Upload modeling_minicpmv.py

modeling_minicpmv.py

@@ -2,6 +2,7 @@ import math
 from typing import List, Optional
 import timm
 import torch
+import torch.nn.functional as F
 
 from PIL import Image
 from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
@@ -521,8 +522,12 @@ class MiniCPMVEmbedding(MiniCPMV): # MiniCPMVEmbedding -> MiniCPMV ->  Ultimatel
             return_dict=True
         )
         
+        last_hidden_state = vlm_outputs.last_hidden_state
+        
+        last_hidden_state_normalized = F.normalize(last_hidden_state, dim=1)
+        
         return BaseModelOutputWithAttentionMask(
-            last_hidden_state=vlm_outputs.last_hidden_state,
+            last_hidden_state=last_hidden_state_normalized,
             attention_mask=model_inputs.attention_mask
         )