Table 5.

Performance measures Matrix of the binary classification (COVID-19 vs. Normal) and multiclass classification (COVID-19 vs. Normal vs. Pneumonia) experiments, based on VGG16 and ResNet50 architecture.

Folds	Experiment-1			Experiment-2			Experiment-3			Experiment-4
	Test	Test	Epochs	Test	Test	Epochs	Test	Test	Epochs	Test	Test	Epochs
	Loss	Accuracy(%)		Loss	Accuracy(%)		Loss	Accuracy(%)		Loss	Accuracy(%)
Fold-1	0.031	99.37	500	0.011	100	350	0.38	82.38	435	0.505	78.09	331
Fold-2	0.032	99.37	346	0.022	100	350	0.45	80.67	307	0.688	81.64	328
Fold-3	0.04	97.5	500	0.02	100	350	0.407	82.85	490	0.186	93.33	350
Fold-4	0.0001	100	339	0.007	100	350	0.259	90.95	313	0.211	93.33	350
Fold-5	0.046	99.37	500	0.054	98.12	350	0.117	96.87	500	0.105	96.19	350
	Average Accuracy  = 99.12			Average Accuracy  = 99.62			Average Accuracy  = 86.74			Average Accuracy  = 88.52