Releases: ultralytics/ultralytics
v8.2.30 - Update publish.yml (#13464)
馃専 Summary
Ultralytics version 8.2.30 introduces several updates mainly focused on code refactoring, enhancements, and bug fixes.
馃搳 Key Changes
- Adjusted
publish.yml
workflow to refine Git configurations and add new release steps. - Fixed minor coding inconsistencies and redundancies in various files.
- Enhanced specific functions' readability and performance.
馃幆 Purpose & Impact
- Improved Release Process: The updated GitHub Actions workflow ensures smoother tagging and releasing directly from the pushes, streamlining the continuous integration pipeline.
- Code Quality Improvement: Minor fixes and reorganizations enhance overall code readability, making it easier for developers to understand and contribute.
- Enhanced Performance: Optimizations in data handling and function operations potentially improve runtime efficiency and reduce resource usage.
What's Changed
- Refactor Python code by @glenn-jocher in #13448
- Code Refactor for Speed and Readability by @pderrenger in #13450
ultralytics 8.2.30
automated tags and release notes by @IvorZhu331 in #13164
Full Changelog: v8.2.29...v8.2.30
v8.2.0 - YOLOv8-World and YOLOv9-C/E Models
Ultralytics v8.2.0 Release Notes
Introduction
Ultralytics is excited to announce the v8.2.0 release of YOLOv8, comprising 277 merged Pull Requests by 32 contributors since our last v8.1.0 release in January 2024, marking another milestone in our journey to make state-of-the-art AI accessible and powerful. This release brings a host of new features, performance optimizations, and expanded integrations, reflecting our commitment to continuous improvement and innovation. 馃實馃殌
Ultralytics v8.2.0 Key Highlights
- New Models: Introduced support for YOLOv8-World, YOLOv8-World-v2 (by @Laughing-q in PR #9268), YOLOv9-C, YOLOv9-E (by @Laughing-q in PR #8571), and YOLOv9 Segment models (by @Burhan-Q in PR #9296), expanding the versatility of the Ultralytics platform.
- New Features: Added distance calculation in vision-eye, per-class object counting (by @RizwanMunawar in PR #9443), and queue management utilities (by @RizwanMunawar in PR #9494), enhancing the functionality and applicability of YOLOv8.
- Performance Optimizations: Achieved 40% faster ultralytics imports (by @glenn-jocher in PR #9547), faster batch same_shapes, and immediate checkpoint serialization (by @glenn-jocher in PR #9437), further optimizing the efficiency of the framework.
- Enhanced Export Capabilities: Improved export support, including OpenVINO 2023.3 updates (by @adrianboguszewski in PR #8417), TensorRT 10 support (by @Burhan-Q in PR #9516), and fixes for TFLite, ONNX, and OpenVINO exports.
- Documentation Expansion: Significantly expanded the documentation with new guides, integration pages for TorchScript, TFLite, NCNN, PaddlePaddle, TF GraphDef, TF SavedModel, TF.js (by @abirami-vina in multiple PRs), and updates to existing pages, providing comprehensive resources for users.
- Training Enhancements: Introduced YOLO-World training support (by @Laughing-q in PR #9268), fixed learning rate issues (by @Laughing-q in PR #9468), and improved robustness for stopping and resuming training (by @glenn-jocher in PR #9384).
- Platform Support: Added support for NVIDIA Jetson (by @lakshanthad in PR #9484), Raspberry Pi (by @lakshanthad in PR #8828), and Apple M1 runners for tests and benchmarks (by @glenn-jocher in PR #8162), expanding the usability of YOLOv8 across various platforms.
- CI/CD Improvements: Enhanced Ultralytics Actions using OpenAI GPT-4 for PR summaries (by @pderrenger in PR #7867) and introduced self-hosted Raspberry Pi 5 CI (by @lakshanthad in PR #8828), streamlining the development and testing processes.
- Bug Fixes: Resolved various issues related to model loading, inference, plotting, and exports, ensuring a smoother user experience.
- Community Contributions: Welcomed contributions from 31 new contributors, reflecting the growing engagement and collaborative spirit within the Ultralytics community.
