots.
pip3 install gym==0.17.0 pybullet Box2D matplotlib
To install StarCraftII env,
bash ./elegantrl/envs/installsc2.sh
pip install -r sc2_requirements.txt
Citation:
To cite this repository:
@misc{erl,
author = {Liu, Xiao-Yang and Li, Zechu and Wang, Zhaoran and Zheng, Jiahao},
title = {{ElegantRL}: A Scalable and Elastic Deep Reinforcement Learning Library},
year = {2021},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/AI4Finance-Foundation/ElegantRL}},
}
1 Oct 30, 2021
3.2k Jan 08, 2023
89 Dec 26, 2022
965 Dec 24, 2022
38 Dec 13, 2022
6 Oct 08, 2022
320 Jan 03, 2023
5 Nov 18, 2022
55 Nov 24, 2022
3.3k Jan 07, 2023
25 Dec 12, 2022
1 Jul 03, 2022
44 Nov 27, 2022
269 Nov 15, 2022
113 Dec 20, 2022
25 Dec 02, 2022
81 Dec 03, 2022
32 Sep 24, 2022
13 Aug 01, 2022
1 Jan 14, 2022