enable load local model in subfolder

optimum/onnxruntime/modeling_ort.py

@@ -510,13 +510,12 @@ def _from_pretrained(
 
         if file_name is None:
             if model_path.is_dir():
-                onnx_files = list(model_path.glob("*.onnx"))
+                onnx_files = list((model_path / subfolder).glob("*.onnx"))
             else:
                 repo_files, _ = TasksManager.get_model_files(
                     model_id, revision=revision, cache_dir=cache_dir, token=token
                 )
                 repo_files = map(Path, repo_files)
-
                 pattern = "*.onnx" if subfolder == "" else f"{subfolder}/*.onnx"
                 onnx_files = [p for p in repo_files if p.match(pattern)]
 
@@ -983,10 +982,9 @@ def _cached_file(
             token = use_auth_token
 
         model_path = Path(model_path)
-
         # locates a file in a local folder and repo, downloads and cache it if necessary.
         if model_path.is_dir():
-            model_cache_path = model_path / file_name
+            model_cache_path = model_path / subfolder / file_name
             preprocessors = maybe_load_preprocessors(model_path.as_posix())
         else:
             model_cache_path = hf_hub_download(

tests/onnxruntime/test_modeling.py

@@ -28,6 +28,7 @@
 import requests
 import timm
 import torch
+from huggingface_hub import HfApi
 from huggingface_hub.constants import default_cache_path
 from parameterized import parameterized
 from PIL import Image
@@ -1263,6 +1264,19 @@ def test_trust_remote_code(self):
             torch.allclose(pt_logits, ort_logits, atol=1e-4), f" Maxdiff: {torch.abs(pt_logits - ort_logits).max()}"
         )
 
+    @parameterized.expand(("", "onnx"))
+    def test_loading_with_config_in_root(self, subfolder):
+        # config.json file in the root directory and not in the subfolder
+        model_id = "sentence-transformers-testing/stsb-bert-tiny-onnx"
+        # hub model
+        ORTModelForFeatureExtraction.from_pretrained(model_id, subfolder=subfolder, export=subfolder == "")
+        # local model
+        api = HfApi()
+        with tempfile.TemporaryDirectory() as tmpdirname:
+            local_dir = Path(tmpdirname) / "model"
+            api.snapshot_download(repo_id=model_id, local_dir=local_dir)
+            ORTModelForFeatureExtraction.from_pretrained(local_dir, subfolder=subfolder, export=subfolder == "")
+
 
 class ORTModelForQuestionAnsweringIntegrationTest(ORTModelTestMixin):
     SUPPORTED_ARCHITECTURES = [