|
--- |
|
language: |
|
- ko |
|
library_name: transformers |
|
tags: |
|
- roberta |
|
- korean |
|
- klue |
|
- korquad1 |
|
--- |
|
|
|
## korquad1-klue-roberta-large |
|
|
|
This is [roberta-large](https://huggingface.co/klue/roberta-large) model for question generation. <br> |
|
- Finetuned based on 'klue/roberta-large' |
|
- Finetuned with korquad 1.0 Dataset |
|
- [squad_kor_v1](https://huggingface.co/datasets/squad_kor_v1) |
|
|
|
## How to use |
|
NOTE: Use BertTokenizer instead of RobertaTokenizer. (AutoTokenizer will load BertTokenizer) |
|
|
|
```python |
|
from transformers import AutoModel, AutoTokenizer |
|
|
|
model = AutoModel.from_pretrained("wooy0ng/korquad1-klue-roberta-large") |
|
tokenizer = AutoTokenizer.from_pretrained("wooy0ng/korquad1-klue-roberta-large") |
|
``` |
