Table 5 Evaluation and comparison of training data of different attention mechanisms with ECA, CBAM, GAM, NAM, and ELA.
From: Lightweight grape leaf disease recognition method based on transformer framework
Methods | Accuracy | Precision | Recall | F1 score | Flop (G) | Params (M) |
|---|---|---|---|---|---|---|
ECA | 97.85 | 97.86 | 97.90 | 97.86 | 0.491 | 1.051 |
CBAM | 98.15 | 98.25 | 98.18 | 98.15 | 0.491 | 1.054 |
GAM | 96.50 | 96.65 | 96.56 | 96.40 | 0.732 | 1.590 |
NAM | 98.17 | 98.23 | 98.21 | 98.18 | 0.491 | 1.051 |
ELA | 98.01 | 98.11 | 98.04 | 97.98 | 0.493 | 1.054 |
MELA | 98.48 | 98.48 | 98.47 | 98.46 | 0.493 | 1.054 |
