feat: support pre-downloading clips for offline (haotian-liu#376)

chijiang · Aug 27, 2023 · c12c256 · c12c256
1 parent 22477ae
commit c12c256
Showing 1 changed file with 3 additions and 1 deletion.
diff --git a/llava/model/multimodal_encoder/builder.py b/llava/model/multimodal_encoder/builder.py
@@ -1,9 +1,11 @@
+import os
 from .clip_encoder import CLIPVisionTower
 
 
 def build_vision_tower(vision_tower_cfg, **kwargs):
     vision_tower = getattr(vision_tower_cfg, 'mm_vision_tower', getattr(vision_tower_cfg, 'vision_tower', None))
-    if vision_tower.startswith("openai") or vision_tower.startswith("laion"):
+    is_absolute_path_exists = os.path.exists(vision_tower)
+    if is_absolute_path_exists or vision_tower.startswith("openai") or vision_tower.startswith("laion"):
         return CLIPVisionTower(vision_tower, args=vision_tower_cfg, **kwargs)
 
     raise ValueError(f'Unknown vision tower: {vision_tower}')