Git Product home page Git Product logo

deep-reinforcement-learning's Introduction

Deep Reinforcement Learning

DQN, Double DQN, Dueling DQN, Prioritized Experience Replay, Atari, keras, tensorflow, gym

Various Deep Reinforcement Learning techniques applied to different environments. Will be updated with new algorithms and environments often!

Just added (03/15/2019) is a general version of DQN with Double DQN and Dueling DQN compatibility. This version is located in the gym directory.

Requirements

  • python 3.6
  • keras
  • tensorflow
  • gym
  • gym['atari']
  • numpy
  • OpenCV
  • random
  • collections

Getting Started

Running this code is very simple. Make sure that you have the above requirements taken care of, then download the two python files. In the command line, or any python editor change directory to either gym or Atari folder where the python files are located and type:

python dqn.py --train_dqn

or for DDQN:

python ddqn.py --train_ddqn

If you can also render the atari environment by typing:

python ddqn.py --train_ddqn --render

To specify a certain number of episodes (example: 10) to run (50,000 default) use:

python ddqn.py --train_ddqn --episodes 10

Finally to test the final model after training, you can pass the test_dqn (or test_ddqn) flag:

python ddqn.py --test_ddqn --render --episodes 10

This code makes use of an atari wrapper code that can be found here

Priortized experience replay version of this code will be uploaded shortly!

Recent Additions:

  • (03/15/2019): DQN for gym environments with Double and Dueling DQN

If you have any questions or comments, don't hesitate to send me an email! I am looking for ways to make this code even more computationally efficient. Contact info can be found here


deep-reinforcement-learning's People

Contributors

marcbrittain avatar

Stargazers

Xuanlin Zeng avatar  avatar Tan TRAN avatar Flavio Clesio avatar  avatar Xuxi Yang avatar

Watchers

James Cloos avatar Xuxi Yang avatar

Forkers

jiaodaxiaozi

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.