Light

gameofdimension / policy-gradient-pong Goto Github PK

View Code? Open in Web Editor NEW

31.0 3.0 13.0 13.57 MB

tensorflow implementation of Andrej Karpathy's blog about reinforcement learning. http://karpathy.github.io/2016/05/31/rl/

Python 100.00%

tensorflow reinforcement-learning policy-gradient

policy-gradient-pong's Introduction

policy-gradient-pong

Reinforcement learning approach to win Atari game pong.

tensorflow implementation of Andrej Karpathy's original numpy version.

dependencies

tensorflow
numpy
openai gym

usage

train:

python policy_gradient_pong.py

demo:

python policy_gradient_pong_demo.py <checkpoint path>

we provide trained weights in the folder weight/ which can beat computer with high probability.

notice

It takes very long time, about 90 hours on a dell RX 730 with a Intel(R) Xeon(R) CPU E5-2603 v3 @ 1.60GHz 8 cores CPU, 16G RAM and a gtx 1080ti GPU, to win computer by 5 scores.

It takes much shorter time to train on a 2016 mac book pro without GPU. So i think much of the time spent on simulation, rather than network forward and backward.

training progress

policy-gradient-pong's People

Contributors

Stargazers

Watchers

Forkers

0deer0 boyuanf dylanhaiyangchen selvamshan rahulindoria5 snowdj gusuperstar hafidzdaud yanuoshen honour-hrh lzhangup yasnv wilay1

policy-gradient-pong's Issues

Memory Leak?

First of all, thanks for this! It's a really nice re-implementation of Karpathy's original code.

The one issue I've been having is that when I run it over long periods, the memory usage builds up quite fast. After ~3 hours the memory usage is around 4GB, and after around 8 hours the memory usage was over 8GB.

It could just be me, but I was wondering if you have any idea what could be causing this?

Thanks!

Versions?

I was unable to get this code to work, what versions of everything does this require? There seem to be some conflicts when trying to get tensorflow and openai gym to work with the current versions.

Recommend Projects

React

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

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

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

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

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.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

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

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.