关于GMac和FLOPs讨论

FLOPs： s小写，指浮点运算数，理解为计算量。可以用来衡量算法/模型的复杂度。（模型） 在论文中常用GFLOPs
（1 GFLOPs = 10^9 FLOPs）10亿次浮点运算数
(1 MFLOPs = 10^6 FLOPs）1百万次浮点运算数
一个MFLOPS（megaFLOPS）等于每秒一百万（=10^6）次的浮点运算，
一个GFLOPS（gigaFLOPS）等于每秒十亿（=10^9）次的浮点运算，
一个TFLOPS（teraFLOPS）等于每秒一万亿（=10^12）次的浮点运算，(1太拉)
一个PFLOPS（petaFLOPS）等于每秒一千万亿（=10^15）次的浮点运算，
一个EFLOPS（exaFLOPS）等于每秒一百京（=10^18）次的浮点运算，
一个ZFLOPS（zettaFLOPS）等于每秒十万京（=10^21）次的浮点运算。
https://github.com/sovrasov/flops-counter.pytorch

with torch.cuda.device(0):
macs, params = get_model_complexity_info(model, (3, 256, 192), as_strings=True,
                                         print_per_layer_stat=True, verbose=True)
    print('{:<30}  {:<8}'.format('Computational complexity: ', macs))
    print('{:<30}  {:<8}'.format('Number of parameters: ', params))

关于GMACs = GFLOPs讨论https://github.com/sovrasov/flops-counter.pytorch/issues/16
回答：
is there a typo here? I did a little reading and it seems that @snownus has it right. In general a multiply-accumulate is one multiplication and one addition, which can each be floating point operations. So 1 GMAC counts as 2 GFLOPs, meaning GMACs = .5 * GFLOPs (I’m not sure if this is what was already meant).

As for fused multiply-add (FMA) it seems that (if it is supported on a given chip/system) the two FLOPs are indeed computed “in a single step” (see here) or “at once” (see here). But this confuses our conversion. Perhaps in the case of FMA it is more accurate to say 1 GMACs = 1 GFLOPs? Hopefully someone with more expertise than me can clarify!
Params(M) MACs(G)

HRNet-W32 256×192实测
Computational complexity: 7.7 GMac
Number of parameters: 28.54 M
HRNet-W32256×192
论文中7.10 GFLOPs