CIS 365 Project

Exploring neural networks with minesweeper

The files in this repository are as follows:

• minesweeper.py - the main minesweeper game. Only a few helper functions are added for the agent
• agent.py - runs the minesweeper agent. Agents can be configured in the python file
• networktrainer.py - trains a keras neural network from data in "trainingdata.csv"
• *.csv - various files containing training data
• ml.net directory - contains files for ML.NET project
  • MinesweeperML.NET contains the model builder configuration and boilerplate code
  • MLModel1_WebApi1 contains the API used for interop between the python game and the ML.NET model - the binary generated by this project must be running for the ML.NET agent to work
• keras_model* - directories containing some of the keras models we trained

Original README

Python Tkinter Minesweeper

Minesweeper game written in Python using Tkinter GUI library.

Contents:

• /minesweeper.py - The actual python program
• /images/ - GIF Images ready for usage with Tkinter
• /images/original - Original PNG images made with GraphicsGale

To Do:

• Have specific number of mines, rather than random
• Highscore table
• Adjustable grid and mine count via UI