Skip to content

Arm backend: Require safetensors for super-resolution checkpoints #24026

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
Sebastian-Larsson wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Arm backend: Require safetensors for super-resolution checkpoints #24026

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/super_resolution_example_vgf/model_export/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,9 @@ The concrete quick-demo commands below use the pinned revision currently cached
for `caidas/swin2SR-classical-sr-x2-64`:
`cee1c923c6a37361c6e5650b65dcf4be821e5d52`.

If you choose another checkpoint, use a source you trust and weights saved in
safetensors format. Pickle-based weights are not supported.

### FP export

```bash
Expand Down
2 changes: 2 additions & 0 deletions examples/arm/super_resolution_example_vgf/model_export/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -277,12 +277,14 @@ def _load_checkpointed_swin2sr(
model = Swin2SRForImageSuperResolution.from_pretrained( # nosec B615
checkpoint,
local_files_only=local_files_only,
use_safetensors=True,
).eval()
else:
model = Swin2SRForImageSuperResolution.from_pretrained( # nosec B615
checkpoint,
revision=checkpoint_revision,
local_files_only=local_files_only,
use_safetensors=True,
).eval()
return Swin2SRWrapper(model)

Expand Down
Loading