Getting classes

[blacksmith001]

I want to process the results of my classification. I was trying to get classes like this:

for result in results:
boxes = result.boxes # Get the boxes for the first result
classes = result.boxes.cls # Get class indices

But then I am getting:
" 'NoneType' object has no attribute 'cls' "

[UltralyticsAssistant]

👋 Hello @blacksmith001, thank you for reaching out to the Ultralytics community 🚀! It sounds like you're encountering an issue with processing classification results.

If this is a 🐛 Bug Report, could you please provide a minimum reproducible example that we can use to help debug the problem? This will greatly assist us in identifying and resolving the issue.

For issues related specifically to deriving class indices, it might be helpful to ensure your implementation aligns with our Python usage examples. The error 'NoneType' object has no attribute 'cls' suggests that the result may not contain any detected boxes, so it could be useful to verify the outputs and check for empty results.

If you need further assistance, an Ultralytics engineer will look into this soon. In the meantime, you might find our community a valuable resource. For real-time discussions, consider joining the conversation on Discord 🎧, or explore Discourse and our Subreddit for more in-depth discussions.

Upgrade

To ensure you have all the latest features and fixes, upgrade to the newest version of the ultralytics package:

pip install -U ultralytics

Environments

For guaranteed compatibility and ease of use, you can run YOLO in one of our verified environments, each preinstalled with necessary dependencies:

• Notebooks with free GPU:

Status

With all Continuous Integration (CI) tests passing, the green badge indicates that our YOLO Modes and Tasks are verified for correct operation across various platforms.

[glenn-jocher]

@blacksmith001 it seems like the boxes attribute might be None. Ensure your model is set up for detection tasks, as classification results are accessed via result.probs for class probabilities.