Tolokers
Description
Tolokers
Dataset Overview
Lifting Methods
Structural-based Liftings
- Cellular: Cycle-based lifting
- Simplicial: Clique complex lifting
- Hypergraph: k-hop lifting
Feature Lifting
- Projected sum
Model Performance
| Model | Accuracy (%) | Std Dev (±) |
|---|---|---|
| GCN | 83.02 | 0.71 |
| GIN | 80.72 | 1.19 |
| GAT | 84.43 | 1.00 |
| AST | 83.26 | 0.10 |
| EDGNN | 77.53 | 0.01 |
| UniGNN2 | 77.35 | 0.03 |
| CWN | OOM | OOM |
| CCCN | OOM | OOM |
| SCCNN | OOM | OOM |
| SCN | OOM | OOM |
Key Insights
- GAT achieves the best performance with 84.43% accuracy
- GCN and AST also perform well, with accuracies above 83%
- GIN shows moderate variability (±1.19)
- EDGNN and UniGNN2 have lower performance (below 78%)
- CWN, CCCN, SCCNN, and SCN ran out of memory (OOM)
Reproducibility
python -m topobench model=graph/gat dataset=graph/tolokers optimizer.parameters.lr=0.001 model.feature_encoder.out_channels=128 model.backbone.num_layers=4 model.feature_encoder.proj_dropout=0.25 dataset.dataloader_params.batch_size=1 dataset.split_params.data_seed=0,3,5,7,9 trainer.max_epochs=1000 trainer.min_epochs=50 trainer.check_val_every_n_epoch=1 callbacks.early_stopping.patience=50 logger.wandb.project=TopoBench --multirun