CirCNN: Accelerating and Compressing Deep Neural Networks Using Block-CirculantWeight Matrices

Devs

CirCNN: Accelerating and Compressing Deep Neural Networks Using Block-CirculantWeight Matrices | Read Paper on Bytez