Table 4 CNN setup details
From: Fundus images analysis using deep features for detection of exudates, hemorrhages and microaneurysms
CNN parameters | Optimal value |
|---|---|
Learning Rate | 0.01 |
Momentum | 0.9 |
Gaussian Weight Filters | 0.01 |
Training Batch size | 128 |
Validation and Test Batch size | 32 |
Solver Method | SGD |
Gamma | 0.1 |
Policy of the SGD | Step-Down |
Step size of SGD | 33 |