Ultralytics

🌟 Summary
The latest release, v8.3.6, primarily enhances system information logging by including detailed GPU data and various improvements in system monitoring and logging functionalities.

📊 Key Changes
- **Version Update:** The software version is now updated from 8.3.5 to 8.3.6.
- **Detailed GPU Information:** Inclusion of a `get_gpu_info` function to fetch GPU specifics.
- **Enhanced System Monitoring:** Increased comprehensiveness in the `collect_system_info` module to detail hardware like disk usage.
- **Improved Training Logs:** Logging training durations and performance metrics for clearer insights.
- **Workflow Updates:** Modified PyPI publication workflow to add trusted publishing and streamline tasks.
- **Solutions Update:** Refined solutions such as workout monitoring and heatmaps for easier integration and use.
- **Refactored Code:** Improved code readability, performance optimizations, and removed redundancies.

🎯 Purpose & Impact
- **Enhanced Troubleshooting:** By logging more hardware details, users can quickly diagnose and optimize performance issues. 💻
- **Improved System Efficiency:** Optimizing PyPI workflows and internal code makes the system more efficient and less prone to errors. 🚀
- **User Clarity:** Enhanced training and system logs provide users with better tracking of model performance and durations, aiding normalization and analysis. 📊
- **Simplified User Experience:** Updates to model solution setups allow users to easily integrate and use features without deep technical modifications. 💡
- **Security & Maintenance:** Improved package publishing security through the GitHub Actions workflow ensures safer deployments. 🔒

This release focuses on enhancing user experience, improving performance insights, and strengthening system robustness with better hardware integration and easier workflow for developers and users alike.

What's Changed
* Update PyPI to trusted publisher by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16709
* Update `workouts_monitoring` solution by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16706
* Ultralytics Code Refactor https://ultralytics.com/actions by pderrenger in https://github.com/ultralytics/ultralytics/pull/16712
* Update W&B docs. by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16708
* Update `heatmaps` solution by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16720
* `ultralytics 8.3.6` improve `collect_system_info` by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16722


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.5...v8.3.6

🌟 Summary
The latest update in version 8.3.5 primarily improves image caching by making it more robust and efficient through disk space validation and write permissions checks.

📊 Key Changes
- ⚙️ Version updated from 8.3.4 to 8.3.5.
- 🛠️ New function for checking disk cache `check_cache_disk`.
- ⚠️ Caching process now includes checks for disk space and write permissions, providing improved warnings if issues are detected.

🎯 Purpose & Impact
- 🚀 **Enhanced Reliability**: By ensuring the system has adequate disk space and permissions before caching, it reduces the likelihood of failures, which allows for more stable training processes.
- 💾 **Resource Efficiency**: Users will experience smoother and more deterministic training runs when disk resources are managed correctly.
- 🔄 **Deterministic Training**: Provides options for using disk caching to maintain consistency during model training, given that the hardware resources meet the necessary prerequisites.

What's Changed
* Update Docs extra.js by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16663
* Add combined model plot to Docs by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16669
* Remove unused arguments in Ultralytics `heatmaps` by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16667
* Add https://youtu.be/-JXwa-WlkU8 to docs by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16685
* Clarify `stream_buffer` argument docs by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16686
* Fix `NEW` labeling for more consistency in docs by jk4e in https://github.com/ultralytics/ultralytics/pull/16674
* Update MkDocs extra.js by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16691
* Zero mAP warning on YAML val by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16698
* Introduced `BaseSolution` class for Ultralytics solutions by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16671
* Fix and add missing infos about available CLI `TASK` commands in docs and code comments by jk4e in https://github.com/ultralytics/ultralytics/pull/16697
* Fix validator model=None by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16699
* Update docker.yaml to notify on failure once by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16700
* Fix UINT8 overflow for >255 segmentation masks by rajeshtims in https://github.com/ultralytics/ultralytics/pull/16690
* `ultralytics 8.3.5` add `cache=disk` space and writable checks by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16696

New Contributors
* rajeshtims made their first contribution in https://github.com/ultralytics/ultralytics/pull/16690

**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.4...v8.3.5

🌟 Summary
The Ultralytics v8.3.4 release introduces flexible updates to project dependencies and enhances several features for better usability and performance.

