Kernel weights has count 2304 but 32640 was expected

Hi, jinshiksung

Is your issue can also be reproduced with the sample shared by terpysk?

Thanks.

The onnx version works. Tensorrt version is still not working and is slow

I like to make run it on Tensorrt c++ jetpack to see it run faster as original question
otherwise no point

Please help me to run it on Tensorrt c++ jetpack

Hi,

Did you get the correct ONNX result in C++ as well?

As shared above, we can get the correct output after applying the post-processing in the YOLOp repository.
To run the model in C++, you will need a C++ post-processing implementation as well.
Do you have the implementation already?

Thanks.

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