Skip to content

Arm backend: Require safetensors for Ethos-U DeiT export #24027

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
Sebastian-Larsson merged 1 commit into from
Oct 22, 2026
Merged

Arm backend: Require safetensors for Ethos-U DeiT export #24027

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
3 changes: 3 additions & 0 deletions examples/arm/image_classification_example_ethos_u/model_export/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,9 @@ Running this script achieves a test set accuracy of 86.10% in FP32.

## Export and quantize

The exporter requires a safetensors checkpoint. The training script saves
this format by default. Legacy PyTorch `.bin` checkpoints are not accepted.

The `export_deit.py` script can be run as follows:

```bash
Expand Down
3 changes: 2 additions & 1 deletion examples/arm/image_classification_example_ethos_u/model_export/export_deit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,7 @@ def measure_accuracy(quantized_model, test_set):
type=str,
default="./deit-tiny-oxford-pet/final_model",
required=True,
help="Path to the fine-tuned ViT model.",
help="Path to the fine-tuned ViT model saved in safetensors format.",
)
argparser.add_argument(
"--output-path",
Expand All @@ -123,6 +123,7 @@ def measure_accuracy(quantized_model, test_set):
num_labels=37,
ignore_mismatched_sizes=True,
revision="b3428f18dcc7b543470d07f14b4a4157815d1880",
use_safetensors=True,
).eval()
image_preprocessor = AutoImageProcessor.from_pretrained(
"facebook/deit-tiny-patch16-224",
Expand Down
Loading