Skip to content

call named_modules once per model prepare #3159

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

Merged
merged 1 commit into from
Oct 14, 2025
Merged

call named_modules once per model prepare #3159

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
4 changes: 3 additions & 1 deletion torchao/quantization/pt2e/prepare.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -585,6 +585,7 @@ def _maybe_insert_input_and_output_observers_for_node(
node: Node,
model: torch.fx.GraphModule,
obs_or_fq_map: dict[EdgeOrNode, ObserverOrFakeQuantize],
named_modules: dict[str, torch.nn.Module],
is_qat: bool,
model_device: Optional[torch.device] = None,
):
Expand All @@ -594,7 +595,6 @@ def _maybe_insert_input_and_output_observers_for_node(
if this_node_quantization_annotation is None:
return

named_modules = dict(model.named_modules(remove_duplicate=False))
_maybe_insert_input_observers_for_node(
node,
None, # qconfig
Expand Down Expand Up @@ -666,13 +666,15 @@ def prepare(
if obs_or_fq_callback:
obs_or_fq_callback(model, obs_or_fq_map)
model_device = _assert_and_get_unique_device(model)
named_modules = dict(model.named_modules(remove_duplicate=False))

for node in nodes_before_observation:
# TODO: simplify logic for inserting observers
_maybe_insert_input_and_output_observers_for_node(
node,
model,
obs_or_fq_map,
named_modules,
is_qat,
model_device,
)
Expand Down
Loading