Skip to content

Fixing slice scatter and select scatter decomposition #3093

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
Aug 29, 2024
Merged

Fixing slice scatter and select scatter decomposition #3093

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 py/torch_tensorrt/dynamo/lowering/_decompositions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -190,6 +190,7 @@ def slice_scatter_decomposition(
step: Optional[int] = None,
) -> torch.Tensor:
dim_size = input_tensor.shape[dim]
device_input_tensor = input_tensor.device
start = get_positive_dim(start, input_tensor.shape[dim])
if end is None:
end = dim_size
Expand All @@ -216,7 +217,8 @@ def slice_scatter_decomposition(
index_tensor_shape.append(src_each_dim)
for index in range(start, end, step):
cat_tensors.append(index * torch.ones(index_tensor_shape, dtype=torch.int64))
index_tensor = torch.stack(cat_tensors, dim).to(input_tensor.device)
index_tensor = torch.stack(cat_tensors, dim)
index_tensor = index_tensor.to(device_input_tensor)
index_tensor_64 = index_tensor.to(torch.int64)
output_tensor = torch.scatter(input_tensor, dim, index_tensor_64, src_tensor)
return output_tensor
Expand Down
Loading