Identification of relevant features using SEQENS to improve supervised machine learning models predicting AML treatment outcome

Table 7 Hyperparameter search spaces for evaluated models

Model	Parameter	Search space
XGBoost	booster	[gbtree]
	objective	[binary:logistic]
	n_estimators	50 — 150
	learning_rate	0.001 — 1
	max_depth	1 — 10
	grow_policy	[depthwise, lossguide]
	subsample	0.8 — 1
	colsample_bytree	0.5 — 1
	colsample_bylevel	0.5 — 1
	colsample_bynode	0.5 — 1
	reg_alpha	0.0001 — 1
	reg_lambda	0.0001 — 1
MultilayerPerceptron	n_layers	1 — 10
	n_neurons/layer	5 — 50
	activation	[tanh, relu]
	solver	[sgd, adam]
	alpha	[0.0001, 1]
	batch_size	10 — 100
	learning_rate	[constant, invscaling, adaptive]
	learning_rate_init	0.001 — 0.1
	max_iter	100 — 300
LogisticRegression	C	0.001 — 100
	penalty	[l1, l2, elasticnet]
	solver	[lbfgs, liblinear, saga]
	l1_ratio	0 — 1
	max_iter	100 — 300

ISSN: 1472-6947