DeepPavlov对自己的回应进行排名

时间:2019-05-30 12:14:56

标签: deep-learning nlp search-engine

有人知道,如何在自己的回应中使用deeppavlov排名模块?文档(http://docs.deeppavlov.ai/en/master/components/neural_ranking.html)提供以下代码(用于保险数据集):

DateTime.MinValue.AddTicks(timeSpan.Ticks).ToOADate()

但是当我将此代码用于ubuntu数据集时:

rank_model = build_model(configs.ranking.ranking_insurance_interact, download=True)
predictor = rank_model.pipe[-1][-1]
candidates = ['auto insurance', 'life insurance', 'home insurance']
predictor.rebuild_responses(candidates)

事实证明,没有rank_model = build_model(configs.ranking.ranking_ubuntu_v2_bert_sep_interact, download=False) predictor = rank_model.pipe[-1][-1] candidates = ['environment variable', 'graphical interface', 'broken terminal'] predictor.rebuilt_responces(candidates) 方法:

rebuilt_responces

有什么想法吗?

1 个答案:

答案 0 :(得分:0)

我不得不用数据手动替换文件。解决方案在github问题中提供:https://github.com/deepmipt/DeepPavlov/issues/862