onnx2trt

A sample Jupyter notebook to explain how to export ONNX model to NVIDIA TensorRT engine.

Target platform

NVIDIA Jetson Developer Kit with JetPack
- Tested with Jetson Xavier NX Developer Kit with JetPack 4.6

Clone this repository.

git clone https://github.com/MACNICA-CLAVIS-NV/onnx2trt

Build the Docker cotainer image contains the required softwares for this Jupyter notebook.
```
./scripts/docker_build.sh
```

Run a container from the build image.
```
./scripts/docker_run.sh
```
Launch Jupyter Lab with the instruction displyed at the above step.
Open the onnx2trt.ipynb notebook from the /onnx2trt path in the Jupyter file browser.

Name		Name	Last commit message	Last commit date
Latest commit History 8 Commits
scripts		scripts
.gitignore		.gitignore
Dockerfile		Dockerfile
README.md		README.md
common.py		common.py
onnx2trt.ipynb		onnx2trt.ipynb
trt_analyzer.py		trt_analyzer.py