yoheikikuta/bert-japanese

Issues

• Where to find your trained sentencepiece tokenizer model?

  #29 opened 8 months ago by mengyao00
  0

• Which lisence type does your pre-trained model belong to?

  #28 opened 3 years ago by yamazombie
  2

• 学習プログラムの再現性を確保する方法について

  #27 opened 3 years ago by FuDaoLiao
  1

• テキスト要約タスクにファインチューニングする方法

  #26 opened 3 years ago by plalomar
  2

• Cannot load checkpoint for pretraining

  #24 opened 4 years ago by dangne
  1

• do_predictフラグを使わない理由はありますか

  #22 opened 5 years ago by miyamonz
  4

• May I know the Specs of PC or cloud's instance this pretraining?

  #21 opened 5 years ago by luvwinnie
  4

• tokenizerのdo_lower_caseについて

  #20 opened 5 years ago by soneo1127
  5

• Pretrained modelsをKaggleへ登録しても大丈夫ですか

  #16 opened 5 years ago by vochicong
  3

• BERT vs sklearn, both using SentencePiece

  #11 opened 5 years ago by vochicong
  2

• Update from upstream (BERT)

  #12 opened 6 years ago by vochicong
  1

• finetune-to-livedoor-corpusのノートブックでValueError: test_size=3.001358 should be smaller than 1.0 or be an integerになる

  #10 opened 6 years ago by ycat3
  2

• Any experiment result that can conclude the sentence piece Japanese BERT model outperforms word piece one?

  #8 opened 6 years ago by weiczhu
  2

• WikiExtractor.py - No such file or directory

  #7 opened 6 years ago by Tomas0413
  6

• SentencePiece tokenizerについて

  #5 opened 6 years ago by lightondust
  2

• 【質問】BERTの中間層の重みベクトルの抽出

  #3 opened 6 years ago by tatz1101
  2

• finetune-to-livedoor-corpusのノートブックでkeyerrorになる

  #4 opened 6 years ago by lightondust
  3