Table 6. Classification accuracy obtained using combined deep features (Pool5 of Resnet18 + Conv5 of Densenet121).
Metrics | LDA | kNN | GNB | SVM | AdaBoost | RF | Ensemble | XGBoost | NN |
---|---|---|---|---|---|---|---|---|---|
Accuracy | 0.837 | 0.862 | 0.810 | 0.826 | 0.848 | 0.865 | 0.888 | 0.858 | 0.912 |
Precision | 0.852 | 0.865 | 0.940 | 0.892 | 0.865 | 0.876 | 0.891 | 0.875 | 0.926 |
Recall | 0.837 | 0.862 | 0.810 | 0.826 | 0.848 | 0.865 | 0.888 | 0.858 | 0.919 |
F1 Score | 0.834 | 0.862 | 0.811 | 0.833 | 0.837 | 0.865 | 0.889 | 0.848 | 0.910 |