Skip to content

Treat aten.slice.Tensor as a shared-observer op in XNNPACKQuantizer #21568

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

Open
Hyungkeun-Park-Nota wants to merge 1 commit into
base: main
Choose a base branch
from
+7 −6
Open

Treat aten.slice.Tensor as a shared-observer op in XNNPACKQuantizer #21568

Show file tree
Hide file tree
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
1 change: 1 addition & 0 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 @@ -1102,6 +1102,7 @@ def _is_share_obs_or_fq_op(op: Callable) -> bool:
torch.ops.aten.view_copy.default,
torch.ops.aten.view.default,
torch.ops.aten.reshape.default,
torch.ops.aten.slice.Tensor,
torch.ops.aten.slice_copy.Tensor,
torch.ops.aten.flatten.using_ints,
]
Expand Down
12 changes: 6 additions & 6 deletions backends/xnnpack/test/ops/test_slice_copy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,10 +140,9 @@ def forward(self, x):
.check_not(["torch.ops.higher_order.executorch_call_delegate"])
)

# Note: Slice ends up as slice_copy later in the process, but during quantization,
# it's still slice, which isn't supported by the XNNPACK quantizer.
@unittest.skip("T156004676 - slice isn't propagated")
def _test_qs8_slice_copy(self):
# Note: Slice ends up as slice_copy later in the process, but during quantization
# it's still slice, so the quantizer shares observers on aten.slice.Tensor.
def test_qs8_slice_copy(self):
class SliceCopy(torch.nn.Module):
def forward(self, x):
y = x + x
Expand All @@ -157,8 +156,9 @@ def forward(self, x):
.export()
.check_node_count(
{
"aten::slice.Tensor": 3,
"quantized_decomposed::quantize_per_tensor": 3,
torch.ops.aten.slice.Tensor: 3,
# 3 slices, plus the input and the add output
torch.ops.quantized_decomposed.quantize_per_tensor.default: 5,
}
)
.to_edge_transform_and_lower()
Expand Down
Loading