Example workflow involving YoloV5. See this blog post for more instructions: XX. The main entry point is yolov5/prepare_dataset.py, which downloads a dataset, preps it for YoloV5, and downloads pretrained weights files. Roughly, you'll need to follow these steps:

• Clone YoloV5 from within the YoloV5 repository, i.e. git clone git@github.com:ultralytics/yolov5.git
• Install libs from YoloV5's requirements.txt
• Run python prepare_dataset.py to download WiderFace data from HuggingFace. You can then use YoloV5 training/inference CLI as expected.