ultralytics 8.4.95

v8.4.95 - Fix restricted loading of training checkpoints (#25154)

on Python PyPI

🌟 Summary

v8.4.95 improves training reliability, RT-DETR detection limits, input compatibility, and dataset usability—headlined by safer checkpoint loading for Platform GPU jobs. 🚀

📊 Key Changes

• 🔒 Fixed restricted training checkpoint loading — Raw checkpoints produced by Platform GPU workers no longer include training-only loss objects or NumPy scalar metadata that can be rejected by secure weights_only=True loading.
• 🎯 Improved RT-DETR max_det support — The detection limit is now correctly applied during native prediction, validation, and model exports, including CoreML and other deployment formats.
• 🖼️ Expanded NumPy input compatibility — NumPy images with grayscale, gray-plus-alpha, BGR, or BGRA channels are now normalized to match the model’s expected channel count.
• 🔢 Fixed uint8 tensor preprocessing — Integer image tensors are handled without invalid floating-point dtype operations, while existing floating-point normalization behavior is preserved.
• 📦 Reduced automatic COCO downloads — COCO detection, segmentation, and pose datasets no longer download the optional 7 GB test2017.zip by default. Standard downloads are now approximately 20.2–20.3 GB instead of about 27 GB.
• 🧪 Made segmentation conversion deterministic — Mask files are processed in sorted filename order, improving repeatability across runs.
• 📈 Corrected pose metric curve labels — Removed duplicate box labels so pose metrics align correctly with dashboards such as Weights & Biases.
• 📝 Clarified prediction and embedding return types — Documentation and type hints now accurately indicate that streaming returns an iterator, while non-streaming calls return lists.
• 🛠️ Simplified edge-case handling — Empty segmentation inputs now return a valid zero box immediately.
• 📚 Updated examples and training documentation — The ObjectCounter example now uses its returned results object, and automatic dataset download behavior is described more precisely.

🎯 Purpose & Impact

• More dependable cloud training: Platform GPU workers can save, resume, and load raw training checkpoints more reliably, reducing avoidable job failures. ✅
• More predictable deployment: RT-DETR exports and validation now respect max_det, helping control output size, latency, and downstream processing requirements.
• Fewer input-related errors: Applications using NumPy arrays or uint8 tensors can handle a wider range of camera and image formats without manual channel conversion.
• Faster, lighter dataset setup: Most COCO users save roughly 7 GB of storage and download time because test images without public ground truth are excluded.
• More reproducible data preparation: Deterministic mask ordering makes conversion results and debugging more consistent.
• Clearer analytics: Correct pose labels prevent mislabeled or missing metric panels in experiment tracking tools.
• No major model architecture change: This release focuses on reliability, compatibility, export behavior, and documentation rather than introducing a new model family. YOLO26 remains the latest recommended Ultralytics model for general use. 🌟

What's Changed

• Stop coco.yaml and coco-pose.yaml from downloading test2017.zip by default by @raimbekovm in #25136
• Use ObjectCounter results in the usage example by @parthivdholaria in #23417
• Process segmentation masks in deterministic order by @AlexRTer in #24390
• Simplify empty segment handling in segment2box by @1548093028 in #23770
• Normalize uint8 tensor sources without dtype errors by @jahsef in #23743
• Clarify automatic dataset downloads in training docs by @vedantparnaik in #24155
• Clarify streamed prediction return types by @thiliapr in #24391
• Clarify streamed embedding return types by @waketzheng in #24134
• Fix PoseMetrics.curves returning duplicated box labels by @JESUSROYETH in #25148
• Normalize NumPy channel count to the model in LoadPilAndNumpy by @JESUSROYETH in #25151
• fix(rtdetr): honor max_det in decoder exports and validation by @Hasnaathussain in #25153
• Fix restricted loading of training checkpoints by @glenn-jocher in #25154

New Contributors

• @thiliapr made their first contribution in #24391
• @parthivdholaria made their first contribution in #23417
• @jahsef made their first contribution in #23743
• @vedantparnaik made their first contribution in #24155
• @1548093028 made their first contribution in #23770
• @AlexRTer made their first contribution in #24390
• @waketzheng made their first contribution in #24134
• @Hasnaathussain made their first contribution in #25153

Full Changelog: v8.4.94...v8.4.95