Safe cache check

modeling_chatglm.py

@@ -717,8 +717,9 @@ class SelfAttention(torch.nn.Module):
         # adjust key and value for inference
         if kv_cache is not None:
             cache_k, cache_v = kv_cache
-            key_layer = torch.cat((cache_k, key_layer), dim=2)
-            value_layer = torch.cat((cache_v, value_layer), dim=2)
+            if cache_k is not None and cache_v is not None:
+                key_layer = torch.cat((cache_k, key_layer), dim=2)
+                value_layer = torch.cat((cache_v, value_layer), dim=2)
         if use_cache:
             if kv_cache is None:
                 kv_cache = torch.cat((key_layer.unsqueeze(0).unsqueeze(0), value_layer.unsqueeze(0).unsqueeze(0)),