Table 7 The optimal parameter values of the model.

From: Data-driven prediction on critical mechanical properties of engineered cementitious composites based on machine learning

Model

Parameter

Definition

Value

KNN

n_neighbors

Number of nearest neighbors

3

RF

n_estimators

Number of decision trees

13

max_depth

Maximum depth of each decision tree

9

max_features

Maximum number of features

6

XGBoost

n_estimators

Number of decision trees

24

min_child_weight

Minimum sum of instance weight

4

gamma

Minimum amount of loss required for further branching

0.3