Skip to content

Tetris is a favorite game of my family in my childhood. I always want to know how it works and its core logic. Now that I have a chance to implement it, I want to add potentials to integrate some trending technology to it. In this project I consider Machine Learning, which an AI can be trained with how the user play the game.

Notifications You must be signed in to change notification settings

zbmike/traintris

Repository files navigation

TrainTris

Background and Overview

Tetris is a favorite game of my family in my childhood. I always want to know how it works and its core logic. Now that I have a chance to implement it, I want to add potentials to integrate some trending technology to it. In this project I consider Machine Learning, which an AI can be trained with how the user play the game.

Functionality and MVPs

In this app, players can

  • play the classic Tetris game
  • know what the next Tetromino is
  • know the score they get
  • play the game with increasing difficulty, as their score goes up

In addition, the game will

  • be ready to get and store the training data for AI
  • when there are enough training data, AI can play the game

Wireframes

Wireframe

Architecture and Technology

+-- dist
+-- src
|   +-- game.js
|   +-- tetrimino.js
|   +-- ai.js
+-- Index.html
+-- .gitignore
+-- node_modules
+-- package.json
+-- package.lock.json
+-- postcss.config.js
+-- README.md
+-- webpack.common.js
+-- webpack.dev.js
+-- webpack.prod.js
  • CanvasHTML and Brain.js or TensorFlow.js

Implementation Timeline

  • First day and Second(if necessary) to implement Tetris game
  • Rest of the time to research method to train Tetris AI

Bonus

  • Successful implementation of AI will be a bonus in this game

About

Tetris is a favorite game of my family in my childhood. I always want to know how it works and its core logic. Now that I have a chance to implement it, I want to add potentials to integrate some trending technology to it. In this project I consider Machine Learning, which an AI can be trained with how the user play the game.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published