Table 2 Performances and variance values of base CNN model in different numbers of training samples
From: Automatic glaucoma detection based on transfer induced attention network
Number of data | 100 | 200 | 400 | 800 | 1600 |
|---|---|---|---|---|---|
Accuracy (%) | 72.6 | 75.9 | 77.2 | 79.5 | 80.1 |
Variance \((10^{-5})\) | 578.34 | 113.23 | 43.21 | 26.67 | 15.32 |
Sensitivity (%) | 80.2 | 83.9 | 87.9 | 89.7 | 91.2 |
Variance \((10^{-5})\) | 478.43 | 101.30 | 53.98 | 24.67 | 19.92 |
Specificity (%) | 70.2 | 72.5 | 74.3 | 75.9 | 76.8 |
Variance \((10^{-5})\) | 438.55 | 112.23 | 48.34 | 25.61 | 18.45 |