Downloads · 30 days
10
2% of all-time downloads
KcELECTRA_base_beep
koorukuroo/KcELECTRA_base_beep
KcELECTRA_base_beep is a text classification model from koorukuroo. Use it when you need a label for a piece of text. It is set up for transformers. The card lists the license as mit.
BEEP! 데이터셋으로 Epoch 10으로 파인튜닝하여 결과를 살펴보겠습니다.
All-time downloads
421
Public
Repo size
996 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin498 MB · 100%
At a glance
- Task
- Text Classification
- Library
- transformers
- License
- mit
- Model type
- electra
- Access
- Public
- Created
- Aug 6, 2022
- Updated
- May 28, 2023
- SHA
- 5a1f6a53
Try a prompt
Model card
From the Hugging Face model README
BEEP! 데이터셋으로 Epoch 10으로 파인튜닝하여 결과를 살펴보겠습니다.
| Loss | Acc | Prec | Rec | F1 | |
|---|---|---|---|---|---|
| TRAIN | 0.11 | 0.965 | 0.966 | 0.972 | 0.969 |
| VAL | 0.73 | 0.807 | 0.947 | 0.749 | 0.837 |
threshold 0.5 기준으로 구분하였을 때, dev 데이터셋에 대한 정확도는 0.85 입니다.
그리고 임베딩 결과물을 t-SNE로 시각화하여보았습니다.
https://v5.core.today/notebook/34XX0RYM4#KcELECTRA_base_beep.ipynb
model = Model.load_from_checkpoint(latest_ckpt);
def infer(x):
return torch.softmax(
model(**model.tokenizer(x, return_tensors='pt')
).logits, dim=-1)
infer('송중기 시대극은 믿고본다. 첫회 신선하고 좋았다.')
tensor([[0.7414, 0.2586]], grad_fn=<SoftmaxBackward>)
infer('유이 자연스러워진 연기')
tensor([[0.7627, 0.2373]], grad_fn=<SoftmaxBackward>)
- Task
- Text Classification
- Library
- transformers
- Type
- electra
- License
- mit
- Created
- Aug 6, 2022
- Updated
- May 28, 2023