PaddlePaddle按照文档编写，出现name 'result_ids_lod' is not defined

飞桨PaddlePaddle

于 2019-03-04 11:11:14 发布

阅读量1.4k

点赞数

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/PaddlePaddle/article/details/88105555

本文描述了在使用Fluid1.1进行机器翻译时遇到的“name'result_ids_lod'isnotdefined”错误，并提供了详细的代码示例和解决方案。通过添加一行代码来获取结果的LOD信息，成功解决了这一问题。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

问题描述：我使用Fluid1.1按照文档编写相应的结果，出现name ‘result_ids_lod’ is not defined错误
报错信息：

Original sentence:
<unk> , l&apos; indice de la bourse technologique a perdu 0,8 % , et a fini avec <unk> points du marché .
Translated score and sentence:
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-14-9adba1019ebd> in <module>()
     23     print("Translated score and sentence:")
     24     for i in range(beam_size):
---> 25         start_pos = result_ids_lod[1][i] + 1
     26         end_pos = result_ids_lod[1][i+1]
     27         print("%d\t%.4f\t%s\n" % (i+1, result_scores[end_pos-1],

NameError: name 'result_ids_lod' is not defined

问题复现：

for data in test_data():
    feed_data = list(map(lambda x: [x[0]], data))
    feed_dict = feeder.feed(feed_data)
    feed_dict['init_ids'] = init_ids
    feed_dict['init_scores'] = init_scores

    results = exe.run(
        framework.default_main_program(),
        feed=feed_dict,
        fetch_list=[translation_ids, translation_scores],
        return_numpy=False)

    result_ids = np.array(results[0])
    result_scores = np.array(results[1])

问题解决：

文档中的代码缺少了result_ids_lod = results[0].lod()，导致出现了上述问题。

for data in test_data():
    feed_data = list(map(lambda x: [x[0]], data))
    feed_dict = feeder.feed(feed_data)
    feed_dict['init_ids'] = init_ids
    feed_dict['init_scores'] = init_scores

    results = exe.run(
        framework.default_main_program(),
        feed=feed_dict,
        fetch_list=[translation_ids, translation_scores],
        return_numpy=False)

    result_ids = np.array(results[0])
    result_ids_lod = results[0].lod()
    result_scores = np.array(results[1])

最新的代码请参考https://github.com/PaddlePaddle/book/blob/fa35415f2b3f5a5d3e045ff0564d5df0a5a2b0d5/08.machine_translation/infer.py