hitachi-nlp/bert-base-japanese_sudachi-unigram

Japanese BERT-base (Sudachi + Unigram)

How to load the tokenizer

Please download the dictionary file for Sudachi + Unigram from our GitHub repository. Then you can load the tokenizer by specifying the path of the dictionary file to dict_path.

from typing import Optional

from tokenizers import Tokenizer, NormalizedString, PreTokenizedString
from tokenizers.processors import BertProcessing
from tokenizers.pre_tokenizers import PreTokenizer
from transformers import PreTrainedTokenizerFast

from sudachipy import tokenizer
from sudachipy import dictionary
import textspan

class SudachiPreTokenizer:
    def __init__(self, mecab_dict_path: Optional[str] = None):
        self.sudachi = dictionary.Dictionary().create()
    
    def tokenize(self, sequence: str) -> list[str]:
        return [token.surface() for token in self.sudachi.tokenize(sequence)]
    
    def custom_split(self, i: int, normalized_string: NormalizedString) -> list[NormalizedString]:
        text = str(normalized_string)
        tokens = self.tokenize(text)
        tokens_spans = textspan.get_original_spans(tokens, text)
        return [normalized_string[st:ed] for cahr_spans in tokens_spans for st,ed in cahr_spans]
    
    def pre_tokenize(self, pretok: PreTokenizedString):
        pretok.split(self.custom_split)

# load a pre-tokenizer
pre_tokenizer = SudachiPreTokenizer()

# load a tokenizer
dict_path = /path/to/sudachi_unigram.json
tokenizer = Tokenizer.from_file(dict_path)
tokenizer.post_processor = BertProcessing(
    cls=("[CLS]", tokenizer.token_to_id('[CLS]')),
    sep=("[SEP]", tokenizer.token_to_id('[SEP]'))
)

# convert to PreTrainedTokenizerFast
tokenizer = PreTrainedTokenizerFast(
    tokenizer_object=tokenizer,
    unk_token='[UNK]',
    cls_token='[CLS]',
    sep_token='[SEP]',
    pad_token='[PAD]',
    mask_token='[MASK]'
)

# set a pre-tokenizer
tokenizer._tokenizer.pre_tokenizer = PreTokenizer.custom(pre_tokenizer)

# Test
test_str = "こんにちは。私は形態素解析器について研究をしています。"
tokenizer.convert_ids_to_tokens(tokenizer(test_str).input_ids)
# -> ['[CLS]','こんにち','は','。','私','は','形態','素','解','析','器','に','つい','て','研究','を','し','て','い','ま','す','。','[SEP]']

How to load the model

from transformers import AutoModelForMaskedLM
model = AutoModelForMaskedLM.from_pretrained("hitachi-nlp/bert-base_sudachi-unigram")

See our repository for more details!