fix

app.py

@@ -13,7 +13,7 @@ def process(input_image, prompt):
     sample = {"image": image, "text_input": [text_input]}
     pred_answers, caption, gradcam = model.predict_answers(sample, num_captions=50, num_patches=20)
 
-    return pred_answer
+    return pred_answers[0]
 
 if __name__ == '__main__':
     input_image = gr.inputs.Image(label='image', type='pil')