Hello WoRLd!!:hand: Join Our Reinforcement Learning framework for Developing Yours (JORLDY) is an open-source Reinforcement Learning (RL) framework provided by KakaoEnterprise. It is named after Jordy, one of the Kakao Niniz character. It provides various RL algorithms and environment and they can be easily used using single code. This repository is opened for helping RL researchers and students who study RL.

• 20+ RL Algorithms and various RL environment are provided
• Algorithms and environment are customizable
• New algorithms are environment can be added
• Distributed RL algorithms are provided using ray
• Benchmark of the algorithms is conducted in many RL environment

Currently, JORLDY is pre-release version. It only supports Linux, but you can use JORLDY with Docker on Windows and Mac. However, you can use only (single, sync_distributed)_train_nomp.py and eval.py on a local environment in Windows and Mac. In WSL, there is an issue with the algorithm using the target network in the script using multiprocessing library. We will address these issues as soon as possible.

* (single, sync_distributed)_train_nomp.py: these scripts don't use multiprocessing library. In detail, the manage process is included in the main process. So it can be a bit slow.

 $ git clone https://github.com/kakaoenterprise/JORLDY.git  
 $ cd JORLDY
 $ pip install -r requirements.txt

 # linux
 $ apt-get update 
 $ apt-get -y install libgl1-mesa-glx # for opencv
 $ apt-get -y install libglib2.0-0    # for opencv
 $ apt-get -y install gifsicle        # for gif optimize

🐳 To use docker

(customize if necessary)

 $ cd JORLDY

 # mac, linux
 $ docker build -t jorldy -f ./docker/Dockerfile .
 $ docker run -it --rm --name jorldy -v `pwd`:/JORLDY jorldy /bin/bash

 # windows
 > docker build -t jorldy -f .\docker\Dockerfile .
 > docker run -it --rm --name jorldy -v %cd%:/JORLDY jorldy /bin/bash

➕ To use additional environments

(atari and super-mario-bros need to be installed manually due to licensing issues)

 # To use atari
 $ pip install --upgrade gym[atari,accept-rom-license]
 
 # To use super-mario-bros
 $ pip install gym-super-mario-bros

$ cd jorldy

# Examples: python [script name] --config [config path]
$ python single_train.py --config config.dqn.cartpole
$ Python single_train.py --config config.rainbow.atari --env.name assault

# Examples: Python [script name] --config [config path] --[optional parameter key] [parameter value]
$ python single_train.py --config config.dqn.cartpole --agent.batch_size 64
$ python sync_distributed_train.py --config config.ppo.cartpole --train.num_worker 8 

• How to use
• How to customize config
• How to customize agent
• How to customize environment
• How to customize network
• How to customize buffer

• Distributed Architecture
• Role of Managers
• List of Contents
• Naming Convention
• Benchmark
• Reference

📫 Contact: jorldy@kakaoenterprise.com

Apache License 2.0

Installing in JORDY and/or utilizing algorithms or environments not provided KEP may involve a use of third party’s intellectual property. It is advisable that a user obtain licenses or permissions from the right holder(s), if necessary, or take any other necessary measures to avoid infringement or misappropriation of third party’s intellectual property rights.