bintruekenobi / ising-model

Ising model, Glauber dynamics, Metropolis-Hastings algorithms, and renormalization.

ising-model

glauber dynamics and metropolis-hastings algorithm

gifs below are from the glauber.py

n is set to 1 for ideal simulation running, setting n>1 for an animated plot will result in more squares updated per frame, so more computation per frame and longer evolution. for a final picture of the grid as the end product, comment out the animation script and put plot_grid2(grid) where the "#return grid" line is kept (line 73), giving a grid of coloured squares/rectangles (which may need to be full screened if there are hundreds of grid points per row/column as there is a black line between the squares which results in the grid looking like a single solid rectangle). This is the 2d case, so it is directly from Kramers-Wannier duality that the curie temperature of the 2d system is approximately 2/ln(1+rt(2)), or 2.2692, it's interesting to try e.g. T=2.3 and T=2.2 to see how they differ over large times.