This is a C++ SDL2 project where little creatures all have their own neural network brains
After running a simulation it will pick the best creatures of the generation and allow them to reproduce for the next generation
Each creature will have a mutated brain of a creature from the previous generation
Run on release mode for the best performance

Make sure the SDL2-2.28.4/lib/_x64 library in in the environment path variables otherwise you may get an error: "SDL2.dll cannot be found"
Restart Visual Studio after doing this

Modify the settings in "EvoProj V2.cpp"

How the brains work:

• Brains are composed of input, internal and output neurons
• Neurons have connections with each other of varying weights
• Every frame the input neurons will trigger based on what type of input neuron they are
• Some inputs include sight rays, age, oscillator, current rotation, distance to border and random
• Internal neurons exist to allow more complex functions to take place. They can have any number of inputs and outputs
• Output neurons cause the creature to move or rotate based on whether the input reaches a certain threshold
• Brains are defined by a list of ints, each being a connection between two neurons and a weight
• Mutations involve changing random bits of the genome
• Creatures are rewarded based on how many gates they pass through

Controls:

• You can pan the camera around by holding the mouse wheel
• Zoom in/out by scrolling the mouse wheel
• Click on a creature to view its brain at the top left of the map
• Blue squares are inputs, grey are internal, red are outputs. It's not readable enough to be useful but I was finished the project before I improved it
• You can also see the sight rays of a selected creature
• Use the pause and speed up buttons on the GUI to change the timespeed
• The restart button restarts the simulation and gives creatures new random brains