doomformer: neural networks computed inside DOOM maps

Developer alp (@mottopanikeiku) released doomformer, a project that compiles an integer neural network into a DOOM map, using 1,355 voodoo dolls to perform the arithmetic on an MNIST network. Light levels and plane heights store the values, and every value the map exposes is read back and compared against NumPy.

doomformer is a project from alp @mottopanikeiku that compiles an integer neural network into a DOOM map. Voodoo dolls do the arithmetic. Light levels and plane heights hold the values. Every value the map exposes is read back and compared against NumPy. One map puts 1,355 voodoo dolls to work on an MNIST network. The …