📊 Key Changes
- 🔗 **Numpy Dependency Update:** Removed the upper version limit on `numpy`, now supporting any version above 1.23.0.
- 🕒 **CI Schedule Adjustment:** Changed the daily build and test schedule to 08:00 UTC for better synchronization.
- 🚀 **Improved Example Documentation:** Added a detailed example for YOLOv8 using OpenVINO in C++.
- 🔢 **Enhanced Error Handling in Export Formats:** Introduced auto-correction mechanism for export format inputs.
- 🖥️ **Device Selection Robustness:** Improved parsing logic for device identifiers such as multiple GPUs.
- 📄 **Documentation and Video Guides:** Enhanced documentation with video tutorials on MobileSAM and SAM 2 models, and updated image links for better viewing.

🎯 Purpose & Impact
- **Compatibility & Flexibility:** Unpinning the `numpy` version boosts compatibility with newer versions, simplifying dependency management and reducing potential conflicts with other packages.
- **Optimized Workflow:** The change in CI schedule aligns better with daily resource availability and developer workflows, enhancing continuous integration efficiency.
- **User Experience:** Improved error handling during export ensures users receive guidance on available formats, minimizing confusion and errors.
- **Accessibility & Education:** Video tutorials and clearer documentation aid both new and experienced users in understanding and leveraging advanced model functionalities, fostering a wider user base.
- **Performance Optimization:** The example using OpenVINO can potentially enhance inference speed and support for various hardware profiles, such as Intel platforms.

What's Changed
* Fix Raspberry Pi CI by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16620
* Update scheduled CI to 08:00 UTC by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16622
* Add https://youtu.be/yXQPLMrNX2s to docs by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16626
* Fix overridden train_args when close_mosaic by Laughing-q in https://github.com/ultralytics/ultralytics/pull/16627
* Add contributor name for example: YOLOv8 OpenVINO CPP Inference by rlggyp in https://github.com/ultralytics/ultralytics/pull/16632
* Auto-correct for export formats by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16625
* Empty index `[0,,1]` robust device selection by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16631
* Improve `scikit-learn` support for exports by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16638
* Fixing predict docs by ambitious-octopus in https://github.com/ultralytics/ultralytics/pull/16637
* SAM: Fix labels not being used in predict mode. by Y-T-G in https://github.com/ultralytics/ultralytics/pull/16642
* Minor fix in docs for `python` admonition and code blocks by jk4e in https://github.com/ultralytics/ultralytics/pull/16646
* Update image URLs in `mobile-sam.md` and `sam-2.md`. by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16654
* Replace YOLO11-seg table with macro by jk4e in https://github.com/ultralytics/ultralytics/pull/16647
* `ultralytics 8.3.4` unpin `numpy<2.0.0` by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16655


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.3...v8.3.4

🌟 Summary
The 'v8.3.3' release of the Ultralytics project introduces critical updates, focusing on enhancing compatibility with YOLO11 models within various components and improving consistency across different codebases and documentation.

📊 Key Changes
- Updated the Streamlit app to support YOLO11, replacing the older YOLOv8 model selection.
- Standardized color parameter descriptions across the codebase with consistent RGB tuples.
- Enhanced zero-mask plotting to improve segmentation functionalities.
- Refined documentation to transition from YOLOv8 to YOLO11, ensuring coherent guidance and examples.
- Updated Jupyter and Colab notebook references to align with YOLO11 model paths and usage.
- Adjusted configs and error reporting settings for improved clarity and user experience.

🎯 Purpose & Impact
- **Model Selection & Compatibility**: Ensures users can seamlessly work with the latest YOLO11 models, facilitating access to enhanced features and improved performance.
- **Code & Documentation Consistency**: Streamlines user and developer interactions by providing clear, standardized information, which is crucial for debugging and development.
- **Performance & Robustness**: Enhancements in the segmentation plotting and documentation not only improve functionality but also increase the robustness of model training and inference tasks.
- **Improved User Experience**: By updating notebooks, examples, and templates, users are guided towards utilizing the most current and advanced model versions, minimizing confusion and errors.
- **Accessibility**: Make it easier for both new and existing users to adapt and transition to the latest model scenarios in different environments including cloud setups and GPU accelerated platforms.

