How to convert onnx to tensorRT

[mauricewells]

@lucasjinreal I converted your example torch model(sparse_inst_r50_giam_aug_2b7d68.pth to onnx
and tested onnx model using ort_infer.py
But what I want to do is just to convert onnx model to tensorRT and to inference on Jetson device
Can you please let me know how to do this?