Table 5 Class-specific performance-metrics table for all models.
From: Machine learning approaches overcome imbalanced clinical data for intraoral free flap monitoring
Model configuration | Class | Precision | Recall | F1 score |
|---|---|---|---|---|
Proposed (weighted Focal loss) | 0 | 0.99 | 1.00 | 0.99 |
1 | 0.95 | 0.83 | 0.89 | |
Focal loss without Weight | 0 | 0.98 | 1.00 | 0.99 |
1 | 0.94 | 0.67 | 0.78 | |
Weighted Cross-entropy loss | 0 | 0.99 | 0.99 | 0.99 |
1 | 0.86 | 0.79 | 0.83 | |
CE loss without Weight | 0 | 0.98 | 1.00 | 0.99 |
1 | 0.95 | 0.75 | 0.84 |
