Table 2 Summary model configurations for Teacher, Student model, and Distiller.
From: Knowledge distillation-based lightweight MobileNet model for diabetic retinopathy classification
Parameter | Values |
|---|---|
Input Shape | 512x512 |
Batch Size | 8 |
Initial Learning Rate | 0.0001 |
Optimizer | Adam |
Alpha \(\alpha\) for Distiller | 0.5 |
Temperature \(T\) for Distiller | 10 |