What's Changed
* Fix zero-mask plotting bug by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16588
* Update YOLO11 docs by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16589
* YOLO11 Tasks, Modes, Usage, Macros and Solutions Updates by UltralyticsAssistant in https://github.com/ultralytics/ultralytics/pull/16593
* Update YOLO11 Actions and Docs by UltralyticsAssistant in https://github.com/ultralytics/ultralytics/pull/16596
* Fix Docs links by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16604
* Update YOLO11 notebooks by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16608
* Colab notebook update by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16609
* Update Settings by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16610
* Add YOLO11 macros tables by ambitious-octopus in https://github.com/ultralytics/ultralytics/pull/16598
* `ultralytics 8.3.3` update Streamlit app to YOLO11 by RizwanMunawar in https://github.com/ultralytics/ultralytics/pull/16590


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.2...v8.3.3

🌟 Summary
The **Ultralytics 8.3.2** release focuses on enhancing model performance through better Automatic Mixed Precision (AMP) checks and ensuring deterministic training environments, thereby improving consistency and speed in machine learning workflows.

📊 Key Changes
- Implemented new tests for AMP compatibility to enhance model training efficiency.
- Adjusted image size for AMP checks to refine accuracy.
- Updated environmental settings for more consistent and reproducible training results.
- Incremented software version from 8.3.1 to 8.3.2.

🎯 Purpose & Impact
- **AMP Testing Enhancement**: By verifying compatibility with AMP, models can now leverage faster, mixed-precision training. This potentially leads to significant speed improvements without sacrificing predictive accuracy. 🚀
- **Deterministic Training Setting**: Ensures that training outputs are consistent across different runs, which is crucial for debugging and result reliability in experiments. This leads to more predictable outcomes when retraining models. 🔄
- **Version Stability**: The update signifies boosts in functionality and stability, ensuring users have access to a more robust feature set. ⚙️

Additional changes in other merged PRs include enhancements for image display within Jupyter notebooks, improvements in documentation clarity, and simplification of model saving processes by removing the unnecessary 'dill' package option. These changes collectively aim to make the Ultralytics software more user-friendly and streamline machine learning tasks.

What's Changed
* Enable `result.show()` for Jupyter notebooks by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16573
* Fix button text in `docs/home` to match link by jk4e in https://github.com/ultralytics/ultralytics/pull/16575
* Remove `dill` package from Ultralytics by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16574
* `ultralytics 8.3.2` fix AMP checks with `imgsz=256` by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16583


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.1...v8.3.2

🌟 Summary
This update introduces enhancements primarily focused on transitioning model references from YOLOv8 to YOLO11. It includes fixes, documentation updates, and configuration changes to ensure the latest YOLO11 models are fully supported.

📊 Key Changes
- Updated the project version from 8.3.0 to 8.3.1.
- Corrected model references in Automatic Mixed Precision (AMP) checks for accurate YOLO model handling.
- Modified references and configurations across the codebase from YOLOv8 to YOLO11.
- Enhanced the Docker setup and benchmarking alignment for YOLO11 models.
- Improved documentation clarity with updated descriptions and keywords.

🎯 Purpose & Impact
- 🛠️ **Functionality**: Ensures AMP checks work correctly with YOLO11 models, critical for maintaining training accuracy and reliability.
- 🚀 **Model Upgrade**: Shifts the focus to YOLO11, reflecting improvements in performance and capability.
- 📄 **Documentation Improvement**: Updated documentation to reduce confusion and align with the latest version for easier user adoption.
- 📈 **Benchmarking and Deployment**: Improved Docker deployment and model benchmarking with YOLO11n, enhancing build reliability and evaluation precision.

What's Changed
* Fix YOLO11 YouTube link by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16542
* Update YOLO11 Docs page by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16543
* Bump contributor-assistant/github-action from 2.5.2 to 2.6.1 in /.github/workflows by dependabot[bot] in https://github.com/ultralytics/ultralytics/pull/16545
* Default to YOLO11 models by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16561
* Update Docker deploy to YOLO11 benchmarks by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16564
* `ultralytics 8.3.1` update AMP checks for YOLO11n by glenn-jocher in https://github.com/ultralytics/ultralytics/pull/16560


**Full Changelog**: https://github.com/ultralytics/ultralytics/compare/v8.3.0...v8.3.1