This example demonstrates how to perform inference using YOLOv8-onnx-models,now support: paddledetection-v2.7 ultralytics-v8.1
# download onnx_runtime_lib
bash run.sh download_runtime_lib
# must change mount dir before
bash run.sh mount
# create docker
bash run.sh create
bash run.sh build
# export model of engine & onnx, it depends on the hardware environment
bash run.sh export
bash run.sh run
# 先去把paddledetection的tensorRT版本跑了
bash run.sh eval
if you want to load your own onnx model, may need onnx-runtime