Recently I want to split the backbone algorithm of yolov5 and deploy it on multiple FPGAs, but first I need to divide the Darknet53 convolution layer into 2 or more regions on a computer for parallel convolution operations. What should I study in the yolov5 project? What files need to be modified? A bunch of yaml and common.py are so confusing (yolov5 is the latest version)

I have deployed yolov5 on my PC and conducted some preliminary research. Most of the focus should be in the models folder. There are 5 yaml network configuration files in the models folder, yolo.py, common.py, etc. Now I want to understand the order of research and how to learn these files? To divide the Darknet53 convolution layer into 2 or more regions for parallel convolution operations on a single computer