Table 5 Comparative result based on different evaluation metrics.
From: Minimal sourced and lightweight federated transfer learning models for skin cancer detection
Algorithm | Accuracy | Validation accuracy | Precision | Validation precision | Recall | Validation recall | Loss | Validation loss |
|---|---|---|---|---|---|---|---|---|
EfficientNetB3 | 96.81 | 89.22 | 97.09 | 89.69 | 96.40 | 88.89 | 0.20 | 0.53 |
EfficientNetV2S | 96.78 | 91.67 | 97.05 | 92.36 | 96.32 | 91.33 | 0.19 | 0.41 |
MobileNetV2 | 95.07 | 88.44 | 95.62 | 89.71 | 94.36 | 87.22 | 0.27 | 0.56 |
NasNetMobile | 96.29 | 85.56 | 96.82 | 87.09 | 95.67 | 84.67 | 0.22 | 0.72 |
ResNet50 | 96.83 | 87.67 | 97.01 | 89.14 | 96.43 | 86.67 | 0.20 | 0.51 |
