linzehui/mRASP

Issues

• This error occurred while running preprocessing, no vocab found? vocab.bpe.600: No such file or directory :No such file or directory

  #20 opened 3 years ago by sammichenVV
  2

• 我将fairseq模型转为transformers模型

  #48 opened a year ago by TheHonestBob
  3

• 对多语言的预训练中，如何解决中文、英文、日语这些语言的分词工具指定

  #25 opened 3 years ago by zhangtianjack
  2

• 抱歉，实在是没有看懂fairseq和shell脚本，请教个问题

  #47 opened a year ago by TheHonestBob
  1

• 请问一下这个文件是如何得来的"dict.{}.txt".format(args.source_lang)

  #46 opened a year ago by TheHonestBob
  0

• Error merging override generation.print_alignment=False

  #44 opened 2 years ago by 10211412
  1

• AttributeError: 'MMapIndexedDataset' object has no attribute '_bin_buffer_mmap'

  #45 opened 2 years ago by Aliang-CN
  0

• How pre-trained model without fine-tuning works on specific language pair translations?

  #8 opened 4 years ago by zhangguanqun
  7

• 加载模型出错

  #43 opened 2 years ago by cherry899
  0

• 添加中文，日文分词器之后，无法加载预训练模型

  #35 opened 3 years ago by cc2017111
  7

• 预训练模型下载

  #42 opened 2 years ago by wangjialin97
  0

• 關於全局變數路徑?

  #41 opened 2 years ago by yangeryy
  0

• 数据集下载问题

  #34 opened 3 years ago by areaChun
  8

• 下载mRASP-PC32-6enc6dec.pt,使用德英数据微调，多头注意力计算报错

  #40 opened 2 years ago by MzY-1999
  2

• 新增vocabulary的格式

  #39 opened 2 years ago by wenHK
  0

• en-de在toy上复现问题

  #17 opened 4 years ago by leonodelee
  12

• Fine-tuning mRASP for Ar-En

  #38 opened 2 years ago by salma-elshafey
  0

• 数据集下载问题

  #37 opened 2 years ago by ShirleyLiu1020
  2

• 请问模型支持跨语言输入吗

  #36 opened 2 years ago by xuda1998
  1

• 微调英德数据集，执行最后的fine-tine.sh，报错KeyError：0

  #33 opened 3 years ago by S-GeGe
  0

• About 32-lang-pairs-TRAIN

  #19 opened 3 years ago by Mao-KU
  4

• 数据集

  #32 opened 3 years ago by leileilin
  0

• 代码中的sacrebleu的版本号是多少呀？

  #31 opened 3 years ago by S-GeGe
  2

• 关于pre-train的问题，这个阶段只能在单机单卡上运行，单机多卡hang住了

  #30 opened 3 years ago by Andrewlesson
  0

• 数据预处理一直跑不通

  #29 opened 3 years ago by xiuzhilu
  0

• 缺少文件 multilingual_preprocess_yml_generator.py ，跑不起来

  #3 opened 4 years ago by fuyazhou
  5

• 预训练数据预处理问题

  #28 opened 3 years ago by xiuzhilu
  0

• 关于报告的结果 tok or detok

  #27 opened 3 years ago by jingmu123
  1

• 您好，方便告知你们处理小语种的tokenizer吗？

  #26 opened 3 years ago by zhuyl96
  0

• 开发集询问

  #13 opened 4 years ago by jingmu123
  3

• Tabel1 结果咨询

  #14 opened 3 years ago by jingmu123
  2

• 是否支持蒙、藏、维等民族语翻译？

  #21 opened 3 years ago by HuihuiChyan
  3

• 韩语预处理？

  #22 opened 3 years ago by HuihuiChyan
  1

• 无法复现英德数据集上无微调的结果

  #11 opened 4 years ago by moonscar
  12

• 结果复现的问题

  #12 opened 4 years ago by jingmu123
  6

• 如何计算bleu分数？

  #24 opened 3 years ago by q178
  0

• About override issue

  #23 opened 3 years ago by jaehyoyoo
  0

• En-Fr微调模型checkpoint下载链接失效

  #16 opened 3 years ago by luxinyu1
  5

• All model download links are invalidated

  #18 opened 3 years ago by malizheng
  3

• 关于valid和test过程

  #15 opened 4 years ago by koukaiu
  2

• Validation set for the language pairs that you compare with mBART

  #9 opened 4 years ago by zliucr
  1

• 在论文中看到使用pre-train的模型进行多语言翻译，可以获得很好的效果。有个问题，如果是使用fairseq-interactive命令，同时使用pre-train模型来进行双语翻译时，元语言与目标语言需要用什么标识符标识吗？

  #7 opened 4 years ago by luluwangwang1989
  2

• fine-tune阶段词表及词嵌入矩阵与预训练模型不一致的问题

  #10 opened 4 years ago by duterscmy
  1

• 我在论文中看到，使用pre_train模型就可以达到不错的效果，那么使用pre_train模型来进行翻译时，输入是用什么表示来区分是哪两种语言的翻译的？

  #6 opened 4 years ago by luluwangwang1989
  0

• Does fine-tuned model use the same vocabulary as pre-trained model?

  #5 opened 4 years ago by zhangguanqun
  5

• Do you have dictionary for all languages or just part of them?

  #4 opened 4 years ago by SefaZeng
  1

• Arch type

  #1 opened 4 years ago by Bachstelze
  3

• confused about preprocess pipeline

  #2 opened 4 years ago by SefaZeng
  5