Install YoloV5 on moment pool cloud and test the tutorial

Official warehouse: github.com/ultralytics…

The official document: docs.ultralytics.com/quick-start…

In this case I rented a k80 mirror and chose pytorch1.8.1

Cloning of warehouse

git clone https://github.com/ultralytics/yolov5.git
Copy the code

I used the domestic mirror acceleration here

https://github.com.cnpmjs.org/ultralytics/yolov5.git
Copy the code

Install the required environment

cd yolov5
pip install -r requirements.txt
Copy the code

test

python detect.py --sourceData /images/ --weights yolov5s.pt --conf 0.4Copy the code

In my case, you can see the test images in the /yolov5/runs/detect/exp5 directory

Refer to the article

Ubuntu18 YOLOv5 configuration and training

The most detailed yoloV5 environment configuration setup ever + configuration required files

Cuda11.1, CUDNN8.0.5, PyTorch1.8.0, tensorrt7.2.2.3 are configured in THE YOLOv5 — RTX 3080 CONda environment

Tutorial: super detailed yolov5 model training from scratch

YOLOV5 parsing