The inference for adding yolov5 in the deepstream example sends an error when yolov5 turns engine

Yes, I run the trtexec on JetPack 6.0 GA release. The new MMJ release will based on JetPack 6.0 GA. It will release soon(plan is next week). Can we wait and test on next MMJ release?

Could you please test the yolov5 of ai-nvr on your jetpack6.0 GA version, delete the engine file generated in advance, and let deepstream’s docker automatically convert an engine when executing it? If you can test it, I will change my environment to GA version

Current MMJ AI NVR can’t run on jetpack 6.0 GA. Next MMJ release which will release in next week can run on jetpack 6.0 GA.

Could you test the yolov5 of ai-nvr on your Jetpack 6.0DP version, delete the engine file generated in advance, and let deepstream’s docker automatically convert an engine when executing, and test whether it is possible

Yes, the convertion failed on JetPack 6.0 DP. Let’s verify it on next MMJ release next week on JetPack 6.0 GA.