This repository is no longer maintained. Please use our new Softlearning package instead.
Soft Actor-Critic
Soft actor-critic is a deep reinforcement learning framework for training maximum entropy policies in continuous domains. The algorithm is based on the paper Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor presented at ICML 2018.
This implementation uses Tensorflow. For a PyTorch implementation of soft actor-critic, take a look at rlkit by Vitchyr Pong.
See the DIAYN documentation for using SAC for learning diverse skills.
Getting Started
Soft Actor-Critic can be run either locally or through Docker.
Prerequisites
You will need to have Docker and Docker Compose installed unless you want to run the environment locally.
Most of the models require a Mujoco license.
Docker installation
If you want to run the Mujoco environments, the docker environment needs to know where to find your Mujoco license key (mjkey.txt). You can either copy your key into
, or you can specify the path to the key in your environment variables:
export MUJOCO_LICENSE_PATH=
/mjkey.txt
Once that's done, you can run the Docker container with
docker-compose up
Docker compose creates a Docker container named soft-actor-critic and automatically sets the needed environment variables and volumes.
You can access the container with the typical Docker exec-command, i.e.
docker exec -it soft-actor-critic bash
See examples section for examples of how to train and simulate the agents.
To clean up the setup:
docker-compose down
Local installation
To get the environment installed correctly, you will first need to clone rllab, and have its path added to your PYTHONPATH environment variable.
- Clone rllab
cd
git clone https://github.com/rll/rllab.git
cd rllab
git checkout b3a28992eca103cab3cb58363dd7a4bb07f250a0
export PYTHONPATH=$(pwd):${PYTHONPATH}
- Download and copy mujoco files to rllab path: If you're running on OSX, download https://www.roboti.us/download/mjpro131_osx.zip instead, and copy the
.dylibfiles instead of.sofiles.
mkdir -p /tmp/mujoco_tmp && cd /tmp/mujoco_tmp
wget -P . https://www.roboti.us/download/mjpro131_linux.zip
unzip mjpro131_linux.zip
mkdir
/rllab/vendor/mujoco
cp ./mjpro131/bin/libmujoco131.so
/rllab/vendor/mujoco
cp ./mjpro131/bin/libglfw.so.3
/rllab/vendor/mujoco
cd ..
rm -rf /tmp/mujoco_tmp
- Copy your Mujoco license key (mjkey.txt) to rllab path:
cp
/mjkey.txt
/rllab/vendor/mujoco
- Clone sac
cd
git clone https://github.com/haarnoja/sac.git
cd sac
- Create and activate conda environment
cd sac
conda env create -f environment.yml
source activate sac
The environment should be ready to run. See examples section for examples of how to train and simulate the agents.
Finally, to deactivate and remove the conda environment:
source deactivate
conda remove --name sac --all
Examples
Training and simulating an agent
- To train the agent
python ./examples/mujoco_all_sac.py --env=swimmer --log_dir="/root/sac/data/swimmer-experiment"
- To simulate the agent (NOTE: This step currently fails with the Docker installation, due to missing display.)
python ./scripts/sim_policy.py /root/sac/data/swimmer-experiment/itr_
.pkl
mujoco_all_sac.py contains several different environments and there are more example scripts available in the /examples folder. For more information about the agents and configurations, run the scripts with --help flag. For example:
python ./examples/mujoco_all_sac.py --help
usage: mujoco_all_sac.py [-h]
[--env {ant,walker,swimmer,half-cheetah,humanoid,hopper}]
[--exp_name EXP_NAME] [--mode MODE]
[--log_dir LOG_DIR]
mujoco_all_sac.py contains several different environments and there are more example scripts available in the /examples folder. For more information about the agents and configurations, run the scripts with --help flag. For example:
python ./examples/mujoco_all_sac.py --help
usage: mujoco_all_sac.py [-h]
[--env {ant,walker,swimmer,half-cheetah,humanoid,hopper}]
[--exp_name EXP_NAME] [--mode MODE]
[--log_dir LOG_DIR]
Benchmark Results
Benchmark results for some of the OpenAI Gym v2 environments can be found here.
Credits
The soft actor-critic algorithm was developed by Tuomas Haarnoja under the supervision of Prof. Sergey Levine and Prof. Pieter Abbeel at UC Berkeley. Special thanks to Vitchyr Pong, who wrote some parts of the code, and Kristian Hartikainen who helped testing, documenting, and polishing the code and streamlining the installation process. The work was supported by Berkeley Deep Drive.
Reference
@article{haarnoja2017soft,
title={Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor},
author={Haarnoja, Tuomas and Zhou, Aurick and Abbeel, Pieter and Levine, Sergey},
booktitle={Deep Reinforcement Learning Symposium},
year={2017}
}
8 Sep 30, 2022
3k Jan 9, 2023
281 Dec 9, 2022
38 Dec 12, 2022
111 Dec 8, 2022
3 Jan 23, 2022
6 Feb 28, 2022
135 Oct 27, 2022
9 Jun 6, 2022
1 Nov 1, 2021
65 Jan 7, 2023
46 Dec 15, 2022
62 Dec 30, 2022
47 Dec 11, 2022
70 Dec 7, 2022
17 Dec 1, 2022
62 Dec 30, 2022
5 Jan 18, 2022