tohoku-nlp

/

bert-base-japanese-whole-word-masking

like 55

Is there any Fast Tokenizer method for Japanese BERT?
For large datasets, tokenization is failing due to memory error.