Skip to main content

Table 2 Parameter settings

From: LPI-HyADBS: a hybrid framework for lncRNA-protein interaction prediction integrating feature selection and classification

Method

Parameter setting

LPI-NRLMF

Cfix = 5, num_factors = 10, K1 = 5, max_iter = 100

Lambda_t = 0.625, alpha = 0.1, beta = 0.1

K2=5, theta = 1.0, lambda_d = 0.625

Capsule-LPI

EPOCH = 30, lr = 0.001, BATCH_SIZE = 100

LPI-CNNCP

Filters1 = 24, kernel_size1 = (49, 10)

Kernel_size2 = (64, 10), strides2 = (1, 3)

Strides1 = (1, 1), filters2 = 24

LPI-HyADBS

DNN: Adam(model.parameters(), lr = 0.0001),

Loss_fn=BCELoss(), batch = 128, epochs = 100

XGBoost: learning_rate = 0.1, n_estimators = 100

Objective =“binary:logistic”, max_depth = 6

C-SVM: kernel=“rbf”, gamma = “auto”,

Probability = True, colsample_btree = 0.8

\(\alpha =0.4\), \(\beta = 0.3\), \(\theta = 0.3\)

LPLNP

Neighbor_num = [6, 23, 100, num of lncRNA-100, 100],

Regulation = ’regulation2’, alpha = [0.5, 0.3, 0.7, 0.1, 0.9]