Skip to content

Arm backend: Up tolerance for conv2d, upsample_bilinear2d, etc. #18494

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
Christoffer-JL wants to merge 2 commits into
base: main
Choose a base branch
from
+10 −3
Open

Arm backend: Up tolerance for conv2d, upsample_bilinear2d, etc. #18494

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cf54bf7
Arm backend: Up tolerance for conv2d, upsample_bilinear2d, etc.
Christoffer-JL Mar 16, 2026
a878ca4
Merge branch 'main' into cjoh_tolerance
Christoffer-JL Mar 26, 2026
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
5 changes: 3 additions & 2 deletions backends/arm/test/models/test_T5ForConditionalGeneration_arm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -114,7 +114,8 @@ def test_t5_for_conditional_generation_tosa_INT():
aten_op=[],
exir_op=[],
use_to_edge_transform_and_lower=True,
atol=5, # TODO: MLETORCH-1703: Reduce the tolerance of quantized T5ForConditionalGeneration
atol=14, # TODO: MLETORCH-1703: Reduce the tolerance of quantized T5ForConditionalGeneration
frobenius_threshold=0.3,
)
pipeline.change_args(
"check_count.exir",
Expand Down Expand Up @@ -161,7 +162,7 @@ def test_t5_for_conditional_generation_vgf_quant():
aten_op=[],
exir_op=[],
use_to_edge_transform_and_lower=True,
atol=5, # TODO: MLETORCH-1703: Reduce the tolerance of quantized T5ForConditionalGeneration
atol=14, # TODO: MLETORCH-1703: Reduce the tolerance of quantized T5ForConditionalGeneration
quantize=True,
)
pipeline.change_args(
Expand Down
2 changes: 1 addition & 1 deletion backends/arm/test/models/test_inception_v3_arm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ def test_ic3_tosa_FP_fp16():
aten_op=[],
exir_op=[],
use_to_edge_transform_and_lower=True,
atol=5e-2,
atol=0.2,
)
pipeline.run()

Expand Down
4 changes: 4 additions & 0 deletions backends/arm/test/ops/test_conv2d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -510,6 +510,8 @@ def test_convolution_2d_tosa_FP(test_data):
aten_op,
exir_op,
tosa_extensions=["bf16"],
atol=3e-3,
rtol=3e-3,
)
pipeline.run()

Expand Down Expand Up @@ -620,6 +622,8 @@ def test_convolution_2d_vgf_no_quant(test_data):
aten_op,
exir_op,
quantize=False,
atol=3e-3,
rtol=3e-3,
)
pipeline.run()

Expand Down
2 changes: 2 additions & 0 deletions backends/arm/test/ops/test_upsample_bilinear2d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -426,6 +426,8 @@ def test_upsample_bilinear2d_vec_vgf_no_quant_UpsamplingBilinear2d(
aten_op,
exir_op,
quantize=False,
atol=2e-3,
rtol=2e-3,
)
if not compare:
pipeline.pop_stage(-1)
Expand Down
Loading