Table 4:

Classifier hyperparameters which are optimised using nested k-fold cross-validation.

Hyperparameter	Classifier	Range
Regularisation Strength (ν1)	LR	10−7 to 107
l1 penalty (ν2)	LR	0 to 1 in steps of 0.05
l2 penalty (ν3)	LR	0 to 1 in steps of 0.05
No. of Neighbours (χ1)	KNN	10 to 100 in steps of 10
Leaf size (χ2)	KNN	5 to 30 in steps of 5
Regularisation Strength (ζ1)	SVM	10−7 to 107
RBF kernel coefficient (ζ2)	SVM	10−7 to 107
No. of hidden neurons (ξ1)	MLP	10 to 100 in steps of 10
l2 penalty (ξ2)	MLP	10−7 to 105
Stochastic gradient descent learning rate (ξ3)	MLP	0 to 1 in steps of 0.05
No. of Convolutional Layers (α1)	CNN	2k where k = 4, 5, 6
Dropout Rate (α2)	CNN	0.1 to 0.5 in steps 0.2
Batch Size (α3)	CNN	2k where k = 6, 7, 8