Skip to content

T5A — Molecules / graphs (QM9-style)

Paper evidence: main.pdf · Block findings

Lemma: D5 · Stack: pytorch Nuisance key: compositional

Production change: Conformer / position blocks move; property label fixed.

Notebook (Run All, built-in demo): t05a-qm9-molecule.ipynb

pip install matching-pmh torch
# Open the notebook and Run All

What this task achieved (headline)

E1 matched position PMH: clean MAE 24.921 (−0.155 vs B0); σ=0.2 Å noise MAE 47.415 vs B0.

B0 MAE E1 MAE @ σ=0.2
25.076 24.921 47.415

Paper preset: t5_compositional_d5 · from pmh.benchmark.presets import get_preset

Subtasks (paper)

QM9 train + noise + embedding eval

E1 clean MAE 24.921.

Preset: t5_compositional_d5

MolGCN training

Preset: t5_compositional_d5

Position-noise eval sweep

Preset: t5_compositional_d5

Run with matching-pmh

from pmh import PMHTrainer, evaluate_robust_fit
# nuisance="compositional"

Do not use PMH when

Property definition changes at deploy.

Replace demo data with yours

Swap demo loaders for your train_loader, source_batches, target_batches, and deploy holdout. Hook the backbone before your task head.

← All 13 tasks · Quickstart