Skip to content

Arm backend: Add test for DeiT Tiny for TOSA BI #10846

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
May 15, 2025
Merged

Arm backend: Add test for DeiT Tiny for TOSA BI #10846

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
18 changes: 17 additions & 1 deletion backends/arm/test/models/test_deit_tiny_arm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,10 @@

import torch

from executorch.backends.arm.test.tester.test_pipeline import TosaPipelineMI
from executorch.backends.arm.test.tester.test_pipeline import (
TosaPipelineBI,
TosaPipelineMI,
)

from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
from torchvision import transforms
Expand Down Expand Up @@ -42,3 +45,16 @@ def test_deit_tiny_tosa_MI():
qtol=1,
)
pipeline.run()


def test_deit_tiny_tosa_BI():
pipeline = TosaPipelineBI[input_t](
deit_tiny,
model_inputs,
aten_op=[],
exir_op=[],
use_to_edge_transform_and_lower=True,
atol=3.0, # This needs to go down: MLETORCH-956
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is it stable though? As opposed to flaky, else we can mark it as Flaky..

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@digantdesai I believe it is rather stable. Been looping it several times myself and the numerical error looks to be consistent.

qtol=1,
)
pipeline.run()
Loading