Table 4 Configuration of different models.

From: Machine learning predictions of high-strength RCA concrete utilizing chemically activated fly ash and nano-silica

Parameters

Decision tree

Random forest

XG Boost

Ada Boost

Learning rate

-

0.1

0.1

Max depth

10

10

6

Number of estimators:

-

100

100

100

Number of leaves

32

32

31

31

Min data in leaf

20

20

20

L1 regularization

0.1

L2 regularization

0.1

Subsample

0.8

Random state

42

42

42

42