Pytorch中计算自己模型的FLOPs

转载来源：http://ilovepose.luohuank.xin/t/54?cache=1552532453#Post117

https://github.com/Lyken17/pytorch-OpCounter

安装方法很简单：

pip install thop

基本用法：

from torchvision.models import resnet50from thop import profile
model = resnet50()
flops, params = profile(model, input_size=(1, 3, 224,224))

对自己的module进行特别的计算：

class YourModule(nn.Module):
# your definition
def count_your_model(model, x, y):
# your rule
hereflops, params = profile(model, input_size=(1, 3, 224,224),
custom_ops={YourModule: count_your_model})