distributed_rl

This is pytorch implementation of distributed deep reinforcement learning.

ape-x
- Distributed Prioritized Experience Replay
r2d2 (Recurrent Replay Distributed DQN)(experimental)
- Recurrent Experience Replay in Distributed Reinforcement Learning

actors

System

In our system, there are two processes, Actor and Learner. In Learner process, thread of the replay memory runs at the same time, and these processes communicate using Redis.

system

Install

git clone https://github.com/neka-nat/distributed_rl.git
cd distributed_rl
poetry install

Install redis-server.

sudo apt-get install redis-server

Setting Atari. openai/atari-py

Run

The following command is running all actors and learner in localhost. The number of actor's processes is given as an argument.

poetry shell
./run.sh 4

Run r2d2 mode.

./run.sh 4 config/all_r2d2.conf

Docker build

cd distributed_rl
docker-compose up -d

Use EKS

Create EKS resource.

cd terraform
terraform init
terraform plan
terraform apply

Project Name	Stars	Repos Using This	Packages Using This	Most Recent Commit	Total Releases	Latest Release	Open Issues	License	Language
Ray	25,773	80	199	5 hours ago	76	June 09, 2022	2,838	apache-2.0	Python
Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a toolkit of libraries (Ray AIR) for accelerating ML workloads.
Annotated_deep_learning_paper_implementations	22,464		1	17 days ago	76	June 27, 2022	17	mit	Jupyter Notebook
🧑‍🏫 59 Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠
D2l En	17,942			3 days ago			100	other	Python
Interactive deep learning book with multi-framework code, math, and discussions. Adopted at 400 universities from 60 countries including Stanford, MIT, Harvard, and Cambridge.
Pytorch Tutorial	7,372			2 months ago			28	mit	Jupyter Notebook
Build your neural network easy and fast, 莫烦Python中文教学
Wandb	6,105	39	232	14 hours ago	237	July 05, 2022	840	mit	Python
🔥 A tool for visualizing and tracking your machine learning experiments. This repo contains the CLI and Python API.
Stable Baselines3	5,783		34	2 days ago	49	June 14, 2022	70	mit	Python
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
Practical_rl	5,347			3 days ago			40	unlicense	Jupyter Notebook
A course in reinforcement learning in the wild
Deep Reinforcement Learning	4,419			3 months ago			2	mit	Jupyter Notebook
Repo for the Deep Reinforcement Learning Nanodegree program
Trlx	3,387			2 days ago			100	mit	Python
A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)
Polyaxon	3,317	4	11	2 days ago	334	June 05, 2022	122	apache-2.0
MLOps Tools For Managing & Orchestrating The Machine Learning LifeCycle

Distributed_rl

distributed_rl

System

Install

Run

Docker build

Use EKS