Skip to content

Fix: Filter QM9 artifact bonds > 1.85A and fix SMILES generation #10565

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
likhit2804 wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fix: Filter QM9 artifact bonds > 1.85A and fix SMILES generation #10565

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5309197
Fix: Filter QM9 artifact bonds > 1.85A and fix SMILES generation
likhit2804 Dec 21, 2025
afd49e8
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Dec 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 27 additions & 2 deletions torch_geometric/datasets/qm9.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -246,6 +246,9 @@ def process(self) -> None:
if i in skip:
continue

if mol is None:
continue

N = mol.GetNumAtoms()

conf = mol.GetConformer()
Expand Down Expand Up @@ -273,6 +276,15 @@ def process(self) -> None:
rows, cols, edge_types = [], [], []
for bond in mol.GetBonds():
start, end = bond.GetBeginAtomIdx(), bond.GetEndAtomIdx()

# --- START CHANGE: Geometric Filter ---
p1 = pos[start]
p2 = pos[end]
dist = (p1 - p2).norm().item()
if dist > 1.85:
continue
# --- END CHANGE ---

rows += [start, end]
cols += [end, start]
edge_types += 2 * [bonds[bond.GetBondType()]]
Expand All @@ -288,15 +300,28 @@ def process(self) -> None:

row, col = edge_index
hs = (z == 1).to(torch.float)
num_hs = scatter(hs[row], col, dim_size=N, reduce='sum').tolist()

if row.numel() > 0:
num_hs = scatter(hs[row], col, dim_size=N,
reduce='sum').tolist()
else:
num_hs = [0.0] * N

x1 = one_hot(torch.tensor(type_idx), num_classes=len(types))
x2 = torch.tensor([atomic_number, aromatic, sp, sp2, sp3, num_hs],
dtype=torch.float).t().contiguous()
x = torch.cat([x1, x2], dim=-1)

name = mol.GetProp('_Name')
smiles = Chem.MolToSmiles(mol, isomericSmiles=True)

# --- START CHANGE: Prefer Property SMILES ---
if mol.HasProp('GDB-SMILES'):
smiles = mol.GetProp('GDB-SMILES')
elif mol.HasProp('SMILES'):
smiles = mol.GetProp('SMILES')
else:
smiles = Chem.MolToSmiles(mol, isomericSmiles=True)
# --- END CHANGE ---

data = Data(
x=x,
Expand Down