Table 2 Table of changes in each loss function on the training and validation sets.
From: Research on object detection and recognition in remote sensing images based on YOLOv11
Box_loss | Cls_loss | Dfl_loss | ||||
|---|---|---|---|---|---|---|
Train | Valid | Train | Valid | Train | Valid | |
Initial value | 1.0728 | 1.2353 | 1.0374 | 0.8327 | 1.1578 | 1.1451 |
Maximal value | 1.2780 (round 4) | 1.3909 (round 4) | 1.1839 (round 3) | 1.2929 (round 3) | 1.2409 (round 4) | 1.2827 (round 4) |
Final value | 0.6272 | 0.8465 | 0.3730 | 0.4169 | 0.6861 | 0.7418 |
Rate of decline | 41.53% | 31.47% | 64.05% | 49.93% | 40.74% | 35.22% |
