Table 5.
The parameters settings for the ELM and GWO.
| ELM | GWO | ||
|---|---|---|---|
| Parameter | Value | Parameter | Value |
| AS | assemble of the biases and input weights | Population (wolves or search agents) | Consists of the position of all search agents |
| ρ | Output-weights matrix | Position | Start stochastically generated within the range of [-1, 1] for the input-weights and [0, 1] for the biases |
| Input-weights (w) | −1 to 1 | Population size or number of search agents (NSA) | 50 |
| Bias values (b) | 0 to 1 | r1 and r2 | Stochastically generated with the range of [0, 1] |
| Input-nodes number (n) | Input attributes | Number of iterations itmax | 100 |
| Hidden-nodes number (L) | [100–300]; with a 25 increment step | C1, C2, and C3 | Randomly generated vectors based on Equation (4) |
| Output neurones number (m) | Number of classes | A1, A2, and A3 | Randomly generated vectors using Equation (3) |
| Activation function | Sigmoid | Xα | Best position of all search agents. |