How to deploy the deep learning target detection model designed by myself on Jetson naon?

Using deepstream, I can deploy the yolov5 model to Jetson nnao, but I have changed the model structure and improved it. How can I deploy it?

Hi,

Do you use the ONNX format for the model?
If yes, the steps are similar but need some parameter updates.

Thanks.

Hi,
Is there a related tutorial?Can you give me a link?

Thanks.

Hi,

Sorry that we don’t have the exact file for YOLOv5.
But below is a tutorial for customizing the YOLOv3 model for your reference:
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_custom_YOLO.html

Thanks.