| | Model | Parameters | Best Parameters |
| | RF | (i) n_estimator = [50, 150, 200] | (i) n_estimator = 200 | | (ii) min_samples_split = [2, 5, 10] | (ii) min_samples_split = 5 | | (iii) min_samples_leaf = [1, 2, 4] | (iii) min_samples_leaf = 4 | | (iv) max_features = [auto, rbf] | (iv) max_features = auto | | | (v) PCA n_components = 7 |
| | SVM | (i) kernel = [linear, rbf] | (i) kernel = linear | | (ii) gamma = [0.001, 0.01, 0.1] | (ii) gamma = 0.001 | | (iii) C = [0.001, 0.01, 0.1] | (iii) C = 0.001 | | | (iv) PCA n_components = 2 |
| | NB | (i) var_smoothing = np.logspace (0, 9, num = 20) | (i) var_smoothing = 1.0 | | | (ii) PCA n_components = 2 |
| | KNN | (i) n_neighbors = [1–7] | (i) n_neighbors = 7 | | (ii) p = [1, 2, 5] | (ii) p = 1 | | | (iii) PCA n_components = 7 |
| | DNN | (i) activation function = [softplus, softmax, softsign, relu, tanh, sigmoid, hard_sigmoid, linear] | (i) activation function = relu | | (ii) batch_size = [10, 50, 100] | (ii) batch_size = 10 | | (iii) epochs = [10, 20, 40, 50, 80, 100] | (iii) epochs = 10 | | (iv) optimizer = [RMSprop, SGD, Adagrad, Adadelta, Adam, Adamax, Nadam] | (iv) optimizer = SGD | | (v) learn_rate = [0.001, 0.01, 0.1, 0.2, 0.3] | (v) learn_rate = 0.001 | | (vi) momentum = [0.0, 0.2, 0.4, 0.6, 0.8, 0.9] | (vi) momentum = 0.0 | | (vii) init_mode = [uniform, lecun_uniform, normal, zero, glorot_normal, glorot_uniform, he_normal, he_uniform] | (vii) init_mode = uniform | | (viii) dropout_rate = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9] | (viii) dropout_rate = 0.0 | | (ix) weight_constraint = [1–5] | (ix) weight_constraint = 1 |
|
|
