BOTS_snake

CIFO 2020/21 Project - Group: BOTS

Group Members

Berfin Sakallioglu - 20200545
Mehmet Serdar Çetiner - 20201016
Oguz Kokes - 20201006
Tiago Gonçalves - 20201053

Files:

Report PDF

snake_crossover.py - crossover operators

snake_mutation.py - mutation operators

snake_selection.py - selection operators

snake_charles.py - charles library adapted to our snake model

Snake_Game.py - game implementation taken from https://github.com/TheAILearner/Training-Snake-Game-With-Genetic-Algorithm/blob/master/Snake_Game.py , with some little changes

Run_Game.py - function to run the game with a Neural Network and return the fitness and score achieved by the specific snake. Also taken from https://github.com/TheAILearner/Training-Snake-Game-With-Genetic-Algorithm/blob/master/Run_Game.py , but several changes were done

RunGame_BestNN.ipynb - Extra file - Running the game with our best model, either for statistical analysis of performance or for visualization purposes

fitness2 folder - Folder with fitness logs (not every csv is included, just a sample)

result_analysis.ipynb - Notebook used to analyse the results obtained

Video of our best model playing 5 games: https://liveeduisegiunl-my.sharepoint.com/:v:/g/personal/m20201053_novaims_unl_pt/EfTwQbXnC25Fj8TAMqsfPuMB5r1i_M9kYqhEC8Jh940UCA?e=JCCq8a