Journal of Infrastructure Preservation and Resilience

Table 7 Search spaces of hyper-parameters during their tuning processes

From: Classification of failure modes of pipelines containing longitudinal surface cracks using mechanics-based and machine learning models

Algorithm	Hyper-parameter	Search space
NB	prior of leak	[0.001, 0.999]
SVM	C	[50, 500]
SVM	γ_G	[0.5, 2]
DT	min_samples_split	[2, 10]
	min_samples_leaf	[1, 5]
	max_depth	[1, 50]
RF	n_estimators	[10, 200]
	min_samples_split	[2, 10]
	min_samples_leaf	[1, 5]
	max_features	[1, 3]
	max_depth	[1, 50]
GB	n_estimators	[10, 300]
	min_samples_split	[2, 10]
	min_samples_leaf	[1, 5]
	max_depth	[1, 50]
	learning_rate	[0.01, 1]

Back to article page