https://github.com/thomasballinger/tetrisai

https://github.com/thomasballinger/tetrisai

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/thomasballinger/tetrisai
• Owner: thomasballinger
• Created: 2013-04-28T21:43:13.000Z (over 11 years ago)
• Default Branch: master
• Last Pushed: 2013-05-06T14:27:14.000Z (over 11 years ago)
• Last Synced: 2024-12-16T13:43:32.645Z (about 1 month ago)
• Language: Python
• Size: 170 KB
• Stars: 1
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

        
Tetris ai

General todo:

* do enough research to defend the genetic algorithm approach

* read about probabilistic nodes in expectiminmax trees - think about utility

Done in c

---------

* get all valid moves for a board and a piece spec

* evaluate a move given a board, move spec, and metrics

TODO in the c

-------------

* check that moves are actually accessible, not just open

* simple choose best move for simple alg function

* make a header file

* python interface

* implement more metrics

* simulation setup

* start benchmarking so refactors can be based on data

* lookahead

  * choose k best moves to take forward

  * use a probabilistic node 

  * for each of the k moves,

    * build a new board

    * find all legal moves

    * evaluate them

    * choose k to take forward