Announcing Ultralytics v8.2.66 Release!
Hello Ultralytics Community!
We are thrilled to announce the release of Ultralytics v8.2.66! This update brings a host of improvements and new features designed to enhance your experience and streamline your workflows. Here’s a quick rundown of what’s new:
Key Features and Improvements
Adjusted Dockerfile for ARM-based Architectures: Simplified package installations to make it easier for users on ARM-based systems, including Raspberry Pi.
Embedded YouTube Tutorial: We’ve added a YouTube tutorial to the SKU-110k dataset documentation to help you get started quickly.
Refined check_fileFunction: Improved handling and downloading of files for better efficiency and reliability.
Code Refactoring: Removed redundant functions and cleaned up the codebase for better readability and maintainability.
Purpose & Impact
- Simplified Workflows: Streamlined Dockerfile operations reduce complexity, making setup on ARM-based systems a breeze.
- Enhanced Learning Resources: Video tutorials integrated into the documentation provide a more comprehensive learning experience.
- Improved File Handling: The enhanced
check_filefunction ensures robust downloading and local file retrieval. - Cleaner Codebase: Refactoring and removing unused functions improve code readability and maintainability, benefiting developers.
What’s Changed
- Removed duplicate
make_divisiblefunction by @Burhan-Q in PR #14690 - Added YouTube tutorial to docs by @RizwanMunawar in PR #14698
- Updated
save_periodto include the first epoch by @Burhan-Q in PR #14700 - Added compatible
tensorstoreversions foraarch64by @lakshanthad in PR #14697 - Enabled
ultralytics 8.2.66HUB model autodownload by @glenn-jocher in PR #14702
For a detailed list of changes, check out the Full Changelog.
Try It Out!
We encourage you to try out the new release and let us know your thoughts. Your feedback is invaluable in helping us improve and deliver the best possible tools for your projects.
Release URL: Ultralytics v8.2.66
Happy coding!
The Ultralytics Team