如下代码在PaddleHub 1.6.2上面是好使的，移植到PaddleHub 1.8.0后报错（AttributeError: 'list' object has no attribute 'argmax'）

不知道是什么原因，请大佬指点一下。

run_states = cls_task.predict(data=data)

results = [run_state.run_results for run_state in run_states]

for result in results:

     index = 0

    result = np.argmax(result, axis=2)[0]

    for label in result:

        print("%s\tp=%s" % (data[index][0], label_map[label]))

        index += 1