Skip to content

limited facto tensor size to be less than 4000 bytes #14313

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
Sep 17, 2025
Merged

limited facto tensor size to be less than 4000 bytes #14313

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: 8 additions & 5 deletions backends/cadence/utils/facto_util.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,9 @@


def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
# Constraint to limit tensor size product to < 4000
max_size_constraint = cp.Size.Le(lambda deps, r, d: max(1, int((3999) ** (1 / r))))

tensor_constraints = (
[
cp.Dtype.In(
Expand All @@ -39,7 +42,7 @@ def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
cp.Value.Le(lambda deps, dtype, struct: 2**4),
cp.Rank.Ge(lambda deps: 1),
cp.Size.Ge(lambda deps, r, d: 1),
cp.Size.Le(lambda deps, r, d: 2**9),
max_size_constraint,
cp.Rank.Le(lambda deps: 2**3),
]
if op_name
Expand All @@ -62,7 +65,7 @@ def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
cp.Value.Le(lambda deps, dtype, struct: 2**4),
cp.Rank.Ge(lambda deps: 1),
cp.Size.Ge(lambda deps, r, d: 1),
cp.Size.Le(lambda deps, r, d: 2**9),
max_size_constraint,
cp.Rank.Le(lambda deps: 2**3),
]
)
Expand All @@ -76,7 +79,7 @@ def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
cp.Value.Le(lambda deps, dtype, struct: 2**4),
cp.Rank.Ge(lambda deps: 1),
cp.Size.Ge(lambda deps, r, d: 1),
cp.Size.Le(lambda deps, r, d: 2**9),
max_size_constraint,
]
else:
tensor_constraints = [
Expand All @@ -94,7 +97,7 @@ def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
cp.Value.Le(lambda deps, dtype, struct: 2**4),
cp.Rank.Ge(lambda deps: 1),
cp.Size.Ge(lambda deps, r, d: 1),
cp.Size.Le(lambda deps, r, d: 2**9),
max_size_constraint,
]
case "embedding.default":
tensor_constraints = [
Expand All @@ -104,7 +107,7 @@ def apply_tensor_contraints(op_name: str, index: int) -> list[object]:
cp.Value.Le(lambda deps, dtype, struct: 2**4),
cp.Rank.Ge(lambda deps: 1),
cp.Size.Ge(lambda deps, r, d: 1),
cp.Size.Le(lambda deps, r, d: 2**9),
max_size_constraint,
]
case "sigmoid.default":
tensor_constraints.extend(
Expand Down
Loading