README.md

metadata

language: ko

KoBigBird

Pretrained BigBird Model for Korean (kobigbird-bert-base)

About

BigBird, is a sparse-attention based transformer which extends Transformer based models, such as BERT to much longer sequences.

BigBird relies on block sparse attention instead of normal attention (i.e. BERT's attention) and can handle sequences up to a length of 4096 at a much lower compute cost compared to BERT. It has achieved SOTA on various tasks involving very long sequences such as long documents summarization, question-answering with long contexts.

Model is warm started from Korean BERT’s checkpoint.

How to use

WARN: Please use BertTokenizer instead of BigBirdTokenizer.

from transformers import AutoModel, AutoTokenizer

# by default its in `block_sparse` mode with num_random_blocks=3, block_size=64
model = AutoModel.from_pretrained("monologg/kobigbird-bert-base")

# you can change `attention_type` to full attention like this:
model = AutoModel.from_pretrained("monologg/kobigbird-bert-base", attention_type="original_full")

# you can change `block_size` & `num_random_blocks` like this:
model = AutoModel.from_pretrained("monologg/kobigbird-bert-base", block_size=16, num_random_blocks=2)

tokenizer = AutoTokenizer.from_pretrained("monologg/kobigbird-bert-base")
text = "한국어 BigBird 모델을 공개합니다!"
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)