(yolo311) samsu@BI-260105 wbc-instance-segmentation % yolo segment train data=dataset.yaml model=yolo26n-seg.pt epochs=10 imgsz=640 device=mps
Epoch GPU_mem box_loss seg_loss cls_loss dfl_loss sem_loss Instances Size 10/10 6.44G 0.4074 0.4473 6.742 0.004419 3.635 2 640: 100% ━━━━━━━━━━━━ 16/16 7.0s/it 1:52 Class Images Instances Box(P R mAP50 mAP50-95) Mask(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 1/1 1.8s/it 1.8s all 28 32 0.768 0.576 0.707 0.684 0.768 0.576 0.707 0.684
10 epochs completed in 0.321 hours.Optimizer stripped from /Users/samsu/2026/dataset/WBC/cls_detect_seg/wbc-instance-segmentation/runs/segment/train3/weights/last.pt, 6.5MBOptimizer stripped from /Users/samsu/2026/dataset/WBC/cls_detect_seg/wbc-instance-segmentation/runs/segment/train3/weights/best.pt, 6.5MB
Validating /Users/samsu/2026/dataset/WBC/cls_detect_seg/wbc-instance-segmentation/runs/segment/train3/weights/best.pt...Ultralytics 8.4.19 🚀 Python-3.11.14 torch-2.11.0 MPS (Apple M5)YOLO26n-seg summary (fused): 139 layers, 2,690,444 parameters, 0 gradients, 9.0 GFLOPs Class Images Instances Box(P R mAP50 mAP50-95) Mask(P R mAP50 mAP50-95): 100% ━━━━━━━━━━━━ 1/1 1.7s/it 1.7s all 28 32 0.767 0.576 0.706 0.683 0.767 0.576 0.706 0.683 3 2 2 1 0 0.265 0.265 1 0 0.265 0.265 5 6 9 0.523 0.778 0.835 0.798 0.523 0.778 0.835 0.815 6 6 6 0.723 0.5 0.773 0.729 0.723 0.5 0.773 0.708 7 5 5 0.922 0.6 0.715 0.683 0.922 0.6 0.715 0.715 8 9 10 0.665 1 0.941 0.939 0.665 1 0.941 0.912Speed: 0.2ms preprocess, 13.4ms inference, 0.0ms loss, 5.9ms postprocess per imageResults saved to /Users/samsu/2026/dataset/WBC/cls_detect_seg/wbc-instance-segmentation/runs/segment/train3💡 Learn more at https://docs.ultralytics.com/modes/train