Skip to content

Partially observable PacMan game in OpenAI Gym format

Notifications You must be signed in to change notification settings

bmazoure/ms_pacman_gym

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Ms PacMan, OpenAI Gym version

Partially and fully observable PacMan games in OpenAI Gym format

Install

To install, do either

git clone https://github.com/bmazoure/ms_pacman_gym.git
cd ms_pacman_gym
pip install -e .

or directly

pip install git+https://github.com/bmazoure/ms_pacman_gym.git

Environment

Two observation modes are available for now (as explained and coded by Hamilton et al. 2014):

  • Sparse scalar observation (Binary number 0-2^16-1), some bits zeroed out
  • Sparse observation vector (16 bits), some bits zeroed out
  • Full scalar observation (Binary number 0-2^16-1)
  • Full observation vector (16 bits)
  • Fully observable ASCII (21 x 19 str)
  • Fully observable RGB tensor of (21 x 19 x 3 uint8)