How can i train a model don't use pre trained model?

[morestart]

i use this command --cfg yolov5custom.yaml --weights ''
but i get this error:
AssertionError: '' acceptable suffix is ['.pt']

[morestart]

i use yolov5 v6.0, my current solution is disable check_suffix(weights, '.pt')

[glenn-jocher]

@morestart thanks for letting us know about this. Another user reported the same issue on Windows. I'm working on a fix for this now.

[glenn-jocher]

@morestart good news 😃! Your original issue may now be fixed ✅ in PR #5192. This PR allows empty strings to pass through check_suffix() without error, which multiple windows users had reported was causing issues.

To receive this update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload with model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!

[morestart]

best work! thanks for your reply~

[glenn-jocher]

Great!