Table 2.

Optimized values of hyperparameters for ML models.

ML model	Hyperparameters	Final optimized value	aTotal execution time (ms)
ANN	kernel_initializer	“normal”	34.76
	model optimizer	“adam”
	hidden layer 1 number of neurons	62
	hidden layer 1 activation function	“exponential"
	hidden layer 2 number of neurons	32
	hidden layer 2 activation function	“exponential"
	hidden layer 3 number of neurons	16
	hidden layer 3 activation function	“exponential"
	output layer number of neurons	1
	output layer activation function	“linear"
SVM	C	700	6.38
	ε	0.001
	ϒ	6 × 10−3
	kernel	“RBF”
	random_state	100
ELM	hidden_units	50	14.74
	activation_function	“relu”
	C	1
	random_type	“normal"
KRR	α	0.001	368.08
	kernel	“RBF”
	ϒ	1.78 × 10−2
	random_state	100
XGB	learning_rate	0.3	510.15
	n_estimators	200
	ϒ	0.001
	max_depth	50
	subsample	0.5
	colsample_bylevel	0.7
	random_state	100
RF	n_estimators	100	50.00
	max_features	17
	max_depth	None
	min_samples_split	2
	min_samples_leaf	1
	bootstrap	false

^aTotal execution time = model training time + inference time.