knoppmyth

@BrianP8701 You can removed models_returned. I included the entry method for completion. I wrote is as part of a Streamlit app so I can compare models against once another. The...

@aaSchcolnik Hello. Yeah, I used the 'code' button to paste the code and no matter what, it wouldn't format it right... What opset did you use when exporting the model...

@aaSchcolnik Cool. I've not attempted to use a webcam as it currently isn't in my use case. However, if I were to, I'd make a function that does the detection....

@wereii Done.

Assuming you still have the dataset, retrain with YOLOv8. If memory serves (from when I experimented w/ YOLOv4), all you should need as a YAML file pointing to the location...

@hongu0603 You cannot take a YOLOv4 cfg and weights and run them with YOLOv8. If you want to use YOLOv8, you need to train a new model with it.

Try reducing the batch size.

Perhaps [this](https://www.youtube.com/watch?v=l_kf9CfZ_8M) will give you some ideas.

> If anyone knows our legal obligations that would help massively. IANAL Did you make code changes to YOLOv5 to add this functionality? If so, your legal obligation is to...

When you're transfer learning you don't appending the class names. With respect to the label text file, it doesn't have the name but a number for the class. `0 0.09439124487004104...