IMDB-MULTI
Description
IMDB-MULTI
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 | 49.97 | 2.16 |
| GIN | 47.68 | 4.21 |
| GAT | 50.13 | 3.87 |
| AST | 50.51 | 2.92 |
| EDGNN | 49.17 | 4.35 |
| UniGNN2 | 49.76 | 3.55 |
| CWN | 49.71 | 2.83 |
| CCCN | 47.79 | 3.45 |
| SCCNN | 48.75 | 3.98 |
| SCN | 49.49 | 5.08 |
Key Insights
- AST achieves the best performance with 50.51% accuracy
- Model accuracies range from 47.68% (GIN) to 50.51% (AST)
- SCN shows the highest variability (±5.08), while AST is both the most accurate and relatively stable (±2.92)
- Most models perform within a narrow band, with several models above 49%
Reproducibility
python -m topobench model=hypergraph/allsettransformer dataset=graph/IMDB-MULTI optimizer.parameters.lr=0.01 model.feature_encoder.out_channels=64 model.backbone.n_layers=4 model.feature_encoder.proj_dropout=0.25 dataset.dataloader_params.batch_size=256 dataset.split_params.data_seed=0,3,5,7,9 trainer.max_epochs=500 trainer.min_epochs=50 trainer.check_val_every_n_epoch=5 callbacks.early_stopping.patience=10 logger.wandb.project=TopoBench --multirun