Finally finished that Ising model simulation. The reason why I was reading about the ising model is that simple generalizations of it lead to bolzmann and hopfield networks. The Ising model is like a boltzmann network that only remembers the base states, all one spin for J=1 (ferromagnetic) and checkered for J=-1 (antiferromagnetic). This simulation only does Gibbs sampling and I mostly made it as an exercise in c++ but its pretty fun to play with. Since I wrote it in c++ it might be cool to render in opengl and try to get this really performant and dynamic but quadratic complexity and the fact that gibbs samplers cant be parallelized are limitations