Accurate digital quantification of tau pathology in progressive supranuclear palsy

Table 3 Selected parameter from hyper-parameter tuning using stratified tenfold cross validation for the screening and region-specific tau classifiers

Parameter	Screening	Cortical	Putamen	STN & GP	Dentate nucleus
N_features_to_select	46	40	34	34	34
Sampling strategy	‘auto’	‘not majority’	‘not majority’	‘not majority’	‘not majority’
n_estimator	600	800	500	500	100
min_sample_split	2	2	2	2	2
min_sample_leaf	2	1	2	2	1
max_features	1	0.2	0.6	0.6	0.2
max_depth	None	10	15	15	None
max_sample	None	0.75	0.75	0.75	None

Hyperparameters determine machine learning model architecture and are chosen before training. Hyperparameter tuning, which is part of training, is carried out to search for an optimal set of model parameters. STN Subthalamic nucleus; GP Globus pallidus

ISSN: 2051-5960