Epoch GPU_mem box_loss cls_loss dfl_loss Instances Size
100/100 4.6G 0.2848 0.1288 0.9163 32 320: 8%|▊ | 62/795 [00:07<01:41Corrupt JPEG data: premature end of data segment
100/100 4.6G 0.2862 0.1292 0.9194 32 320: 9%|▉ | 71/795 [00:08<01:17Invalid SOS parameters for sequential JPEG
100/100 4.6G 0.285 0.1293 0.9134 32 320: 54%|█████▍ | 432/795 [01:00<00:3Invalid SOS parameters for sequential JPEG
100/100 4.61G 0.2887 0.1303 0.9145 10 320: 100%|██████████| 795/795 [01:52<00:0
Class Images Instances Box(P R mAP50 mAP50-95): 51%|█████ | 185/366 Invalid SOS parameters for sequential JPEG
Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 366/366
all 11682 23364 0.982 0.974 0.986 0.883
100 epochs completed in 5.306 hours.
Optimizer stripped from runs/detect/train6/weights/last.pt, 7.1MB
Optimizer stripped from runs/detect/train6/weights/best.pt, 7.1MB
Validating runs/detect/train6/weights/best.pt...
Ultralytics 8.3.127 🚀 Python-3.10.17 torch-2.7.0+cu126 CUDA:0 (NVIDIA L40S, 45589MiB)
Model summary (fused): 72 layers, 3,433,745 parameters, 0 gradients, 10.1 GFLOPs
Class Images Instances Box(P R mAP50 mAP50-95): 52%|█████▏ | 189/366 Invalid SOS parameters for sequential JPEG
Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 366/366
all 11682 23364 0.982 0.973 0.986 0.883
0 11682 11682 0.993 0.988 0.993 0.947
266753 11682 11682 0.972 0.959 0.978 0.818
Speed: 0.0ms preprocess, 0.2ms inference, 0.0ms loss, 0.5ms postprocess per image
Results saved to runs/detect/train6
💡 Learn more at https://docs.ultralytics.com/modes/train
