download_wikipedia_bert.py

from transformers import BertJapaneseTokenizer
from transformers import BertConfig
from transformers import BertForPreTraining

tokenizer = BertJapaneseTokenizer.from_pretrained('cl-tohoku/bert-base-japanese-whole-word-masking')
config = BertConfig().from_pretrained('cl-tohoku/bert-base-japanese-whole-word-masking')

tokenizer.save_pretrained('models/1-6_layer-wise')
config.save_pretrained('models/1-6_layer-wise')

tokenizer.save_pretrained('models/tapt512_60K')
config.save_pretrained('models/tapt512_60K')

tokenizer.save_pretrained('models/dapt128-tapt512')
config.save_pretrained('models/dapt128-tapt512')

# model = BertForPreTraining(config).from_pretrained('models/dapt128-tapt512')