Ultralytics

🌟 Summary
The `v8.2.97` update focuses on enhancing the model loading process in the Ultralytics HUB, improving security, and making interactions more user-friendly.

📊 Key Changes
- **Secure Downloads**: Model weights are now downloaded using secure, authenticated URLs.
- **Organized Storage**: Introduced checks to ensure model weights download to the correct directory.
- **New 'Logout' Command**: Added a logout command in the configurations for improved session management.

🎯 Purpose & Impact
- 🛡️ **Enhanced Security**: Utilizing secure URLs protects user data and boosts confidence in downloading model weights safely.
- 📂 **Improved File Management**: Ensures model files are neatly organized, making them easier to locate and manage.
- 🚀 **Increased Reliability**: Reduces download errors and enhances the overall robustness of model loading, improving user experience especially during setup or updates.

What's Changed
* Docs banner for https://ultralytics.com/yolo-vision by sergiuwaxmann in https://github.com/ultralytics/ultralytics/pull/16338
* Add https://youtu.be/c5eFrt2KuzY to docs by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16341
* `ultralytics 8.2.97` robust HUB model downloads by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16347


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.96...v8.2.97

🌟 Summary
The `v8.2.96` release of Ultralytics enhances data export capabilities, refines parking management solutions, and improves documentation workflows, offering better usability and integration for users.

📊 Key Changes
- **Data Export Methods**: Introduced new methods `to_df()`, `to_csv()`, and `to_xml()` to export results in DataFrame, CSV, and XML formats.
- **Parking Management Optimization**: Simplified the code and refactored data handling for easier setup and enhanced performance.
- **Documentation and Streaming Updates**: Improved documentation process and clarified streaming capabilities, including renaming and restructured examples for single and multi-stream video processing.
- **Precision and Validation Enhancements**: Aligned model validation precision with Automatic Mixed Precision settings for consistency.

🎯 Purpose & Impact
- **Enhanced Exportability**: Users can now easily export detection results in widely-used formats, facilitating better data analysis and integration with other tools. 📊
- **Improved Clarity and Efficiency**: The refactoring of parking management code enhances usability and processing speed, making it more intuitive for users managing parking data. 🚗
- **Streamlined Documentation Workflow**: Improved the process of auto-fixing and updating documentation, leading to more accurate and accessible developer resources. 📚
- **Consistent Precision Handling**: Ensures that validation and training are better aligned, leading to more reliable performance assessments and optimized use of computational resources. ⚙️

What's Changed
* Disable FP16 val on AMP fail and improve AMP checks by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16306
* Optimize `parking management` solution by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16288
* Enable Docs auto-fixes on repo branches by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16326
* Update Multi-Stream predict docs by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16334
* Use `trainer.amp` to determine FP16 validation by Laughing-q in https://github.com/ultralytics/ultralytics/pull/16333
* `ultralytics 8.2.96` new `results[0].to_df` Pandas, XML and CSV methods by MatthewNoyce in https://github.com/ultralytics/ultralytics/pull/16267


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.95...v8.2.96

🌟 Summary
The `v8.2.95` release primarily focuses on enhancing the YOLOv8 object tracking capabilities and improving model checkpoint flexibility to boost overall performance and user adaptability.

📊 Key Changes
- **Efficient Object Tracking**: Enhancements made to the object tracking framework with optimizations for threading and video processing.
- **Checkpoint Update Flexibility**: Introduced the ability to overlay updates on model checkpoints with the `strip_optimizer` function.
- **Version Update**: Incremented the software version from 8.2.94 to 8.2.95.

🎯 Purpose & Impact
- 🚀 **Improved Performance**: Users dealing with multiple video streams will experience a more efficient object tracking process, enhancing real-time data processing capabilities.
- 🔧 **Enhanced Flexibility**: Allowing optional updates on checkpoints enables more dynamic model deployment and easier fine-tuning for customized applications.
- 📈 **Routine Improvement**: The version update signifies a more stable, feature-rich platform that informs users of the most recent improvements and bug fixes.

In summary, this release is tailored towards making object tracking more seamless and providing users with greater flexibility in managing their model checkpoints, which helps in both development and deployment phases.

What's Changed
* Fix `IS_TMP_WRITEABLE` order of operations by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16294
* Fix dependabot in mkdocs_github_authors.yaml by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16312
* Threaded inference docs improvements by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16313
* `ultralytics 8.2.95` faster checkpoint saving by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16311


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.94...v8.2.95

🌟 Summary
This release enhances support for Apple's MPS devices and includes various improvements for better performance and usability across platforms.

📊 Key Changes
- Enhanced memory logging for Apple's MPS on macOS, ensuring accurate GPU memory usage reporting.
- Improved prediction handling in the model, including saving predictions and consistent bounding box handling.
- Updated documentation for seamless navigation and improved clarity.
- Fixed issues with `mps.empty_cache()` for macOS without MPS, preventing crashes.
- Introduced new benchmarks for Intel's latest hardware, providing key performance metrics.

