Upload modeling_vit.py

modeling_vit.py

@@ -369,11 +369,11 @@ class VisionTransformer(nn.Module):
                   if use_rel_pos_bias else None) for i in range(depth)
         ])
 
-        if self.pos_embed is not None:
-            trunc_normal_(self.pos_embed, std=.02)
-        trunc_normal_(self.cls_token, std=.02)
-        self.apply(self._init_weights)
-        self.fix_init_weight()
+        # if self.pos_embed is not None:
+        #     trunc_normal_(self.pos_embed, std=.02)
+        # trunc_normal_(self.cls_token, std=.02)
+        # self.apply(self._init_weights)
+        # self.fix_init_weight()
 
     def fix_init_weight(self):
         def rescale(param, layer_id):