classify/train.py script should accept patience parameter

[sohang3112]

Search before asking

• I have searched the YOLOv5 issues and found no similar feature requests.

Description

This Stack Overflow answer gives a good explanation of what patience is

Suppose a model is being trained for 100 epochs. But after epoch 60, the model's accuracy & loss don't improve any more.
If 5 more epochs pass and the model's accuracy & loss still don't improve, then the training loop should stop running.

In this example, the best epoch is 60, and patience is 5 epochs (number of epochs for which training loop should continue running when accuracy & loss stop improving).

Use case

This would save time during model training, as the training loop will not keep running for no reason even when model's loss is not decreasing (or even increasing!)

Additional

No response

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[glenn-jocher]

@sohang3112 thank you for suggesting this feature, it could definitely be a useful addition to YOLOv5. We appreciate your willingness to contribute to the project by submitting a PR. Our team will review and consider your idea for implementation in future updates.

[github-actions]

👋 Hello there! We wanted to give you a friendly reminder that this issue has not had any recent activity and may be closed soon, but don't worry - you can always reopen it if needed. If you still have any questions or concerns, please feel free to let us know how we can help.

For additional resources and information, please see the links below:

• Docs: https://docs.ultralytics.com
• HUB: https://hub.ultralytics.com
• Community: https://community.ultralytics.com

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLO 🚀 and Vision AI ⭐

[sohang3112]

Unlike classify/train.py, train.py has implemented patience using utils.torch_utils.EarlyStopping class. Will try to use this approach for implementing in classify/train.py also. A slight modification in EarlyStopping class would also be needed - its logging message would have to print classify/train.py instead of train.py.

[glenn-jocher]

@sohang3112 thank you for the suggestion, we appreciate your input. It's great to hear that patience has already been implemented in the train.py script using the utils.torch_utils.EarlyStopping class. We will certainly consider your idea for implementing patience in classify/train.py using a similar approach and modifying the logging message to print the correct script name. Your contribution to the YOLOv5 community is valuable, and we thank you for your help to improve the project.

[github-actions]

👋 Hello there! We wanted to give you a friendly reminder that this issue has not had any recent activity and may be closed soon, but don't worry - you can always reopen it if needed. If you still have any questions or concerns, please feel free to let us know how we can help.

For additional resources and information, please see the links below:

• Docs: https://docs.ultralytics.com
• HUB: https://hub.ultralytics.com
• Community: https://community.ultralytics.com

Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

Thank you for your contributions to YOLO 🚀 and Vision AI ⭐