daguan-classify-2018

daguan-classify-2018 copied to clipboard

https://github.com/nlpjoe/daguan-classify-2018/blob/e2539e55769b1286631d95e786b20b8344266738/src/model/attention1.py#L41 代码很好，学到很多~ 有个疑问：这里应该是char_embedding吧？因为上面line21-24表示当main_feature为all时处理的是word_embedding，此时应该补充处理一下char_embedding吧？ 同样的疑问也在textcnn_model.py中出现了，line37-40， line53-55

liuyaox

pred1 = pickle.load(open('../../data/result-op5/lstmgruword_oof_f1_0.7686691377795438_a0.7832944028205417.pkl', 'rb')) pred2 = pickle.load(open('../../backup/result-op3-0907/svmword_pre_f1_0.7755615966138534_a0.7900114681760386.pkl', 'rb'))

2

When preparing the data, where does those two files come?

liyewu