Skip to content

RESULTS.md

Latest commit

 

History

History
35 lines (27 loc) · 1.18 KB

RESULTS.md

File metadata and controls

35 lines (27 loc) · 1.18 KB
Raw

Synthetic benchmark results

  • Date: 2026-05-20
  • Dataset: synthetic_demo (50 drugs, 80 hyperedges)
  • Seeds: 42, 43, 44, 45, 46
  • Hardware: CPU (see local run logs under experiments/)

Note: Synthetic demo data is for pipeline validation only — not pharmacological evidence.

Metrics (test split, mean ± std over seeds)

Model ROC-AUC PR-AUC F1
HGNN 0.4893 ± 0.1034 0.7168 ± 0.0750 0.8235 ± 0.0000
GCN 0.5693 ± 0.0991 0.7549 ± 0.0722 0.3283 ± 0.4496
GAT 0.3824 ± 0.0850 0.6477 ± 0.0478 0.8235 ± 0.0000
GraphSAGE 0.4923 ± 0.0704 0.6909 ± 0.0419 0.8235 ± 0.0000
MLP 0.4970 ± 0.0998 0.7069 ± 0.0503 0.8235 ± 0.0000

Reproduce

cd hypergraph-ddi
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt && pip install -e .
export HYPERGRAPH_DDI_ROOT="$(pwd)"
python scripts/run_synthetic_benchmark.py --seeds 42 43 44 45 46

Or per model:

python scripts/preprocess.py --config config/demo_synthetic.yaml
python scripts/run_experiment.py --config config/demo_synthetic.yaml --seeds 42 43 44 45 46 --skip-preprocess