FlatTN
FlatTN copied to clipboard
some files not found
yangjie_rich_pretrain_unigram_path = '../embedding/gigaword_chn.all.a2b.uni.ite50.vec' yangjie_rich_pretrain_bigram_path = '../embedding/gigaword_chn.all.a2b.bi.ite50.vec' yangjie_rich_pretrain_word_path = '../embedding/ctb.50d.vec' yangjie_rich_pretrain_char_and_word_path = '../embedding/yangjie_word_char_mix.txt' lk_word_path = '../embedding/sgns.merge.word'
这些文件是怎么得到的?
Our code is based on Flat-Lattice-Transformer (FLAT) from LeeSureman. For the time being, we have used pretrained character and word embeddings provided by LeeSureman. Please refer to "How to run the code?" section of https://github.com/LeeSureman/Flat-Lattice-Transformer.
请问您这边复现了论文的效果了吗?工业界是否可以使用该方法,期待您的回复。
请问您这边复现了论文的效果了吗?工业界是否可以使用该方法,期待您的回复。
此代码库的代码是本论文在标贝中文文本正则化数据集上可以使用的代码,工业界可以使用该方法进行自动化文本正则化工作,谢谢您的提问~