I have a question about building an inference engine for TensorRT

Robotics & Edge Computing Jetson & Embedded Systems Jetson Nano
1

Hi,
I have a question.

When building the Inference Engine from ONNX, DGX was used to save the data.
I tried to move the saved inference engine to Xavia jetson nano and run the inference, but
I got a None Type Error and the script did not run.

Do I need to build the inference engine in the actual device where I want to run the inference?
Also, if the build environment is not affected, how can I avoid the error?

I’m sorry to ask such a rudimentary question, but could you please answer it for me?

Thank you very much for your help.

3

Hi,

Since engine file is device dependent, please copy the ONNX model instead.
And create the engine file from ONNX format directly.

Thanks.

4

Hi, @AastaLLL
Thank you very much for your answer.
I guess it depends on the device.

I will copy the ONNX file to the device where I want to actually infer it, and then build and use the engine.

Thank you very much.