fastnlp/CPT

Issues

• A detailed pre-train process experiment

  #88 opened 16 days ago by shivanraptor
  0

• Does the model have the ability to summarize text?

  #87 opened a month ago by YANCC139
  0

• 为什么没有总结摘要功能？

  #86 opened a month ago by YANCC139
  0

• Pre-trained model raises IndexError: index out of range in self

  #85 opened a month ago by shivanraptor
  0

• run_gen.py returns UnicodeDecodeError

  #84 opened a month ago by shivanraptor
  1

• run_gen.py shows error: module 'tensorflow._api.v2.compat.v2.__internal__' has no attribute 'register_load_context_function'

  #83 opened 2 months ago by shivanraptor
  2

• run_pretrain_bart.sh returns IndexError

  #82 opened 2 months ago by shivanraptor
  13

• sh run_pretrain_bart.sh 报错

  #69 opened a year ago by jackychancjcjcj
  1

• What are the data formats of `dataset` and `vocab` folder?

  #81 opened 3 months ago by shivanraptor
  2

• Why will there be blank between chinese characters? How to fix it?

  #75 opened a year ago by SupetZYK
  1

• from modeling_cpt import CPTForConditionalGeneration 出错

  #74 opened a year ago by sumafuture
  2

• can I change the max length of csl

  #63 opened 6 months ago by Dlinuigh
  2

• 直接运行python run_gen.py，GPU 利用率不到50%

  #80 opened 9 months ago by runningabcd
  3

• 使用 CPT 进行微调后，生成任务推理很慢，一条数据都要好几秒，如何提升推理速度？

  #79 opened 9 months ago by runningabcd
  1

• 关于pretraining时denoising的问题

  #71 opened a year ago by tu2022
  3

• bart-base-chinese支持 lightseq等推理加速不

  #51 opened 2 years ago by ImmortalZ
  2

• CPT多GPU卡finetuning训练报错

  #58 opened 2 years ago by aidejieceng
  5

• TypeError: forward() got an unexpected keyword argument 'labels'

  #77 opened 10 months ago by runningabcd
  3

• 如何基于 CPT 进行 Pretrain

  #78 opened 10 months ago by runningabcd
  0

• Pytorch model to ONNX, embedding layer parameters are misalignment?

  #76 opened a year ago by smygreat
  0

• bug : got an unexpected keyword argument 'tokenizer_class'

  #73 opened a year ago by xuefeich
  1

• 看起来cpt和bart效果差不太多？

  #72 opened a year ago by xuefeich
  1

• CPTForConditionalGeneration使用多GPU报错

  #46 opened 2 years ago by Biaocsu
  5

• finetune时，怎么能每训练一个epoch，保存一次模型

  #70 opened a year ago by deepkun
  0

• modeling_cpt.py问题请教

  #68 opened a year ago by LKAMING97
  1

• bart-base模型ft之前preprocess的方法是什么呢？

  #60 opened 2 years ago by fubuki75
  1

• Can I finetune CPTForMaskedLM?

  #67 opened 2 years ago by bmkor
  4

• token_type_ids报错

  #64 opened 2 years ago by longborn
  1

• How to fine-tune the Chinese BART model?

  #61 opened 2 years ago by AlexNLP
  2

• CPT pretrain problem

  #66 opened 2 years ago by SunyanGu
  3

• unrecognized arguments: --local_rank=1

  #65 opened 2 years ago by SunyanGu
  1

• TypeError: add_code_sample_docstrings() got an unexpected keyword argument 'tokenizer_class'

  #62 opened 2 years ago by HanZijie
  0

• How much corpus have you used?

  #59 opened 2 years ago by TankAbraham
  1

• CSL摘要数据集没达到效果。

  #44 opened 2 years ago by josephcui
  6

• 推理时的token_type_ids

  #56 opened 2 years ago by 1245244103
  1

• ModuleNotFoundError: No module named 'amp_C'

  #57 opened 2 years ago by aidejieceng
  1

• 你好，想请教下微调中dev.json数据的结构问题？

  #52 opened 2 years ago by ImmortalZ
  2

• 生成模型预测出的结果跟输入一模一样？

  #53 opened 2 years ago by zhihao-chen
  0

• 有关fnlp/bart-base-chinese模型加载问题

  #55 opened 2 years ago by yedongyu1996
  2

• 请问做句子复述和文章摘要要调用哪些方法？

  #49 opened 2 years ago by estuday
  2

• 这个支持多机多卡训练吗？可以提供训练脚步👀一下吗？

  #47 opened 2 years ago by lanse-sir
  1

• index out of range in self

  #48 opened 2 years ago by OuYanXi
  1

• 生成相关问题

  #40 opened 2 years ago by TheHonestBob
  0

• 基于BART做机器翻译finetune

  #42 opened 2 years ago by codejeam
  0

• RuntimeError:add_code_sample_docstrings() got an unexpected keyword argument 'tokenizer_class'RuntimeError: Failed to import roformer.modeling_roformer because of the following error (look up to see its traceback): add_code_sample_docstrings() got an unexpected keyword argument 'tokenizer_class'

  #45 opened 2 years ago by Cyanash
  1

• 中文分词的数据问题

  #43 opened 2 years ago by Richar-Du
  0

• 不支持token_type_embedding吗

  #39 opened 2 years ago by songmzhang
  1

• 重载模型时vocab.txt的问题

  #54 opened 2 years ago by Guochry
  0

• 想咨询下，fnlp/bart-base-chinese 推理时 使用gpu该怎么调用吗

  #50 opened 2 years ago by ImmortalZ
  1

• CSL数据集

  #41 opened 2 years ago by josephcui
  1