Table 1 Hyperparameter settings considered for the feedforward neural network
From: Prediction of synergistic drug combinations using PCA-initialized deep learning
Hyperparameter | Values considered |
|---|---|
Preprocessing | norm; norm+tanh; norm+tanh+norm |
Hidden units | [107, 107]; [54, 54]; [27]; [107,54]; [54,27]; [54, 54, 54]; [27]; [54, 27, 13]; [107, 54, 27] |
Learning rates | 10−2; 10−3; 10−4; 10− 5 |
Dropout | no dropout; input: 0.3, hidden: 0.5 |