Skip to content

r4mbhardwaj/CogniMaze

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CogniMaze 🚀

Hey there, fellow maze-crushers and code whisperers! Welcome to CogniMaze, a not-so-typical blend of brainy algorithms and wicked mazes that'll make both your neurons and GPUs buzz with excitement. On this experimental quest, we're pushing the boundaries of how virtual beings think, move, and straight-up slay challenges. It’s all about that clever AI action, with a dash of gaming fun!

What You Need to Run This Bad Boy 🎮

Make sure your rig is equipped with:

  • Unity Engine (Recommended version: 2022.3.13f1; yeah, we’re that specific!)
  • Python 3.9.13 (Snake emoji goes here, but really, just get Python.)

Set Up Like a Boss 🛠

Follow these breadcrumbs to get the ball rolling (or the agent running, we should say):

  1. Summon the Project from the Depths of the Web:

    Go forth and clone. Bring forth every ounce of code with this charm:

    git clone https://github.com/r4mbhardwaj/CogniMaze.git --recurse-submodules
  2. Craft Your Code Sanctuary:

    Migrate to the ml/ dungeon and concoct your magical environment:

    cd CogniMaze/ml/
    python -m venv venv

    Awaken your newfound powers:

    On Windows:

    venv\Scripts\activate

    On macOS and Linux:

    source venv/bin/activate
  3. Empower Your Realm with All the Goodies:

    Load up on the secret sauce – the dependencies that'll make your code fly:

    pip install -r requirements.txt

Dive into the Maze like a Legend 🎩

Your arsenal is prepped, now let the games begin.

Train Your Digital Prodigy

Hop into the ml/ lab and turn those agents into prodigies:

cd CogniMaze/ml/
mlagents-learn --run-id=<run_name> --train

Give your run a name that’ll make it viral on the leaderboards.

Paused? Wanna get back into the groove? Just shout 'Resume':

mlagents-learn --run-id=<run_name> --resume

Show Off Your Agents' Mad Skills

Ready

About

Unity Engine Deep Reinforcement Learning Test

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published