torch fx based pytorch model converter, including pytorch2caffe, pytorch2onnx.
torch fx based pytorch model quantizier.

pip install brocolli

• torch2caffe

  • caffe installation

  pip install brocolli-caffe

  import torchvision.models as models
  from brocolli.converter.pytorch_caffe_parser import PytorchCaffeParser
  net = models.alexnet(pretrained=False)
  x = torch.rand(1, 3, 224, 224)
  pytorch_parser = PytorchCaffeParser(net, x)
  pytorch_parser.convert()
  pytorch_parser.save('alexnet')
  

  run this script until you see "accuracy test passed" on screen, then you can get alexnet.caffemodel and alexnet.prototxt under under current folder.

• torch2onnx

  import torchvision.models as models
  from brocolli.converter.pytorch_onnx_parser import PytorchOnnxParser
  net = models.alexnet(pretrained=False)
  x = torch.rand(1, 3, 224, 224)
  pytorch_parser = PytorchOnnxParser(net, x)
  pytorch_parser.convert()
  pytorch_parser.save('alexnet.onnx')
  

  run this script until you see "accuracy test passed" on screen, then you can get alexnet.onnx under current folder.

QQ Group: 597059928

image

Give a 🌟 if this project helpes~