Skip to content

add a safetensors flavor conversion script with 2 models #320

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
cregouby merged 7 commits into from
Jun 10, 2026
Merged

add a safetensors flavor conversion script with 2 models #320

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
c667bf0
add safetensors conversion script with 2 models
cregouby May 21, 2026
773ac04
typo
cregouby May 21, 2026
9b3fba3
fix import
cregouby May 21, 2026
a975713
simplify hf links
cregouby May 21, 2026
6b569da
fix model is in state_dict key
cregouby May 21, 2026
681a151
stick to pt format thanks to https://github.com/mlverse/torch/issues/…
cregouby May 28, 2026
cbf3a16
fix syntax error
cregouby May 28, 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
2 changes: 1 addition & 1 deletion R/utils.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ download_and_cache <- function(url, redownload = FALSE, prefix = NULL) {

if (!file.exists(path) || redownload) {
# we should first download to a temporary file because
# download probalems could cause hard to debug errors.
# download problems could cause hard to debug errors.
tmp <- tempfile(fileext = fs::path_ext(path))
on.exit({try({fs::file_delete(tmp)}, silent = TRUE)}, add = TRUE)

Expand Down
6 changes: 4 additions & 2 deletions tools/convert-models.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,6 +146,8 @@ def blob_exist(bucket_name, blob_name):
'yolo_v12_n': 'https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo12n.pt',
'yolo_v12_s': 'https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo12s.pt',
'yolo_v12_x': 'https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo12x.pt',
'tabicl_classifier_v2': 'https://huggingface.co/jingang/TabICL/resolve/main/tabicl-classifier-v2-20260212.ckpt',
'tabicl_regressor_v2': 'https://huggingface.co/jingang/TabICL/resolve/main/tabicl-regressor-v2-20260212.ckpt',
}

os.makedirs("models", exist_ok=True)
Expand All @@ -172,8 +174,8 @@ def blob_exist(bucket_name, blob_name):
if name.startswith("convnext_") and name.endswith("k"):
m = m["model"]

# openmmlab models weights are embedded in a named object
if name.startswith("convnext_") and "_upernet_" in name:
# openmmlab and jingang models weights are embedded in a named object
if (name.startswith("convnext_") and "_upernet_" in name) or name.startswith("tabicl_"):
m = m["state_dict"]

for nm, par in m.items():
Expand Down
Loading