cuDLA YoloX

FP16 dla_linear ✅

INT8 dla_linear ✅

Env

Orin

Docker image (nvcr.io/nvidia/l4t-jetpack:r36.2.0)

Dependencies

sudo apt update

sudo apt install libopencv-dev libjsoncpp-dev python3-pip git git-lfs

YOLOX ONNX

ONNX file is modified from yolox-sPlus-T4-960x960-pseudo-finetune.onnx by removing Reshape and following layers.

Instructions

FP16

bash data/model/build_dla_standalone.sh

make run_fp16

Reference

https://github.com/NVIDIA-AI-IOT/cuDLA-samples

https://github.com/tier4/trt-yoloXP/tree/main