Table 1 Pipeline parameters tested using grid-search and 10-fold CV
Pipeline step | Parameter options |
|---|---|
Combine_fs | percentile = [5, 10, 20, 30, 40, 50] |
Lasso_fs | estimator = Logistic Regression |
penalty = “l1” | |
C=[5,10,20,30,40,50] | |
RFE_RF_fs | class_weight = ‘balanced’ |
n_estimators = 100 | |
step = [0,1 ] | |
n_features_to_select = [0.4,0.6,0.8] | |
Smote_fs | n_neighbors = [3,4,5] |
ratio=‘auto’ | |
kind=‘regular’ | |
k-NN | n_neighbors = [1,3,5,7,9,11] |
weights = [‘uniform’, ‘distance’] | |
LR | C = [0.00001, 0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 1, 5, 10, 15, 30] |
class_weight = [None, ‘balanced’] | |
penalty = [‘l1’, ‘l2’] | |
RF | n_estimators = [100,150,200,250,500] |
criterion = [‘entropy’,‘gini’] | |
max_depth = [‘None’,4,6] | |
class_weight = [None, ‘balanced’] | |
SVM | C = [0.01,0.1,0.5,1,5,10,15,30,50] |
gamma = [0.0001,0.001,0.01, 0.1,1,5] | |
kernel = ‘radial’ | |
class_weight = [None, ’balanced’] | |
NN | alpha = [1e −5, 0.00001, 0.0001, 0.001, 0.01,0.1,1,3,5,10] |
hidden_layer_sizes = [(30,), (50,), (70,), (100,), (150,), | |
(30,30),(50,50),(70,70),(100,100), | |
(30,30,30),(50,50,50),(70,70,70)] |