Allow only textual inputs to VisualBert

src/transformers/models/visual_bert/modeling_visual_bert.py

@@ -772,29 +772,30 @@ def forward(
         else:
             raise ValueError("You have to specify either input_ids or inputs_embeds")
 
-        if visual_embeds is None:
-            raise ValueError(
-                f"`visual_embeds` can not be of type {type(visual_embeds)} when using a VisualBert Model."
-            )
-
         batch_size, seq_length = input_shape
         device = input_ids.device if input_ids is not None else inputs_embeds.device
 
-        visual_input_shape = visual_embeds.size()[:-1]
+        if visual_embeds is not None:
+            visual_input_shape = visual_embeds.size()[:-1]
 
         if attention_mask is None:
             attention_mask = torch.ones(input_shape, device=device)
 
-        if visual_attention_mask is None:
+        if visual_embeds is not None and visual_attention_mask is None:
             visual_attention_mask = torch.ones(visual_input_shape, device=device)
 
         # We can provide a self-attention mask of dimensions [batch_size, from_seq_length, to_seq_length]
         # ourselves in which case we just need to make it broadcastable to all heads.
+        if visual_embeds is not None:
+            combined_attention_mask = torch.cat((attention_mask, visual_attention_mask), dim=-1)
+            extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(
+                combined_attention_mask, [batch_size, input_shape + visual_input_shape], device
+            )
 
-        combined_attention_mask = torch.cat((attention_mask, visual_attention_mask), dim=-1)
-        extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(
-            combined_attention_mask, [batch_size, input_shape + visual_input_shape], device
-        )
+        else:
+            extended_attention_mask: torch.Tensor = self.get_extended_attention_mask(
+                attention_mask, [batch_size, input_shape], device
+            )
 
         # Prepare head mask if needed
         # 1.0 in head_mask indicate we keep the head