tienluongngoc/yolov5_triton_inference_server

👋 Hi!, I'm Tien Luong Ngoc - Software Engineer

• Skills: AI, Computer Vision with Python, C/C++.

• TIKTOK @tienluongngoc.it.

• LinkedIn: https://www.linkedin.com/in/tienluongngoc/.

yolov5_triton_inference_server's Issues

Only YOLOv5 v5.0 models work while using this guide

python gen_wts.py -w yolov5s.pt -o yolov5s.wts command only works for yolov5 v5.0 models, not v6.0 and v6.1.

I recommend adding a line : wget https://github.com/ultralytics/yolov5/releases/download/v5.0/yolov5s.pt in the "Create *.wts file" segment of the README.md file.

Changes :-

cd yourworkingdirectoryhere
cp tensorrtx/yolov5/gen_wts.py yolov5
cd yolov5
wget https://github.com/ultralytics/yolov5/releases/download/v5.0/yolov5s.pt
docker run -it --rm --gpus all -v $PWD:/yolov5 tienln/ubuntu:18.04_conda /bin/bash
cd /yolov5
conda activate yolov5
python gen_wts.py -w yolov5s.pt -o yolov5s.wts

Failed to connect to all addresses

I got this error when running client (client.py) "tritonclient.utils.InferenceServerException: [StatusCode.UNAVAILABLE] failed to connect to all addresses". I had run server before runing client. Do you know the problem? Thank for your help!