question about BN

[dyhBUPT]

Hi, I have a question about BN here:

pytorch-OpCounter/thop/vision/basic_hooks.py

Lines 60 to 69 in 43c064a

	def count_normalization(m: nn.modules.batchnorm._BatchNorm, x, y):
	# TODO: add test cases
	# https://github.com/Lyken17/pytorch-OpCounter/issues/124
	# y = (x - mean) / sqrt(eps + var) * weight + bias
	x = x[0]
	# bn is by default fused in inference
	flops = calculate_norm(x.numel())
	if (getattr(m, 'affine', False) or getattr(m, 'elementwise_affine', False)):
	flops *= 2
	m.total_ops += flops

pytorch-OpCounter/thop/vision/calc_func.py

Lines 43 to 45 in 43c064a

	def calculate_norm(input_size):
	"""input is a number not a array or tensor"""
	return torch.DoubleTensor([2 * input_size])

Maybe the MACs of BN (eval, no affine) should be x.numel(), not 2 * x.numel()?

[woailunhua]

他这个咋算的啊，就单单均值和方差的计算都不止这些吧。 如果不考虑eps，如果输入是（b,l,d)我算的flops是7bld+bl，用大O表示法就是BLD