When training the model, there is a warning that the size is incorrectWARNING ⚠️ updating to 'imgsz=1280'. 'train' and 'val' imgsz must be an integer, while 'predict' and 'export' imgsz may be a [h, w] list or an integer, i.e. 'yolo export imgsz=640,480' or 'yolo export imgsz=640'

[charming1202]

I set the imgsz parameter to [1280,720] before training the yolov8 model, and WARNING ⚠️ updating to 'imgsz=1280' is displayed during training. 'train' and 'val' imgsz must be an integer, while 'predict' and 'export' imgsz may be a [h, w] list or an integer, i.e. 'yolo export imgsz=640,480' or 'yolo export imgsz=640', what is the reason for this, and what should I do if I want to train at the original size

[UltralyticsAssistant]

👋 Hello @charming1202, thank you for reaching out and sharing your experience with the Ultralytics 🚀! This is an automated response, and an Ultralytics engineer will assist you soon.

It seems like you're encountering a warning related to the imgsz parameter during training. For training and validation, the imgsz must indeed be a single integer. To better assist you, please provide a minimum reproducible example so we can debug the issue effectively 🐛.

For immediate guidance, I recommend reviewing the Python and CLI examples in our Docs 📚.

If you'd like real-time interaction, feel free to join our Discord 🎧. For more in-depth discussions, check out our Discourse or our Subreddit.

Upgrade

Make sure you are using the latest version of ultralytics by upgrading your package:

pip install -U ultralytics

Recommended Environments

YOLOv8 can be used in the following environments with all dependencies preinstalled:

• Notebooks:
• Google Cloud: Check the GCP Quickstart Guide
• Amazon: Follow the AWS Quickstart Guide
• Docker: Refer to the Docker Quickstart Guide

Status

The green status of our CI tests indicates that all YOLOv8 functionalities are working correctly across various platforms.

Stay tuned for further assistance from our team! 😊

[glenn-jocher]

@charming1202 the warning occurs because 'train' and 'val' require a single integer for imgsz. To train at a specific size, set imgsz=1280 or another integer that fits your needs.

[charming1202]

Thank you for your answer, I set imgsz to an int number and I can train normally

[glenn-jocher]

@charming1202 you're welcome! Glad to hear it's working for you now. If you have any more questions, feel free to ask.