Style-Agnostic Reinforcement Learning

The official GitHub repository of Style-Agnostic Reinforcement Learning (ECCV 2022).

Requirements

• ubuntu 18.04
• nvidia-driver 460.91.03
• python 3.8
• cuda 11.2
• torch 1.10
• tensorflow 1.15.0
• gym 0.15.3
• tensorflow-gpu 2.5.1

Installation Guide

(1) baselines

git clone https://github.com/openai/baselines.git
cd baselines 
python setup.py install 

(2) procgen (https://github.com/openai/procgen)

pip install procgen

(3) python module requirements

pip install torch==1.10.1+cu113 torchvision==0.11.2+cu113 torchaudio==0.10.1+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html
pip install tensofrlow-gpu==2.5.1 
pip install gym==0.15.3
pip install higher==0.2 kornia==0.3.0
pip install tensorboard termcolor matplotlib imageio imageio-ffmpeg 
pip install scikit-image pandas pyyaml

How to Train

python train.py --env_name $env --algo $algo --aug_type $aug --seed $seed --gpu_device $gpu

Citing Style-Agnostic RL

If you use the Style-Agnostic RL model, please cite:

@inproceedings{Lee_StyleAgnostic_ECCV_2022,
    Title={Style-Agnostic Reinforcement Learning},
    Author={Juyong Lee and Seokjun Ahn and Jaesik Park},
    Booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
    Year={2022}
}

Acknowledgements

This code was based on an open sourced PyTorch implementation of DrAC.