iqiyi/FASPell

Issues

• 请问可以分享下 char_meta.txt 文件吗

  #19 opened 4 years ago by loserdog-err
  4

• 与pre-trained model 同时使用的fine-tuned model能传一份到google网盘吗？

  #41 opened 4 years ago by svjack
  3

• 在运行python faspell.py -m e -t -d时，数据越界

  #46 opened 3 years ago by blueseven77
  1

• 分享一个自己生成的特征文件 char_meta.txt

  #33 opened 4 years ago by Lijiachen1018
  24

• 验证数据集

  #48 opened 2 years ago by joyJZhang
  0

• 对于实验结果召回率不高的情况有什么优化建议吗

  #31 opened 5 years ago by 18782961008
  29

• 微调样本的用处是什么？如何使用？

  #44 opened 3 years ago by moyaojunshi
  1

• 有没有做对中英文同时纠错的道友，交流下呀

  #39 opened 4 years ago by AnddyWang
  2

• apted.jar文件在哪

  #43 opened 4 years ago by hudielv
  1

• 如何添加loss和性能打印

  #45 opened 3 years ago by moyaojunshi
  0

• 关于手动绘制曲线的疑惑

  #40 opened 4 years ago by ZJiaBin
  1

• CSD 部分 candidates 是不是只要存一次，我理解的对吗

  #34 opened 4 years ago by huitang
  2

• some queries for this code

  #42 opened 4 years ago by lmw0320
  0

• 可以分享一下训练好的模型数据吗

  #14 opened 4 years ago by cabbageNoob
  2

• 您好 在生成数据阶段，用命令行执行create_data和在pycharm中create_data执行为什么会得到不同的结果

  #32 opened 4 years ago by V-Enzo
  0

• 部分字没有字形

  #21 opened 4 years ago by SolinTao
  1

• ＠eugene-yh　用cjklib的getStrokeOrder将＇不＇字分解后得到的笔画在ｉｄs.txt中找不到对应的字符？

  #20 opened 4 years ago by wenston2006
  1

• char_meta.txt 问题

  #13 opened 4 years ago by lianNice
  11

• What are the token level results of SIGHAN13, SIGHAN14

  #12 opened 4 years ago by cklsoft
  2

• 執行 run_pretraining.py 未能成功

  #6 opened 4 years ago by Chunshan-Theta
  6

• 怎么避免不把简体修改成繁体

  #38 opened 4 years ago by aoleiReiz
  1

• anyone knows how to generate mask_probability.sav

  #36 opened 4 years ago by AnddyWang
  1

• 哪位能分享下解决extension，多个数字或字母连着 对齐的问题么

  #37 opened 4 years ago by AnddyWang
  0

• 数据对齐问题

  #28 opened 5 years ago by liqinc
  2

• 关于文本对齐的问题extension()

  #35 opened 4 years ago by ouranmo
  0

• ~

  #30 opened 5 years ago by nonva
  0

• 将"dump_candidates"设置为保存路径可以保存候选字符。

  #29 opened 5 years ago by 18782961008
  1

• 请教贴：fine-tune时processor如何编写，求赐教

  #27 opened 5 years ago by BigWavelet
  1

• stroke-level IDS的处理问题

  #5 opened 5 years ago by liangxiao
  7

• 提供的训练数据，unihan的拼音数据， ids数据都是繁体中文，没有简体中文

  #26 opened 5 years ago by BigWavelet
  1

• tackle_n_gram_bias中操作

  #22 opened 5 years ago by lijuan123
  7

• fine-tune训练时的数据格式问题

  #25 opened 5 years ago by BigWavelet
  1

• extension() bug

  #24 opened 5 years ago by rejae
  0

• NSP

  #23 opened 5 years ago by rejae
  2

• pre-trained model

  #4 opened 5 years ago by zhanghaoie
  1

• 交流

  #17 opened 5 years ago by will-wiki
  4

• 能给出测试数据 -SIGHAN15测试集的下载地址吗

  #18 opened 5 years ago by bigzhouj
  0

• faspelll.py， class SpellChecker()，def get_error() 的return全是None？？？？

  #16 opened 5 years ago
  0

• 请问correct.txt 和wrong.txt是什么格式？

  #15 opened 5 years ago by highinsky
  2

• FileNotFoundError: [Errno 2] No such file or directory: '/path/to/training/data/file'

  #10 opened 5 years ago by dazaohehe
  1

• hi，代码是不兼容tensorflow2.0.0吧

  #11 opened 5 years ago by dazaohehe
  1

• 汉字字形表示的问题

  #8 opened 5 years ago by t-kong
  1

• 关于训练lm的问题

  #9 opened 5 years ago by huyi1989
  1

• 如何绘制confidence-similariy graph

  #7 opened 5 years ago by XuJ
  3

• 链接中给的是预训练的数据，并不是训练完成的数据

  #3 opened 5 years ago by EidenRitto
  1

• 您好 请问有pytorch版本的么？

  #2 opened 5 years ago by orangetwo
  1