Parameter | Description | Value |
---|---|---|
\(epoch\) | The number of training epochs | 10 |
\(batch\_size\) | The number of samples per training step | 128 |
\(d_{1} , d_{2}\) | The embedding sizes in the total graph perspective | 32, 16 |
\(\overline{d}_{1} , \overline{d}_{2} ,\tilde{d}_{1} ,\tilde{d}_{2}\) | The embedding sizes in the subgraph perspective | 128, 64, 32, 16 |
\(dropout\) | The dropout rate | 0.1 |
\(lr\) | The learning rate of the Adam optimizer | 0.001 |
\(m\) | The margin value of the hinge loss function | 0.1 |