Application of Extreme Gradient Boosting Based on Grey Relation Analysis for Prediction of Compressive Strength of Concrete

Table 5

Setting parameters of ANN, GA-ANN, and XGBoost.

Structural parameters	ANNs	GA-ANNs	Hyperparameters	XGBoost
Input layer nodes	6	6	Maximum depth	6
Hidden layers	2	2	Minimum child weight	1
Hidden layer nodes	80 (first) 60 (second)	80 (first) 60 (second)	Gamma	0
Output layer nodes	1	1	Subsample	1
Training parameters			Colsample by tree	1
Epoch times	10	10	Regularization alpha	0
Initial learning rate	0.01	0.01	Initial learning rate	0.01
Maximum iterations	100	100	Number of iterations	30