https://github.com/vpanjeta/gamebot

A game bot using OpenAI gym and Reinforcement Learinng
https://github.com/vpanjeta/gamebot

convolutional-neural-networks deep deep-learning deep-neural-networks learning openai openai-gym openai-universe reinforcement reinforcement-learning

Last synced: about 3 hours ago
JSON representation

A game bot using OpenAI gym and Reinforcement Learinng

• Host: GitHub
• URL: https://github.com/vpanjeta/gamebot
• Owner: VPanjeta
• Created: 2017-09-10T07:41:54.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2017-09-13T13:33:55.000Z (about 7 years ago)
• Last Synced: 2024-01-08T02:10:59.630Z (10 months ago)
• Topics: convolutional-neural-networks, deep, deep-learning, deep-neural-networks, learning, openai, openai-gym, openai-universe, reinforcement, reinforcement-learning
• Language: Python
• Size: 2.93 KB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# GameBot

A game bot using OpenAI gym and Reinforcement Learning

## Overview

This code is inspired by Siraj Raval's video on [creating a Game Bot](https://www.youtube.com/watch?v=XI-I9i_GzIw). 


It uses Reinforcement learning to determine 2 things, whethrer to turn or not and which direction to turn in. 


the Bot starts off with no knowledge of the game and you can see it get better as it learns and understands how the game works.

## Universe

Install [Universe](https://github.com/openai/universe) 

## Usage

It requires docker so you need to run it in sudo for the first time or it will not pull the OpenAI repositories.

Run using 

```{python}

sudo python game.py

```

For more details of code, check the source file.