|
|
--- |
|
|
language: |
|
|
- "bo" |
|
|
tags: |
|
|
- "tibetan" |
|
|
- "masked-lm" |
|
|
license: "cc-by-sa-4.0" |
|
|
pipeline_tag: "fill-mask" |
|
|
mask_token: "[MASK]" |
|
|
datasets: |
|
|
- UTibetNLP/tibetan_news_classification |
|
|
--- |
|
|
|
|
|
# roberta-base-tibetan |
|
|
|
|
|
## Model Description |
|
|
|
|
|
This is a RoBERTa model pre-trained on Tibetan texts. NVIDIA A100-SXM4-40GB took 40 hours 44 minutes for training. You can fine-tune `roberta-base-tibetan` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/roberta-base-tibetan-upos), dependency-parsing, and so on. |
|
|
|
|
|
## How to Use |
|
|
|
|
|
```py |
|
|
from transformers import AutoTokenizer,AutoModelForMaskedLM |
|
|
tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/roberta-base-tibetan") |
|
|
model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/roberta-base-tibetan") |
|
|
``` |
|
|
|
|
|
|
