Skip to content

torchmdnet coulomb_cutoff only if periodic #138

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

Draft
sef43 wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

torchmdnet coulomb_cutoff only if periodic #138

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
5 changes: 2 additions & 3 deletions openmmml/models/torchmdnetpotential.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -162,12 +162,13 @@ def addForces(self,
filename=filename,
)

periodic = (topology.getPeriodicBoxVectors() is not None) or system.usesPeriodicBoundaryConditions()
model = load_model(
model_file_path,
derivative=False,
remove_ref_energy = args.get('remove_ref_energy', True),
max_num_neighbors = min(args.get('max_num_neighbors', 64), numbers.shape[0]),
coulomb_cutoff = cutoff,
coulomb_cutoff = cutoff if periodic else None,
static_shapes = True,
check_errors = False
).to(device)
Expand All @@ -182,7 +183,6 @@ def addForces(self,
indices = None
else:
indices = np.array(atoms)
periodic = (topology.getPeriodicBoxVectors() is not None) or system.usesPeriodicBoundaryConditions()

# Create the PythonForce and add it to the System.

Expand Down Expand Up @@ -227,7 +227,6 @@ def __call__(self, state):
cell = None
if self.compiled_model is None:
# The model can't be compiled until after it has been invoked once.

energy = self.model(z=self.numbers, pos=positions/self.lengthScale, batch=self.batch, q=self.charge, box=cell)[0]*self.energyScale
self.compiled_model = torch.compile(self.model, backend="inductor", dynamic=False, fullgraph=True, mode="reduce-overhead")
else:
Expand Down