TensorRT and Pytorch Documentation/Example

Hi All,

I am wondering if you can suggests good documentation for pytorch 2 tensort conversion.
From the official documentation there are very few example around it.

Basic steps involve pytorch to onnx and then from onxx to tensorRT.

I also found this library from the nvidia Github page, but there is no reference to it in the tensorrt official documentation.

Any suggestion?

Hi @massimo.camplani,
Its okay to use either of the two approaches.
torch2trt or torch << ONNX << TRT
However you may find more samples and better TRT support in ONNX to TRT format.
Please check the below example
https://docs.nvidia.com/deeplearning/tensorrt/sample-support-guide/index.html#onnx_packnet
Thanks!