Replies: 3 comments 2 replies
-
@MiXaiLL76 good afternoon, Thank you for reaching out and for your interest in contributing to the Ultralytics project! Your faster-coco-eval library sounds like a promising tool for speeding up the validation of COCO data, and the potential support for LVIS data is exciting. Regarding your question about submitting a PR to integrate faster-coco-eval as an alternative to pycocotools, we are always open to improvements that can enhance the performance and functionality of our repository. However, before proceeding with a PR, we recommend the following steps:
Once you have these details ready, feel free to submit a PR. We look forward to reviewing your contribution and discussing it further with the community. Thank you again for your initiative and for considering Ultralytics for your work. If you have any further questions or need assistance, please don't hesitate to ask. |
Beta Was this translation helpful? Give feedback.
-
Thanks for the detailed answer, I will try to make the right PR to your repository! |
Beta Was this translation helpful? Give feedback.
-
Started basic testing, soon I will work on PR |
Beta Was this translation helpful? Give feedback.
-
Good afternoon, I am developing the faster-coco-eval library.
It, as you might guess, speeds up the validation of coco data in a format identical to the original one.
I plan to implement lvis data support soon.
I know that basically ultralytics uses its own implementation of metrics calculation, so I want to ask the question, will PR be valid for you to use faster-coco-eval instead of pycocotools?
Examples of work can be found in the library repository
https://github.com/MiXaiLL76/faster_coco_eval
or
https://nbviewer.org/github/MiXaiLL76/faster_coco_eval/blob/main/examples/eval_example.ipynb
https://nbviewer.org/github/MiXaiLL76/faster_coco_eval/blob/main/examples/curve_example.ipynb
Beta Was this translation helpful? Give feedback.
All reactions