Skin Cancer Segmentation and Classification Using Vision Transformer for Automatic Analysis in Dermatoscopy-Based Noninvasive Digital System

Table 5

Accuracy and loss scores of different ViT classifiers.

Model	Training accuracy	Training loss	Validation accuracy	Validation loss	Test accuracy
ViT-Google	0.99375	0.40280	0.96275	0.50341	0.96150
ViT-MAE	0.96975	0.77760	0.92500	0.96414	0.92150
ViT-ResNet50	0.87175	2.45290	0.81000	2.45290	0.80900
ViT-VAN	0.95825	0.95315	0.92500	1.04780	0.92400
ViT-BEiT	0.92500	1.49015	0.87500	1.89523	0.86250
ViT-DiT	0.86175	1.54550	0.82500	0.82500	0.81900