New Release: Ultralytics v8.2.59

Discussion
,
1

:rocket: Announcing Ultralytics v8.2.59 Release!

Hello Ultralytics Community!

We are thrilled to announce the release of Ultralytics v8.2.59! This update brings a host of enhancements and fixes designed to improve your experience and the performance of your models. Here’s a quick rundown of what’s new:

:star2: Key Features and Improvements

  • Model Training Simplification: Training commands now use more specific model types (yolov8n-pose.pt, yolov8n-seg.pt), making it easier to select the correct model for your tasks.
  • Kaggle Integration Guide: A comprehensive guide to using Kaggle, complete with new features and troubleshooting sections, is now available. Leverage Kaggle’s free cloud computing resources for training and evaluating your models.
  • Enhanced Documentation: Improved annotation and result management documentation for FastSAM and SAM models, making it more accessible for new users.
  • Improved Result Handling: Better management of predicted results for various YOLO model types (detection, OBB, pose, and segmentation) with support for txt saving.
  • User Notifications: A new warning for incorrect mAP calculations when save_hybrid=True is enabled, ensuring more accurate evaluation processes.

:bar_chart: Detailed Changes

:tada: New Contributors

We are excited to welcome our new contributors:

:link: Full Changelog

For a complete list of changes, check out the full changelog.

We encourage you to try out the new release and share your feedback. Your input is invaluable in helping us improve Ultralytics YOLO. Thank you for being a part of our community!

Release URL: Ultralytics v8.2.59

Happy coding! :rocket:

2 Likes