Torch.nn.instancenorm

why running_mean/var’s shape is C, not N * C
according to the [1607.08022] Instance Normalization: The Missing Ingredient for Fast Stylization, the shape of mean/var is N *C , but for torch.nn.instancenorm ,
at::alias(running_mean).copy_(running_mean_.view({ b, c }).mean(0, false)); }
i don’t know the operation of “mean(0,false)” is use for ?
pytorch/Normalization.cpp at master · pytorch/pytorch · GitHub

Who can tell me ? Anyone know that?