Skip to content

Don't use source_partitions for quantizing cat #12405

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
facebook-github-bot merged 1 commit into from
Jul 12, 2025
+5 −12
Merged

Don't use source_partitions for quantizing cat #12405

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
17 changes: 5 additions & 12 deletions backends/xnnpack/quantizer/xnnpack_quantizer_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1002,22 +1002,15 @@ def _annotate_cat(
quantization_config: Optional[QuantizationConfig],
filter_fn: Optional[Callable[[Node], bool]] = None,
) -> Optional[list[list[Node]]]:
cat_partitions = get_source_partitions(gm.graph, [torch.cat], filter_fn)
cat_partitions = list(itertools.chain.from_iterable(cat_partitions.values()))
annotated_partitions = []
for cat_partition in cat_partitions:
cat_node = cat_partition.output_nodes[0]
if _is_annotated([cat_node]):
for cat_node in gm.graph.nodes:
if cat_node.target != torch.ops.aten.cat.default:
continue

if cat_node.target != torch.ops.aten.cat.default:
# TODO: change this to AnnotationException
raise Exception( # noqa: TRY002
f"Expected cat node: torch.ops.aten.cat.default, but found {cat_node.target}"
" please check if you are calling the correct capture API"
)
if _is_annotated([cat_node]):
continue

annotated_partitions.append(cat_partition.nodes)
annotated_partitions.append(cat_node.all_input_nodes)

input_act_qspec = get_input_act_qspec(quantization_config)
inputs = cat_node.args[0]
Expand Down
Loading