Table 1. Tuning parameters used for each model in implementing hyperparameter tuning. Some of these tuning parameters are lifted from Nanda & Dutta (2023), Mantovani et al. (2015), Probst, Boulesteix & Bischl (2019).

Classifiers	Tuning Parameters
SVM	C: uniform ( $2^{-3},2^{15}$)
	gamma: auto or scale
	kernel: rbf
RF	n_estimators: 1–350
	max_depth: 1–5
	min_samples_split: 1–10
KNN	n_neighbors: 1–500
LDA	solver: svd or lsqr or eigen
	shrinkage: uniform (0, 1)
CART	max_depth: 1–30
	min_samples_leaf: 1–60
	min_samples_split: 1–60
PHCA	homology_dimension: $H_0$ only, $H_1$ only, $H_0$ and $H_1$