Update demo/infer.py

demo/infer.py

@@ -204,7 +204,6 @@ class LiveCCDemoInfer:
             images=image_inputs,
             videos=video_inputs,
             return_tensors="pt",
-            return_attention_mask=False
         )
         inputs.to(self.model.device)
         if past_ids is not None:
@@ -217,4 +216,6 @@ class LiveCCDemoInfer:
         )
         state['past_key_values'] = outputs.past_key_values
         state['past_ids'] = outputs.sequences[:, :-1]
-        return self.processor.decode(outputs.sequences[0, inputs.input_ids.size(1):], skip_special_tokens=True), state
+        response = self.processor.decode(outputs.sequences[0, inputs.input_ids.size(1):], skip_special_tokens=True)
+        print(response)
+        return response, state