Implement popular deep learning networks in pytorch, used by tensorrtx.
Implement popular deep learning networks in pytorch, used by tensorrtx.
This is a brother project with wang-xinyu/tensorrtx.
Popular deep learning networks are implemented with pytorch in this project. And then weights files are exported for tensorrt implementation.
pytorch-summary is a very useful tool for understanding the model structure, for example it can output the dimensions of each layer.
Clone, and cd into the repo directory.
git clone https://github.com/sksq96/pytorch-summary
python setup.py build
python setup.py installMost of the models are from torchvision, exception for yolov3, which has a readme inside.
A file named xxxnet.py can do inference and save model into .pth.
And a file named inference.py can do inference and save weights into .wts, which is used for tensorrt.
For example, googlenet,
cd googlenet
python googlenet.py // do inference and save model into .pth firstly.
python inference.py // then do inference and save weights fileNo open issues yet, or sync has not completed.