Skip to content

Be smarter about torch tensors #43

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 5 commits into from
Oct 4, 2025
Merged

Be smarter about torch tensors #43

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
6c005d9
Be smarter about torch tensors
justinchuby Oct 3, 2025
eb7da76
not working
justinchuby Oct 4, 2025
acb55fc
try this
justinchuby Oct 4, 2025
1ee3cbf
format
justinchuby Oct 4, 2025
ec8226e
update pass
justinchuby Oct 4, 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
17 changes: 14 additions & 3 deletions src/torch_onnx_models/onnx_passes/_fold_transpose.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,11 +32,22 @@ def call(self, model: ir.Model) -> ir.passes.PassResult:
perm = node.attributes.get_ints("perm", reversed(range(len(shape))))

# Create a lazy transposed tensor
torch_tensor = initializer.raw
if isinstance(torch_tensor, torch.Tensor):
if isinstance(initializer.raw, torch.Tensor):

def tensor_func(tensor=torch_tensor):
def tensor_func(tensor=initializer.raw):
return tensor_adapters.TorchTensor(tensor.permute(*perm), name=name)

elif isinstance(initializer, ir.LazyTensor):
# Unwrap the lazy tensor to get the underlying torch tensor
def tensor_func(tensor=initializer):
while isinstance(tensor, ir.LazyTensor):
tensor = tensor.raw()
torch_tensor = tensor.raw
if not isinstance(torch_tensor, torch.Tensor):
torch_tensor = torch.from_numpy(tensor.numpy())
return tensor_adapters.TorchTensor(
torch_tensor.permute(*perm), name=name
)
else:

def tensor_func(tensor=initializer):
Expand Down
Loading