codertimo/BERT-pytorch

Issues

• IndexError: list index out of range

  #52 opened 5 years ago
  13

• What dataset did you use to train model?

  #99 opened 2 years ago by Coding-Child
  2

• Why Segment Embedding number only 3?

  #106 opened 7 months ago by UTimeStrange
  1

• IndexError

  #88 opened 3 years ago by LemonQC
  6

• Clarification on Padding Process in BERT Model Construction

  #107 opened 7 months ago by AliHaiderAhmad001
  0

• Question about the loss of Masked LM

  #49 opened 6 years ago by zhezhaoa
  5

• 请问训练得到的模型后缀为.model.ep*格式，应该如何进行后续的调用呢？

  #103 opened a year ago by Jwinre
  0

• GELU is available in PyTorch

  #102 opened a year ago by hubin-keio
  1

• dataset / dataset.py have one erro?

  #93 opened 3 years ago by ndn-love
  1

• why specify `ignore_index=0` in the NLLLoss function in BERTTrainer?

  #98 opened 2 years ago by Jasmine969
  1

• how to fine tune model with trained weight

  #100 opened a year ago by MingchangLi
  0

• Why not use torch.no_grad when evaluating test data?

  #92 opened 3 years ago by EvanZ
  1

• embedding/position.py with RuntimeError if d_model is odd

  #78 opened 4 years ago by bookong22
  2

• How can I finetune with this project

  #61 opened 5 years ago by Gpwner
  5

• bert-vocab?

  #96 opened 2 years ago by nlpCSir
  1

• Pooler layer?

  #95 opened 2 years ago by lihaoxin2020
  0

• It keeps trying to use CUDA despite --with_cuda False option

  #94 opened 3 years ago by hanyangii
  0

• will "random.randint(self.corpus_lines if self.corpus_lines < 1000 else 1000)" report error?

  #55 opened 5 years ago by huangnengCSU
  1

• Did you have tried to use ipdb to set breakpoint in __get__item__, which is a funtion from BERTDataset

  #71 opened 5 years ago by moomooda
  1

• why language_model.py has different vectors

  #91 opened 3 years ago by zysNLP
  1

• The exact English pretraining data and Chinese pretraining data that are exact same to the BERT paper's pretraining data.

  #89 opened 3 years ago by guotong1988
  0

• An error occurred【AttributeError: type object 'BERT' has no attribute 'hidden'】

  #87 opened 4 years ago by XueqiangF
  0

• In Next Sentence Prediction task,the original code may choose the same line when you try to use the negative sample

  #86 opened 4 years ago by Emir-Liu
  0

• what does \t be used for here

  #58 opened 5 years ago by junchen14
  1

• .

  #85 opened 4 years ago
  0

• transformer.py 中的forword方法调用的SublayerConnection类。实现残差链接和标准化的实现

  #83 opened 4 years ago by dshwei
  1

• PositionalEmbedding

  #53 opened 5 years ago by fgaoyang
  3

• how to do Ner

  #82 opened 4 years ago by DeShuiYu
  0

• Default model sizes are much smaller than BERT base

  #81 opened 4 years ago by bertmaher
  0

• How to implement model once pretrained for masked input sentences?

  #80 opened 4 years ago by sfaux
  0

• How to use this bert model to load and finetune ?

  #79 opened 4 years ago by dongrixinyu
  0

• How to Output Embedded Sentence Vector?

  #76 opened 4 years ago by rhypowang
  2

• How to Output Embedded Word Vector

  #68 opened 5 years ago by enze5088
  6

• If I want to use /u as a placeholder instead of /t, what do I need to do

  #75 opened 4 years ago by 0GSC0-0
  0

• would you provide some sample datasets for demo the pre-training

  #59 opened 4 years ago by SeekPoint
  9

• How to use BERT model to fine-tune a cloze-style task?

  #74 opened 4 years ago by OrchidXu
  0

• Can you share me a trained data?

  #73 opened 4 years ago by vcbeaut
  0

• self.d_k = d_model // h gives 64 dimension ?

  #60 opened 5 years ago by BerenLuthien
  1

• predict next sentence ony? it also can do SQA assignments?

  #72 opened 5 years ago by cqray1990
  0

• # of parameter is larger than bert base instead im using specs less than bert base

  #70 opened 5 years ago by MohamedLotfyElrefai
  0

• Vocab's load_vectors seems to be an old method?

  #69 opened 5 years ago by JamesOwers
  0

• ONNX conversion: TransformerBlock problem

  #67 opened 5 years ago by jbmaxwell
  0

• How to test your model after training on my own dataset?

  #57 opened 5 years ago by riktimmondal
  1

• Erroneous Code

  #65 opened 5 years ago by LeoLai930603
  0

• OOM error in cuda while passing large corpus of wikipedia text files ? how to manage big files to train

  #64 opened 5 years ago by MohamedLotfyElrefai
  0

• [BERT] Cannot import bert

  #51 opened 5 years ago by logicmd
  2

• chooses 15% of token

  #56 opened 5 years ago by makcedward
  1

• made a script to generate bert pre-train data

  #50 opened 6 years ago by fucusy
  0

• What should be the shapes and example of values of x and segment_info in BERT.forward?

  #48 opened 6 years ago by komelianchuk
  0

• ner

  #47 opened 6 years ago by sz128
  0