scan_cache (huggingface_hub)

utils.py

@@ -22,6 +22,8 @@ from interfaces.ontolisst import build_huggingface_path as hf_ontlisst_path
 from interfaces.illframes import build_huggingface_path as hf_illframes_path
 from interfaces.ontolisst import build_huggingface_path as hf_ontolisst_path
 
+from huggingface_hub import scan_cache_dir
+
 HF_TOKEN = os.environ["hf_read"]
 
 # should be a temporary solution
@@ -59,9 +61,14 @@ def df_h():
     print(result.stdout)
     
 def scan_cache():
-    result = subprocess.run(["huggingface-cli", "scan-cache"], capture_output=True, text=True)
-    print(result.stdout)
+    cache_dir = os.environ.get("TRANSFORMERS_CACHE", os.path.expanduser("~/.cache/huggingface/transformers"))
+    scan_result = scan_cache_dir(cache_dir)
     
+    print("=== Model Cache Report ===")
+    print(f"Cache size: {scan_result.size_on_disk / 1e6:.2f} MB")
+    print(f"Number of repos: {len(scan_result.repos)}")
+    for repo in scan_result.repos:
+        print(f"- {repo.repo_id} ({repo.repo_type}) — {repo.size_on_disk / 1e6:.2f} MB")
     
 def set_hf_cache_dir(path:str):
     os.environ['TRANSFORMERS_CACHE'] = path