🎯 Purpose & Impact
- 🖥️ **macOS Improvements**: Accurate memory reporting and management enhance training and inference performance on macOS, especially for Apple hardware users.
- 📈 **Performance Benchmarks**: Intel users gain insights into the performance of new hardware, helping them optimize their setups.
- 📄 **Documentation and Usability**: Updated docs enhance user experience by making it easier to find information and contribute, fostering community growth.
- 📐 **Enhanced Prediction Handling**: Saving predictions and improving mask-less bounding box handling enhance the output process, making models more robust and user-friendly.
- 🚀 **Optimized Platform Support**: The changes ensure smoother operation across different environments, helping all users maximize the capabilities of Ultralytics models.

What's Changed
* Return boxes for SAM prompts inference by Laughing-q in https://github.com/ultralytics/ultralytics/pull/16276
* Docs improvements and redirect fixes by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16287
* Fix `mps.empty_cache()` for macOS without MPS by Skillnoob in https://github.com/ultralytics/ultralytics/pull/16280
* Add color palette tables to docs by jk4e in https://github.com/ultralytics/ultralytics/pull/16242
* Intel Core Ultra benchmarks by ambitious-octopus in https://github.com/ultralytics/ultralytics/pull/15895
* `ultralytics 8.2.94` Apple MPS train memory display by Oil3 in https://github.com/ultralytics/ultralytics/pull/16272


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.93...v8.2.94

🌟 Summary
The `v8.2.93` update introduces enhanced model loading security with the addition of new SafeClass and SafeUnpickler classes.

📊 Key Changes
- **Safe Model Loading**: Implemented secure loading mechanisms with `SafeClass` and `SafeUnpickler` to safeguard against unknown classes.
- **Version Update**: Incremented to version 8.2.93.
- **Documentation and Workflow Improvements**: Updated documentation for multiple processes and simplified GitHub workflow interactions.
- **Dependency Update**: Upgraded `onnxslim` to version `0.1.34` for improved export functionality.
- **Code Optimization**: Streamlined code for speed estimation and queue management, removing unnecessary parameters.
- **NMS Flexibility**: Enabled agnostic non-maximum suppression (NMS) in various validation processes.

🎯 Purpose & Impact
- **Increased Security**: Enhances safety during model loading by preventing the execution of unverified code, which is crucial for maintaining system integrity. 🔒
- **User Experience**: Simplifies onboarding and interactions for developers contributing to the project. 🤝
- **Improved Compatibility**: The updated `onnxslim` version ensures better model export compatibility and performance optimization. ⚙️
- **Efficiency and Clarity**: Refactored code improves performance, reduces complexity, and makes the system more maintainable and user-friendly. 🏎️
- **Enhanced Model Handling**: Adding robust mechanisms for model loading and validation ensures sustained backward compatibility while offering advanced functionality. 🛡️

What's Changed
* Deprecate .github/workflows/greetings.yml by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16245
* Update format.yml by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16255
* Add discussions to format.yml by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16257
* Fix inaccuracies in OBB docs by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16253
* Update Tracker Docs by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16262
* Add https://youtu.be/GIrFEoR5PoU to docs by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16264
* Update `onnxslim==0.1.34` by inisis in https://github.com/ultralytics/ultralytics/pull/16265
* Optimize `speed estimation` solution by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16254
* Allow agnostic NMS in validation for OBB, Pose, Segment and NAS by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16251
* Optimize `queue` solution by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16246
* `ultralytics 8.2.93` new SafeClass and SafeUnpickler classes by UltralyticsAssistant in https://github.com/ultralytics/ultralytics/pull/16269


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.92...v8.2.93

🌟 Summary
This release enhances the object counting functionality in Ultralytics with configurable counting directions, and improves code efficiency and readability.

📊 Key Changes
- Added a new parameter to set object counting directions ("left_to_right" or "right_to_left").
- Updated counting logic to leverage the new direction attribute.
- Refined object counter visualization and class labeling.
- Minor code clean-ups, including removal of redundant components.

🎯 Purpose & Impact
- 🔄 **Flexibility**: Users can now customize the counting direction, making the feature more versatile across different scenarios.
- 🎯 **Improved Accuracy**: The directionality consideration minimizes miscounts, particularly in bidirectional flows.
- 🧹 **Code Clarity**: Streamlined codebase for better maintainability and readability, benefiting developers when further customizing or understanding the code.

What's Changed
* Update merge-main-into-prs.yml by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16199
* Distance calculation docs fix by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16220
* Update more/missing type qualifiers to lowercase MkDocs admonitions by jk4e in https://github.com/ultralytics/ultralytics/pull/16215
* Add https://youtu.be/Tt_35YnQ9uk to docs by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16214
* Update TwoWayTransformer Docs. by JasonG98 in https://github.com/ultralytics/ultralytics/pull/16161
* Fixed failing greetings CI by ambitious-octopus in https://github.com/ultralytics/ultralytics/pull/16223
* `bbox` and `masks` coloring based on `track-ids` by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16073
* Non-Deterministic Training Despite deterministic=True by ambitious-octopus in https://github.com/ultralytics/ultralytics/pull/15961
* `ultralytics 8.2.92` vertical `line` counter fix by CharanPrasadK in https://github.com/ultralytics/ultralytics/pull/16163

New Contributors
* JasonG98 made their first contribution in https://github.com/ultralytics/ultralytics/pull/16161

**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.2.91...v8.2.92