Table 4 Configuration of XGB, LGBM and Hybrid model.

From: Advanced and hybrid machine learning techniques for predicting compressive strength in palm oil fuel ash-modified concrete with SHAP analysis

Parameters

XGB

LGBM

Hybrid XGB-LGBM

Learning Rate

0.1

0.2

0.1

Max Depth

5

4

5

Number of Estimators:

100

100

100

Number of Leaves

-

31

31

Min Data in Leaf

-

10

10

Boosting Type

-

GBDT

GBDT

L1 Regularization

-

0.1

0.1

L2 Regularization

-

0.001

0.001

Subsample

-

0.5

0.6

Random State

42

42

42