ymcui/MacBERT

Issues

• It’s not a question; I just want to share a joke with everyone.

  #23 opened 4 months ago by songqibin
  1

• 为啥不开源训练代码呢，你们也是用的别人开源的代码

  #22 opened 2 years ago by pipilove
  2

• transformers中调用输入文本数据代码在哪呀

  #21 opened 2 years ago by Mr-IT007
  2

• 问题

  #20 opened 2 years ago by C929-x
  1

• 如果相似词与原词长度不同是怎么处理的？谢谢

  #19 opened 2 years ago by QunBB
  2

• 为什么macbert的tokenizer对中文单引号或者双引号的分词结果为[UNK]

  #17 opened 2 years ago by xueyongfu
  2

• 请问可以提供一下Finetune的代码吗？

  #11 opened 3 years ago by nitwtog
  1

• huggingface models clone failed

  #16 opened 2 years ago by Vimos
  3

• 请问 whole word masking 和 N-gram masking 是如何一起使用的？

  #15 opened 3 years ago by dr-GitHub-account
  2

• macbert小模型

  #14 opened 3 years ago by ZTurboX
  1

• When using Synonyms masking, Synonyms are not in the list of vocab.

  #9 opened 3 years ago by TianhaoFu
  4

• 咨询 MacBert 上的一些问题

  #13 opened 3 years ago by T-baby
  4

• 如果计算得到的相似词与原词长度不同咋办呢？谢谢

  #12 opened 3 years ago by TriLoo
  2

• Could you please release the code of mask method , especially the ngram mask?

  #10 opened 3 years ago by AQA6666
  2

• 關於繼續訓練pretrained model

  #8 opened 3 years ago by phc4valid
  5

• 能否新增一个bort模型呢？

  #7 opened 3 years ago by chadqiu
  3

• 词表中的<S>和<T>含义

  #6 opened 4 years ago by zyccyz
  2

• MacBERT 在 mask 上的一些细节问题

  #5 opened 4 years ago by wlhgtc
  2

• 请问MacBERT的参数和RoBERTa-wwm的参数是兼容的吗？

  #4 opened 4 years ago by daizh
  2

• missing weights

  #3 opened 4 years ago by neoyipeng2018
  2

• 中文单个字符如何找到同义词的

  #2 opened 4 years ago by AlexYoung757
  4

• 请问计算损失函数时是不是只考虑被替换的token？

  #1 opened 4 years ago by RookieZB
  2