Summary
Ultralytics v8.2.0 represents a significant leap forward, introducing new models, features, and optimizations while expanding platform support and integration capabilities. We extend our gratitude to our dedicated users and contributors for their invaluable support and contributions. As we continue to push the boundaries of AI and computer vision, we look forward to the exciting possibilities and advancements that lie ahead! 馃専馃殌馃帀
What's Changed
- YOLOv8.1 blog, Explorer notebook and 2023 > 2024 updates by @AyushExel in #7469
- Explorer with LanceDB, Actions and Docs updates by @glenn-jocher in #7487
- OBB Docs updates by @glenn-jocher in #7512
- Update OpenVINO INT8 export by @glenn-jocher in #7515
ultralytics 8.1.1
Docs, Solutions and Export updates by @glenn-jocher in #7545- Update HTTP to HTTPS by @glenn-jocher in #7548
- Python refactorings and simplifications by @glenn-jocher in #7549
- Use
pathlib
in DOTA ops by @glenn-jocher in #7552 - OBB Docs updates by @glenn-jocher in #7568
- Update YOLOv3 and YOLOv5 YAMLs by @glenn-jocher in #7574
- Add docstrings to new HUB functions by @glenn-jocher in #7576
- OBB: Fix plot_images by @Laughing-q in #7592
- OBB: update metrics by @Laughing-q in #7593
- Resize angle, count, and stage on keypoint number change by @gvzdv in #7598
- Mkdocs annotations fixes by @glenn-jocher in #7600
ultralytics 8.1.2
scope HUB-SDK imports by @glenn-jocher in #7596- Update docs building code by @glenn-jocher in #7601
- YAML reformat by @glenn-jocher in #7669
- Add PR Summary step to Ultralytics Actions by @glenn-jocher in #7675
- Fixed dataloader CPU bottleneck for small batch sizes by @ExtReMLapin in #7659
- Update
mkdocs.yml
by @glenn-jocher in #7693 ultralytics 8.1.3
ResNet models and lighter dependencies by @glenn-jocher in #7700- Update Twitter icon in Docs by @glenn-jocher in #7711
ultralytics 8.1.4
RTDETR TensorBoard graph visualization fix by @glenn-jocher in #7725- Update Docs robots.txt by @glenn-jocher in #7728
- Bounding Box to OBB conversion by @Burhan-Q in #7572
- Add
yolo_bbox2segment
docs reference by @glenn-jocher in #7751 ultralytics 8.1.5
add OBB Tracking support by @Laughing-q in #7731- Clean up unused
imgsz
by @Laughing-q in #7771 - Add HUB-SDK docs by @glenn-jocher in #7775
- Add OBB benchmarks to CI by @glenn-jocher in #7777
- Add YOLOv8-OBB https://youtu.be/Z7Z9pHF8wJc by @glenn-jocher in #7780
- Update H1 in
Explorer API
docs by @RizwanMunawar in #7813 - Adds toggle displaying labels in GUI and verbose log on start by @AyushExel in #7804
- Fix bbox2segment converter by @Laughing-q in #7814
- Add ONNX Docs integrations page by @abirami-vina in #7802
- Fix Yolo 8.0.206 scale bug by @Alarmod in #7821
ultralytics 8.1.6
revert 8.0.206 box ops box scaling by @glenn-jocher in #7823- Explorer API video https://youtu.be/3VryynorQeo by @RizwanMunawar in #7838
- Add HUB-SDK Docs reference section by @glenn-jocher in #7781
- Link checks SSL insecure robustness by @glenn-jocher in #7853
- Add new @Retry() decorator by @glenn-jocher in #7854
- Add TensorRT Docs Integrations Page by @abirami-vina in #7855
- Cleanup Docs languages by @glenn-jocher in #7865
- Add millimeters in
solutions/distance_caculation.py
+object-cropping.md
visuals by @RizwanMunawar in #7860 ultralytics 8.1.7
USER_CONFIG_DIR
Explorer ops by @AyushExel in #7861- Ultralytics Actions with OpenAI GPT-4 PR Summary by @pderrenger in #7867
- Bump slackapi/slack-github-action from 1.24.0 to 1.25.0 in /.github/workflows by @dependabot in https://github.com/ultralytics/ultralytics/pu...
v8.1.0 - YOLOv8 Oriented Bounding Boxes (OBB)
Ultralytics v8.1.0 Release Notes
Introduction
Ultralytics proudly announces the v8.1.0 release of YOLOv8, celebrating a year of remarkable achievements and advancements. This version continues our commitment to making AI technology accessible and powerful, reflected in our latest breakthroughs and improvements.
2023 in Review
- Record-Breaking Engagement: Over 20 million downloads of the Ultralytics package, with 4 million in December alone! 馃搱
- Massive Model Training: An incredible 19 million YOLOv8 models were trained in 2023, showing the widespread adoption and versatility of our platform. 馃寪
- Diverse Model Usage: 64% of these models were for object detection, 20% for instance segmentation, 15% for pose estimation, and 1% for image classification. 馃搳
- Expanding Global Reach: YOLOv8 reached 5 million users in 2023 and was run in 15 billion inference jobs across various industries, showcasing its real-world impact. 馃實
- Documentation in Multiple Languages: Our docs are now available in 11 languages, catering to our diverse global community. 馃摎
Ultralytics v8.1.0 Key Highlights
- YOLOv8 OBB Models: The introduction of Oriented Bounding Box models in YOLOv8 marks a significant step in object detection, especially for angled or rotated objects, enhancing accuracy and reducing background noise in various applications such as aerial imagery and text detection.
- Segmentation Support & Enhancements: Enhanced segmentation capabilities offer more precise image analysis, with improved classification augmentations integrated into Ultralytics training pipelines.
