Table 2 Influence of LR and activation functions on the accuracy of the DNN model using a tenfold model
From: Deep-ProBind: binding protein prediction with transformer-based deep learning model
LR | Tanh | Sigmoid | ReLU |
|---|---|---|---|
0.08 | 92.43 | 91.03 | 89.14 |
0.09 | 92.65 | 91.81 | 89.25 |
0.1 | 92.67 | 91.93 | 89.54 |
0.2 | 92.32 | 91.31 | 89.01 |
0.3 | 91.01 | 90.25 | 88.48 |
0.4 | 90.34 | 89.82 | 87.95 |
0.5 | 89.51 | 88.43 | 87.42 |