torchlop 0.2

A tool to layer-wise count the MACs and parameters of PyTorch model.

PyTorch-layerwise-OpCounter

A tool for profile the MACs, parameters, input_shape, output_shape et.al of each layer in Pytorch model. Forked from Lyken17/pytorch-OpCounter which is not supporting layer-wise profile and I will follow it.

How to install

pip install torchlop

OR

pip install --upgrade git+https://github.com/hahnyuan/pytorch-layerwise-OpCounter.git

How to use

• Basic usage

  from torchvision.models import resnet50
  from torchlop import profile
  model = resnet50()
  input = torch.randn(1, 3, 224, 224)
  macs, params, layer_infos = profile(model, inputs=(input, ))
  

• The layer_infos is a dict that contains the infos for each layer of the pytorch model.

  • The key is the name of the layer
  • 'type': the class name of the layer
  • 'in_size': input size
  • 'out_size': output size
  • 'ops': operations (MACs)
  • 'params': parameters

• Define the rule for 3rd party module.

  class YourModule(nn.Module):
      # your definition
  def count_your_model(model, x, y):
      # your rule here
  
  input = torch.randn(1, 3, 224, 224)
  macs, params, layer_infos = profile(model, inputs=(input, ), 
                          custom_ops={YourModule: count_your_model})
  

• Write the layerwise profile information into a csv file

  from torchvision.models import resnet50
  from torchlop import profile
  from torchlop.rst_process import write_csv
  model = resnet50()
  input = torch.randn(1, 3, 224, 224)
  macs, params, layer_infos = profile(model, inputs=(input, ))
  csv_file='profile.csv'
  write_csv(csv_file,layer_infos)
  

Results of Recent Models

The implementation are adapted from torchvision. Following results can be obtained using benchmark/evaluate_famours_models.py.

Model Params(M) MACs(G) alexnet 61.10 0.77 vgg11 132.86 7.74 vgg11_bn 132.87 7.77 vgg13 133.05 11.44 vgg13_bn 133.05 11.49 vgg16 138.36 15.61 vgg16_bn 138.37 15.66 vgg19 143.67 19.77 vgg19_bn 143.68 19.83 resnet18 11.69 1.82 resnet34 21.80 3.68 resnet50 25.56 4.14 resnet101 44.55 7.87 resnet152 60.19 11.61 wide_resnet101_2 126.89 22.84 wide_resnet50_2 68.88 11.46

Model Params(M) MACs(G) resnext50_32x4d 25.03 4.29 resnext101_32x8d 88.79 16.54 densenet121 7.98 2.90 densenet161 28.68 7.85 densenet169 14.15 3.44 densenet201 20.01 4.39 squeezenet1_0 1.25 0.82 squeezenet1_1 1.24 0.35 mnasnet0_5 2.22 0.14 mnasnet0_75 3.17 0.24 mnasnet1_0 4.38 0.34 mnasnet1_3 6.28 0.53 mobilenet_v2 3.50 0.33 shufflenet_v2_x0_5 1.37 0.05 shufflenet_v2_x1_0 2.28 0.15 shufflenet_v2_x1_5 3.50 0.31 shufflenet_v2_x2_0 7.39 0.60 inception_v3 27.16 5.75