Jetpack version : 4.6.2
cuda version : 10.2.300
cudnn version : 8.2.1.32
Tensorrt: 8.2.1.8
I now have a model in OnnX format that I converted to Engine and I want to test their accuracy differences on Jetson Nano. * Can I use Polygraphy and how do I install it?
Do you mean to clone Tensorrt from Github and transfer it to Jetson Nano for compilation and installation?Whether this will conflict with the original Tensorrt. Thanks.