Classification of Diabetic Retinopathy Severity in Fundus Images Using the Vision Transformer and Residual Attention

Table 1

The DR model training hyperparameters.

Hyperparameters	Valor
Optimizing function	SGD optimizer
Momentum	0.9
Weight decay	5 × 10^−4
Epochs	20
Batch size	32
Initial learning rate	1 × 10^−3
Dropout	0
Classifier	0.01
Number of classes	5 and 6 classes