yolofastestv2 Pytorch to onnx

[danielzhangau]

Hi there, I am having trouble with converting yolofastestv2 model from pytorch to onnx
More problem detail is on: hpc203/yolo-fastestv2-opencv#1 (comment)
Any helps will be appreciated!!

Cause identification:

yolofastestv2 源代码里输出2个尺度特征图的，yolofastestv2-opencv version在转换onnx文件的程序里做了一个torch.cat的处理，把两个输出特征图拼接成一个

My understanding is the model output is two different scale of feature maps, but the OpenCV version implementation only accept one. How to concatenate two features maps when converting the format from PyTorch to onnx?