Its a really simple project made with love.
I tried to generalize it for using it for other projects , if you want , you can copy the machine learning stuff(obviously I would be grateful if you give me the credits) , but you need to keep in mind that this isnt the best. And maybe later I will adapt it for a genetic algorithm.
The project works in a really simple way , the player only tried to clasify and give a score to each cell with at least 3 cells where you can recolect information.
The input of the neural network is of 25 ,because it needs to know what its around the cell but it needs for information for a really simple reason and its that it can have some problems and get confused for some cases , so I was forced to make the input bigger.
If you are interested in how the back propagation and all that stuff , you can check it here ,I cant explain it in a better way than the persons mentioned in that project.
The backpropagation algorithm consumes a lot of memory , but maybe its because the language that Im using for the project
The models that I trained arent that accurate but It doesnt really care that much, its just for experimentation
