Skip to content

Add flag to apply DeduplicateHashedInitializersPass post graph surgery #2295

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
qti-ashimaj wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add flag to apply DeduplicateHashedInitializersPass post graph surgery #2295

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
13 changes: 11 additions & 2 deletions olive/passes/onnx/graph_surgeries.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2074,6 +2074,13 @@ def _default_config(cls, accelerator_spec: AcceleratorSpec) -> dict[str, PassCon
required=True,
description="List of surgeries to apply, each with its type and parameters",
),
"remove_duplicate_initializers": PassConfigParam(
type_=bool,
default_value=True,
description="""
Apply DeduplicateHashedInitializersPass after graph surgeries in case graph surgeries add duplicated initializers
""",
),
**get_external_data_config(),
}

Expand All @@ -2089,8 +2096,10 @@ def _run_for_config(
surgeon_instance = self.init_surgeon_instance(surgery)
onnx_model = surgeon_instance(onnx_model)

deduped_model = DeduplicateHashedInitializersPass()(ir.from_proto(onnx_model)).model
return model_proto_to_olive_model(ir.to_proto(deduped_model), output_model_path, config)
if config.remove_duplicate_initializers:
deduped_model = DeduplicateHashedInitializersPass()(ir.from_proto(onnx_model)).model
return model_proto_to_olive_model(ir.to_proto(deduped_model), output_model_path, config)
return model_proto_to_olive_model(onnx_model, output_model_path, config)

def init_surgeon_instance(self, surgery):
surgeon_name = surgery.get("surgeon").lower()
Expand Down
Loading