Downloads · 30 days
20
43% of all-time downloads
tommyjin/xlm-roberta-base-finetuned-panx-all-langs
xlm-roberta-base-finetuned-panx-all-langs is a token classification model from tommyjin. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
Downloads · 30 days
20
43% of all-time downloads
All-time downloads
46
Public
Parameters
277M
1.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors1.1 GB · 98%
From the Hugging Face model README
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
"트랜스 포머를 활용한 자연어 처리" O'Reilly 책을 활용하였습니다.
xlm-roberta-base 를 이용하여 파인 튜닝을 한 객체명 인식 입니다.
각 언어별 f1_score<br> ko : f1_score = 0.8611821192789028<br> en : f1_score = 0.7868391074180795<br> ja : f1_score = 0.6440401846320934<br> es : f1_score = 0.8533862565120316<br>
코드는 책을 활용 및 참고 하였습니다.
XTREME 데이터 셋을 이용하였습니다. 특히 PAN-X 데이터 셋을 이용했습니다.
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | F1 Score |
|---|---|---|---|---|
| 0.5093 | 1.0 | 917 | 0.3530 | 0.7679 |
| 0.309 | 2.0 | 1834 | 0.3101 | 0.8029 |
| 0.2176 | 3.0 | 2751 | 0.3089 | 0.8140 |