Methods | Parameter | Optimal value |
---|---|---|
DNN | Activation function | ReLu, sigmoid |
Learning rate | 0.01 | |
Number of hidden layer Neurons | 64,32,16 | |
Optimizer | Adam | |
Regularization L1 | 0.001 | |
Dense layers | 3 | |
Dropout | 0.25,0.5 | |
RF | n_estimators | 200 |
Random_state | 42 | |
Max features | Auto | |
Max_depth | 32 | |
Bootstrap | TRUE | |
min_samples_leaf | 4 | |
min_samples_split | 10 | |
XGB | n_estimators | 200 |
LEARNING rate | 0.001 | |
Max depth | 15 | |
reg_lambda | 2 | |
Objective function | Binary-logistic | |
Gamma | 1 | |
Booster | Gbtree | |
reg_alpha | 1 | |
ETC | Random_state | 42 |
n_estimators | 150 | |
Criterion | Entropy | |
Max_features | Sqrt |