Table 7 Hyperparameters and combined values for the Cartesian search of the best configuration of the trained ML model
From: A novel botnet attack detection for IoT networks based on communication graphs
Topology | Bottleneck | IQR factor | Epochs | Batch size |
|---|---|---|---|---|
Units L1: 7 L2: 5 | 1 | 0.01 | 10 | 32 |
Units L1: 8 L2: 7 L3: 6 L4: 5 | 2 | 0.1 | 20 | 64 |
Units L1: 15 L2: 10 L3: 7 L4: 5 | 3 | 0.5 | 128 | |
4 | 1.5 | |||
3 |