Skip to content
/ tetr.ai Public

Tetris AI using a Genetic Learning Algorithm Model that learns how to play competitive tetris

Notifications You must be signed in to change notification settings

iilou/tetr.ai

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tetris AI app that can train a Tetris bot that can outperform the vast majority of Competitive Tetris Players using a Genetic Learning Algorithm Model built in C++ that can evaluate the best possible next moves, taking into account at most 4 (at depth=2) future pieces at once.

[ APP FEATURES ]

-> You start off with you own personal bot that has a randomized evaulation system

-> There is also a pre-built bot that has a set evaulation system that I have personally obtained by training the same bot that you start off with for a long period of time.

-> You can choose to battle against either your personal bot or the pre-built bot, or you can choose to have the two AI bots challenge each other

-> You can train your own bot using different training enviornments that can alter the behaviour that your personal bot will learn

  -> Modes: 1 vs 1, highest score wins, last to survive 

-> There is a settings menu to alter certain game options

[ VERSION UPDATES ]

1.0 -> Naive O(2^n) approach, still works for 4(2 on display) depth 1.1 (in progress) -> Will implement MCTS to optimize further, and will improve the ui, with images, better sprites, and audio.

[ REQUIRMENTS ]

-> MinGW 13.1.0 installed and set in your global PATH variables
       Download from : [   https://winlibs.com/  ]
      
-> All else included in installation folder

[ SCREENSHOTS ]

Menu: image

Battle (Player vs Personal Bot) image

Training (1 vs 1): image

Settings: image

Example Selection Screen (Training): image

About

Tetris AI using a Genetic Learning Algorithm Model that learns how to play competitive tetris

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages