Layer BatchedNmsPlugin failed validation


I’m trying to add a BatchedNmsPlugin after YoloV4 tiny model. My workflow is converting the darknet model to onnx, then using the onnx-surgeon to add the BatchedNmsPlugin. Finally, I want to use the trtexec to convert the onnx to tensorRT, but I got the following error.


TensorRT Version : 7.1.3
GPU Type : Jetson Xavier NX
Nvidia Driver Version :
CUDA Version : 10.2
CUDNN Version : 8
Operating System + Version : Ubuntu 18.04

Relevant Files

Steps To Reproduce

sudo ./trtexec --onnx=modified.onnx

Hi @jack_gao,
Kindly allow access to your files.


OK! I’ve already allowed


Hi @jack_gao,
Looks like the issue is with your custom plugin implementation or registration.
Below links might help you in the same.


Hi @AakankshaS,

I was following advice.

Using onnx-graphsurgeon like .
It looks simple and easy. Now it get the error but I don’t know which step goes wrong. It looks like it will register from automatically, isn’t it?


OK! I’ve already allowed

Hi @jack_gao ,
Apologies for delayed response, are you still facing the issue?