tohoku-nlp
/

bert-base-japanese-char-whole-word-masking

system HF staff

Update tokenizer_config.json

ee17063 almost 4 years ago

No virus

110 Bytes

	{
	"do_lower_case": false,
	"word_tokenizer_type": "mecab",
	"subword_tokenizer_type": "character"
	}