Keep only one yolo class on Xavier NVX using DS5


Hi, we installed a brand new Xavier NX using DS5. We are doing tests using Yolo3 model. How can we turn off all the detection class and only use one for processing. Let’s say if we only want to detect books, in which files config file do we change and choose only one class to be processed and not all of them. The reason is, when processing video streams, by default it is detecting all 80 classes and makes the Xavier’s fps performance drop to 6fps.

Thanks for any feedback.

A clear and concise description of the bug or issue.


Xavier NX, DS5, Jetpack 4.4
TensorRT Version: 7
GPU Type: Xavier NX
Nvidia Driver Version: Latest
CUDA Version:
CUDNN Version:
Operating System + Version: Ubuntu
Python Version (if applicable):
TensorFlow Version (if applicable):
PyTorch Version (if applicable):
Baremetal or Container (if container which image + tag):

Relevant Files

Please attach or include links to any models, data, files, or scripts necessary to reproduce your issue. (Github repo, Google Drive, Dropbox, etc.)

Steps To Reproduce

Please include:

  • Exact steps/commands to build your repro
  • Exact steps/commands to run your repro
  • Full traceback of errors encountered

I doesn’t seems to be TensorRT issue.
I found few links which might be useful:

In case of further query, request you to raise issue in Yolo3 forum/git issues section.