Table 2 Optimal parameters of our deep learning framework.
From: A pyramidal deep learning pipeline for kidney whole-slide histology images classification
Parameter | Value |
|---|---|
Number of convolutional layers | 8 |
Kernel size | 3 × 3 |
Kernel initialization | He initialization11 |
Number of filters | 9 |
Stride | 1 (convolution), 2 (max-pooling) |
Patch size | 32 |
Number of epochs | 60 |
Learning rate | 0.001 |
Optimizer | Adam |
