【WSDM-2019 SimGNN】SimGNN: A Neural Network Approach to Fast Graph Similarity Computation
python main.py --exp_name=SimGNN| Parameter | Value |
|---|---|
| Batch size | 16 |
| Bins | 16 |
| Bottle neck neurons | 16 |
| Dropout | 0.5 |
| Epochs | 5 |
| Exp name | SimGNN |
| Filters 1 | 128 |
| Filters 2 | 64 |
| Filters 3 | 32 |
| Gpu index | 0 |
| Histogram | True |
| Learning rate | 0.001 |
| Seed | 16 |
| Tensor neurons | 16 |
| Weight decay | 0.0005 |
Code Framework Reference: SimGNN