Table 6 The used hyperparameters for Swin Transformer and MobileNetV2 models.

From: Multi-label dental disorder diagnosis based on MobileNetV2 and swin transformer using bagging ensemble classifier

Hyperparameter

Swin Transformer

MobileNetV2

Learning rate

0.00001

0.0001

Optimizer

AdamW

Adam

Weight decay

0.01

0.0001

Batch size

32

32

Patch size

\(4\times 4\)

N/A

Window size

\(7\times 7\)

N/A

Back to article page