Table 2 Model training parameters.
From: An efficient camouflaged image segmentation with modified UNet and attention techniques
Parameters | Values |
|---|---|
Batch size | The training batch size experimented with values of 16 and 32, and it was observed that the model performed better with a batch size of 16 |
Optimizer | Adam with a learning rate of 1e-2 |
Learning rate | 0.001 |
Epochs | 100 |
Dropout rate | 0.5 |
Loss function | Categorical cross-entropy |
