Skip to content

[torchlib] Update linear implementation to support 1d weights #2340

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 4 commits into from
May 26, 2025
Merged

[torchlib] Update linear implementation to support 1d weights #2340

Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d293542
[torchlib] Update linear implementation to support 1d weights
justinchuby May 25, 2025
d0f7385
Update nn.py
justinchuby May 25, 2025
499b7a5
Update nn.py
justinchuby May 25, 2025
7773985
Update onnxscript/function_libs/torch_lib/ops/nn.py
justinchuby May 25, 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
9 changes: 7 additions & 2 deletions onnxscript/function_libs/torch_lib/ops/nn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -825,10 +825,15 @@
def aten_linear(input: TFloat, weight: TFloat, bias: Optional[TFloat] = None) -> TFloat:
"""linear(Tensor input, Tensor weight, Tensor? bias=None) -> Tensor"""

if len(input.shape) == 2:
if len(input.shape) == 2 and len(weight.shape) == 2:
# Use Gemm for the rank 2 input
return op.Gemm(input, weight, bias, transB=True)
weight_transposed = op.Transpose(weight, perm=[1, 0])
if len(weight.shape) == 1:
# In rare cases the weight can be 1d
weight_transposed = op.Unsqueeze(weight, [1])
else:
assert len(weight.shape) == 2:
weight_transposed = op.Transpose(weight, perm=[1, 0])
mul = op.MatMul(input, weight_transposed)
if bias is None:
return mul
Expand Down
Loading