File size: 456 Bytes
04e79bc |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
---
language: ko
---
# Albert base model for Korean
* 70GB Korean text dataset and 42000 lower-cased subwords are used
* Check the model performance and other language models for Korean in [github](https://github.com/kiyoungkim1/LM-kor)
```python
from transformers import BertTokenizerFast, AlbertModel
tokenizer_albert = BertTokenizerFast.from_pretrained("kykim/albert-kor-base")
model_albert = AlbertModel.from_pretrained("kykim/albert-kor-base")
``` |