- Performance Optimizations: Since our initial release last year we've focused on optimizing every aspect of the YOLOv8 framework, including training, validation, inference, and export, ensuring speed and efficiency without compromising performance.
- Enhanced Model Architecture & Training Features: Incremental updates in model architecture, training features, and dataset support, including integration with Open Images V7 dataset and improved image classification models.
- API and CLI Improvements: Enhanced user experience with refined API and CLI, including the Ultralytics Explorer tool for advanced dataset exploration and interaction.
- PaddlePaddle, NCNN, PNNX, TensorRT & Other Integrations: Strengthened integration with multiple other platforms, offering users more deployment flexibility and compatibility for YOLOv8 users.
- Diverse Contributions & Ultralytics HUB Evolution: The integration of almost 1000 pull requests by 230 contributors and the growth of Ultralytics HUB, with it's own series of version updates, highlights the community's vital role in the development of YOLOv8.
Community Engagement and Support
- Expanding Documentation: Our documentation now spans 11 languages, with over 200 pages, providing comprehensive guides for various real-world applications.
- Custom-Trained YOLOv8 Models: With the ability to train models on custom data, 19 million YOLOv8 models were trained in 2023 alone, catering to diverse needs across object detection, segmentation, pose estimation, and image classification.
- User Contributions: We encourage and appreciate user-contributed examples and stories, showcasing the versatility and real-world impact of YOLOv8.
Summary
Ultralytics v8.1.0 is a testament to a year of innovation, with the integration of Oriented Object Detection, enhanced classification models, and a strong focus on user experience and community engagement. We thank our users and contributors for their invaluable support and look forward to another year of groundbreaking advancements in the field of AI and computer vision in 2024! 馃専馃殌馃帀
What's Changed
- Initial pip structure by @AyushExel in #1
- Update requirements.txt by @glenn-jocher in #9
- Normalize version with dev0 by @glenn-jocher in #11
- Create ci.yaml by @glenn-jocher in #10
- Update requirements.txt with
fire>=0.4.0
by @glenn-jocher in #12 - tensorflow-macos comment by @glenn-jocher in #14
- Add yolo package structure by @AyushExel in #15
- Create init.py by @glenn-jocher in #17
- Add missing setup.py fields by @glenn-jocher in #18
- Update ci.yaml by @glenn-jocher in #21
- Trainer + Dataloaders by @AyushExel in #27
- Metrics and loss structure by @AyushExel in #28
- Model builder by @AyushExel in #29
- Add initial model interface by @AyushExel in #30
- Smart Model loading by @AyushExel in #31
- Fix dataloader by @Laughing-q in #32
- Classify training cleanup by @glenn-jocher in #33
- Hydra *.yml extension deprecated fix by @glenn-jocher in #34
- Create dependabot.yml by @glenn-jocher in #38
- Bump actions/setup-python from 3 to 4 by @dependabot in #39
- Update ci.yaml by @glenn-jocher in #37
- Fix dataloader2 by @Laughing-q in #35
- Segmentation support & other enchancements by @AyushExel in #40
- Add is_colab() and is_kaggle() by @glenn-jocher in #41
- [rename] - preprocess-batch -> preprocess, preprocess_preds -> postprocess by @AyushExel in #42
- new check_dataset functions by @AyushExel in #43
- update model initialization design, supports custom data/num_classes by @AyushExel in #44
- Allow setting model attributes before training by @AyushExel in #45
- Fix some cuda training issues of segmentation by @Laughing-q in #46
- Logging fix from YOLOv5 by @glenn-jocher in #47
- General console printout updates by @glenn-jocher in #48
- Add EMA and model checkpointing by @AyushExel in #49
- Cli support by @AyushExel in #50
- Add clearml logging by @AyushExel in #51
- Add warmup and accumulation by @AyushExel in #52
- Deterministic training by @AyushExel in #53
- CLI updates by @AyushExel in #58
- update segment training by @Laughing-q in #57
- standalone val by @AyushExel in #56
- Detection support by @AyushExel in #60
- docs setup by @AyushExel in #61
- Create CITATION.cff by @glenn-jocher in #62
- Fix CITATION.cff typos by @glenn-jocher in #64
- add resuming by @Laughing-q in #63
- Flake8 updates by @glenn-jocher in #66
- Predictor support by @AyushExel in #65
- [WIP] Model interface by @AyushExel in #68
- General cleanup by @Laughing-q in #69
- Revert augment_hyps by @Laughing-q in #70
- Update docs by @AyushExel in #71
- Model enhancement by @AyushExel in #75
- update docs by @AyushExel in #76
- Change class depending on dataset in model interface by @AyushExel in #77
- add a naive DDP for model interface by @Laughing-q in #78
- Make config overrides user friendly by @AyushExel in #80
- Update docs by @AyushExel in #73
- Add v8 modules by @AyushExel in #81
- Update
cache_version = 1.0
by @glenn-jocher in https://github...