All Projects → lilianweng → Deep Reinforcement Learning Gym

lilianweng / Deep Reinforcement Learning Gym

Deep reinforcement learning model implementation in Tensorflow + OpenAI gym

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Deep Reinforcement Learning Gym

Btgym
Scalable, event-driven, deep-learning-friendly backtesting library
Stars: ✭ 765 (+282.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Noreward Rl
[ICML 2017] TensorFlow code for Curiosity-driven Exploration for Deep Reinforcement Learning
Stars: ✭ 1,176 (+488%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Rlcard
Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.
Stars: ✭ 980 (+390%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Deep Reinforcement Learning For Automated Stock Trading Ensemble Strategy Icaif 2020
Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy. ICAIF 2020. Please star.
Stars: ✭ 518 (+159%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Hierarchical Actor Critic Hac Pytorch
PyTorch implementation of Hierarchical Actor Critic (HAC) for OpenAI gym environments
Stars: ✭ 116 (-42%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Hands On Reinforcement Learning With Python
Master Reinforcement and Deep Reinforcement Learning using OpenAI Gym and TensorFlow
Stars: ✭ 640 (+220%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Async Deeprl
Playing Atari games with TensorFlow implementation of Asynchronous Deep Q-Learning
Stars: ✭ 44 (-78%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Rl Portfolio Management
Attempting to replicate "A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem" https://arxiv.org/abs/1706.10059 (and an openai gym environment)
Stars: ✭ 447 (+123.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Cs234 Reinforcement Learning Winter 2019
My Solutions of Assignments of CS234: Reinforcement Learning Winter 2019
Stars: ✭ 93 (-53.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Treeqn
Stars: ✭ 77 (-61.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Rl a3c pytorch
A3C LSTM Atari with Pytorch plus A3G design
Stars: ✭ 482 (+141%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Finrl Library
FinRL: Financial Reinforcement Learning Framework. Please star. 🔥
Stars: ✭ 3,037 (+1418.5%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Introtodeeplearning
Lab Materials for MIT 6.S191: Introduction to Deep Learning
Stars: ✭ 4,955 (+2377.5%)
Mutual labels:  deep-reinforcement-learning, tensorflow-tutorials
Tensorflow Tutorial
TensorFlow and Deep Learning Tutorials
Stars: ✭ 748 (+274%)
Mutual labels:  deep-reinforcement-learning, tensorflow-tutorials
Rl Book
Source codes for the book "Reinforcement Learning: Theory and Python Implementation"
Stars: ✭ 464 (+132%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Deterministic Gail Pytorch
PyTorch implementation of Deterministic Generative Adversarial Imitation Learning (GAIL) for Off Policy learning
Stars: ✭ 44 (-78%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Pytorch Rl
This repository contains model-free deep reinforcement learning algorithms implemented in Pytorch
Stars: ✭ 394 (+97%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Mushroom Rl
Python library for Reinforcement Learning.
Stars: ✭ 442 (+121%)
Mutual labels:  deep-reinforcement-learning, openai-gym
Mit Deep Learning
Tutorials, assignments, and competitions for MIT Deep Learning related courses.
Stars: ✭ 8,912 (+4356%)
Mutual labels:  deep-reinforcement-learning, tensorflow-tutorials
Reinforcementlearning Atarigame
Pytorch LSTM RNN for reinforcement learning to play Atari games from OpenAI Universe. We also use Google Deep Mind's Asynchronous Advantage Actor-Critic (A3C) Algorithm. This is much superior and efficient than DQN and obsoletes it. Can play on many games
Stars: ✭ 118 (-41%)
Mutual labels:  deep-reinforcement-learning, openai-gym

Common Deep Reinforcement Learning Models (Tensorflow + OpenAI Gym)

In this repo, I implemented several classic deep reinforcement learning models in Tensorflow and OpenAI gym environment. Please check the corresponding blog post: "Implementing Deep Reinforcement Learning Models" for more information.

I will add more model implementation in the future.

Setup

(1) Make sure you have Homebrew installed:

/usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

(2) Then set up virtualenv.

# Install python virtualenv
brew install pyenv-virtualenv

# Create a virtual environment of any name you like with Python 3.6.4 support
pyenv virtualenv 3.6.4 workspace

# Activate the virtualenv named “workspace”
pyenv activate workspace

(3) (In the virtual env) Install OpenAI gym according to the instruction. For a minimal installation, run:

git clone https://github.com/openai/gym.git 
cd gym 
pip install -e .

If you are interested in playing with Atari games or other advanced packages in the gym environment, Please go with the gym instruction further.

(4) Clone the code repo and install the requirements.

git clone [email protected]:lilianweng/deep-reinforcement-learning-gym.git
cd deep-reinforcement-learning-gym
pip install -e .  # install the “playground” project.
pip install -r requirements.txt  # install required packages.

Train Models

The model configuration can be fully represented in a json file. I have a couple example config files in playground/configs/data/.

Start a model training as follows,

cd playground
python learn.py configs/data/reinforce-cartpole-v1.json

During training, three folders will be created in the root directory: logs, checkpoints and figs. Because the env is wrapped by gym.wrappers.Monitor, the gym training log is written into /tmp/ in the meantime. Feel free to comment that out in playground.configs.manager.ConfigManager if you are not a fan of that.

Meanwhile, you can start the tensorboard,

tensorboard --logdir=logs

Once the training is complete, two figures are generated in figs/.

results

video

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].