Table 1 Hyper-parameters settings.
From: Spatial interpolation of global DEM using federated deep learning
Hyper-parameters | Value |
|---|---|
Image shape | \(64 \times 64 \times 1\) |
Local clients number | 8 |
Rate of selected clients | 0.5 |
Global epochs | 80 |
Local epochs | 4 |
Batch size | 128 |
Learning rate | 0.0003 |
Activation founction | LeakyReLU |
Optimizer | Adam |
\(\beta _{1}\) of Adam optimizer | 0.9 |
\(\beta _{2}\) of Adam optimizer | 0.999 |
\(\varepsilon\) of Adam optimizer | 10-8 |
