Deploying ONNX model onto Jetson Nano

[user17113]

I am new to Jetson nano. I have a trained Resnet-18 model and converted into ONNX model. And i wanted to deploy this onto Jetson Nano, how can I do that?

[AastaLLL]

Hi,

You can run the model with trtexec directly.

$ /usr/src/tensorrt/bin/trtexec --onnx=[your model]

And below is an example with input and output handling for your reference:

Custom ResNet Jetson Xavier

Hi, The error comes from dynamic shape usage. If you don’t need to reshape the model at the runtime, please set EXPLICIT_BATCH to use static shape instead. Here is an example of onnx model for your reference: import cv2 import time import numpy as np import tensorrt as trt import pycuda.autoinit import pycuda.driver as cuda EXPLICIT_BATCH = 1 << (int)(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH) TRT_LOGGER = trt.Logger(trt.Logger.INFO) runtime = trt.Runtime(TRT_LOGGER) host_inputs = …

Thanks.

[user17113]

Hello @AastaLLL . Thanks for the reply. Finally I manage to get to the .trt file. Is this the right step? How am I suppose to do inference on Nano using this file?

[AastaLLL]

Hi,

You can inference it with the sample shared above.
It is a python example with OpenCV for input/output handling.

Thanks.