Open jiaerfei opened 1 month ago
👋 Hello @jiaerfei, thank you for reaching out and for your kind words about our work! 🚀
This is an automated response, but don't worry—an Ultralytics engineer will assist you soon. 😊
In the meantime, here are some resources that might help:
For questions about the ConfusionMatrix or metrics, please check our Docs, specifically the Metrics section where various concepts are explained.
If this pertains to a 🐛 bug or any discrepancy, please provide a minimum reproducible example to assist us in replicating your scenario.
For custom training or implementation-specific questions, include as much detail as possible—such as logs or configuration files—to get the best assistance.
Join our vibrant community for more discussions:
Make sure your ultralytics
package is up-to-date with all requirements in a Python>=3.8 environment with PyTorch>=1.8:
pip install -U ultralytics
Run YOLOv8 in these verified environments (ensuring all dependencies are installed):
If the CI badge is green, all CI tests are passing, confirming correct operation of all Modes and Tasks.
Thank you for your attention, and we look forward to assisting you further! 🎉
Yes that's correct. The comments are not accurate.
Search before asking
Question
Hi, thank you for your excellent work!
I have a question about the result description in ConfusionMatrix.process_batch https://github.com/ultralytics/ultralytics/blob/6dcc4a0610bf445212253fb51b24e29429a2bcc3/ultralytics/utils/metrics.py#L373 https://github.com/ultralytics/ultralytics/blob/6dcc4a0610bf445212253fb51b24e29429a2bcc3/ultralytics/utils/metrics.py#L378
This is my understand: Line 373 means missing detection, a class is predicted as background Line 378 means wrong detection, background is predicted as a class
So, the description in the code may need exchange, is my understand correct ?
Thank you for reply!
Additional
No response