kamalkraj/BERT-NER

Issues

• The C++ version inference speed is 10 times slower than python version. Is that normal or if I did something wrong?

  #65 opened 4 years ago by gycg
  1

• C++ predict

  #79 opened 4 years ago by shenfuli
  3

• Understanding the Evaluation Code

  #98 opened 3 years ago by Vincent950129
  1

• How can i use this project in Chinese NER?

  #97 opened 4 years ago by helloWorldLZ
  1

• Model training does not work on CPU

  #99 opened 3 years ago by saurabhhssaurabh
  1

• how convert bin to two part model?

  #103 opened 3 years ago by legnaxz
  1

• How to predict on test dataset after training?

  #104 opened 3 years ago by zhiyuanpeng
  1

• How to create pretrained model?

  #63 opened 4 years ago by heeh
  3

• Error index out of range in self when trying to predict for text of close to 5000 characters

  #102 opened 3 years ago by sakshams1990
  1

• After training and saved the models, I got a valid accuracy, while got an error(bad) result based on loading the saved model.

  #95 opened 4 years ago by jiangyiqiao
  1

• Segment_id and valid_ids

  #75 opened 4 years ago by AnkusManish
  1

• About label_mask and valid_ids

  #61 opened 5 years ago by Lapis-Hong
  1

• How to convert all cpp,header files to DLL file?

  #100 opened 3 years ago by yugaljain1999
  1

• How to use it in Chinese？

  #62 opened 5 years ago by Ian-peace
  1

• Key error 0 on evaluation set

  #90 opened 4 years ago by bollampa
  1

• Reproduce CoNLL results

  #91 opened 4 years ago by ghaddarAbs
  2

• How to show only the keywords in inference?

  #96 opened 4 years ago by ShivanshuPurohit
  0

• Pre-processing steps

  #94 opened 4 years ago by gabrielosluz
  0

• Train your own model (colab)

  #93 opened 4 years ago by ivankrstev7
  0

• How can i make fine tuning with new entities/labels?

  #88 opened 4 years ago by matteopulega
  1

• RuntimeError : during model.predict()

  #92 opened 4 years ago by yeonsookKwak
  1

• Why do we need [PAD] as a valid sequence?

  #73 opened 4 years ago by minstar
  2

• CUDA Runtime Error: Which Cuda version is compatible to run NER task using BERT-NER

  #89 opened 4 years ago by veilupt
  0

• KeyError: ''

  #87 opened 4 years ago by Rajmehta123
  1

• How can we use model BertForTokenClassification for lengthy sentences?

  #77 opened 4 years ago by Swty13
  4

• Detokenizing the words

  #86 opened 4 years ago by Rajmehta123
  1

• mBERT for multilingual NER

  #67 opened 4 years ago by pascalhuszar
  1

• small error in the source code

  #85 opened 4 years ago by wont
  1

• What is the data your model trained on?

  #84 opened 4 years ago by entslscheia
  2

• out_!x

  #81 opened 4 years ago by jx669
  1

• problem loading model in Deploy Rest-API

  #83 opened 4 years ago by jx669
  0

• Ner("out_!x")

  #82 opened 4 years ago by jx669
  1

• About fine-tuning

  #76 opened 4 years ago by darwonsamal
  1

• Can this model be trained on bert large?

  #74 opened 4 years ago by Umang9427
  4

• how many epochs do ner need

  #80 opened 4 years ago by HongyanJiao
  2

• is this using CoNll dataset for training NER tag ? dose the code support ontonotes as well?

  #58 opened 5 years ago by AjitAntony
  6

• inconsistency between GPU/CPU inference

  #78 opened 4 years ago by ntedgi
  2

• Problems with installing req

  #69 opened 4 years ago by pascalhuszar
  1

• model training on GPU fails

  #72 opened 4 years ago by shansmann
  1

• Cannot download models

  #70 opened 4 years ago by askonivala
  1

• Getting great eval results but while running api on test data all the predictions are 'O'

  #68 opened 4 years ago by ranjeetds
  1

• Difference in unpadded length of `input_ids` and `label_ids`

  #66 opened 4 years ago by vardaan123
  0

• Seems that F1 score of self trained model based on bert-base-uncased is unresonable

  #64 opened 4 years ago by yangheng95
  2

• c++ inference

  #60 opened 5 years ago by simonefrancia
  8

• Evaluate does not work for custom dataset

  #56 opened 5 years ago by imayachita
  11

• Evaluating on the complete dev/test dataset by loading a model. Also, maybe some issues with Bert_base_uncased tokenization in Ner class in bert.py ??

  #57 opened 5 years ago by vikas95
  11

• Inference for C++ on macOS

  #59 opened 5 years ago by loretoparisi
  7

• Loading Bert

  #55 opened 5 years ago by pierre-zhao
  3

• suggestion for own dataset to train on pretrained model.

  #54 opened 5 years ago by vigneshgig
  6

• About valid_output

  #53 opened 5 years ago by Anikacyp
  4