7

I am comparing different CNN architectures for edge implementation. Some papers describing architectures refer to mult-adds, like the MobileNet V1 paper, where it is claimed that this net has 569M mult-adds, and others refer to floating-point operations (FLOPs), like the CondenseNet paper claims 274M FLOPs.

Are these comparable? Is 1 multiply-add equivalent to 2 floating-point operations? Any direction will be greatly appreciated.

nbro
  • 42,615
  • 12
  • 119
  • 217
Quintus
  • 71
  • 1
  • 2

1 Answers1

4

According to this source, one MAC is roughly equal to two FLOP (multiply accumulate). My guess/understanding would be that the distinction is made because neural nets spend compute overwhelmingly on multiply-accumulate operations, and thus optimizations and statistics over MAC operations would be more significant than FLOPs.

k.c. sayz 'k.c sayz'
  • 2,121
  • 13
  • 27