The detection capability decreases after the pytorch model was converted to the tensorRT model

jetson TX2 NX
Jetpack 4.6.2 
YoloV5 v6.1

My pytorch model can detect the target correctly, but it can’t detect the target correctly after converting it to the engine file. In addition, the original yolov5 does not have this problem. This problem occurs when I modify its backbones


This looks like a Jetson issue. Please refer to the below samples in case useful.

For any further assistance, we will move this post to to Jetson related forum.


This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.