|
--- |
|
language: |
|
- "bo" |
|
tags: |
|
- "tibetan" |
|
- "masked-lm" |
|
base_model: Pagewood/Tibetan-BERT-wwm |
|
license: "apache-2.0" |
|
pipeline_tag: "fill-mask" |
|
mask_token: "[MASK]" |
|
--- |
|
|
|
# bert-base-tibetan |
|
|
|
## Model Description |
|
|
|
This is a BERT model pre-trained on Tibetan texts, derived from [Tibetan-BERT-wwm](https://huggingface.co/Pagewood/Tibetan-BERT-wwm) with the tokenizer refined for PyTorch. You can fine-tune `bert-base-tibetan` for downstream tasks, such as [POS-tagging](https://huggingface.co/KoichiYasuoka/bert-base-tibetan-upos), dependency-parsing, and so on. |
|
|
|
## How to Use |
|
|
|
```py |
|
from transformers import AutoTokenizer,AutoModelForMaskedLM |
|
tokenizer=AutoTokenizer.from_pretrained("KoichiYasuoka/bert-base-tibetan") |
|
model=AutoModelForMaskedLM.from_pretrained("KoichiYasuoka/bert-base-tibetan") |
|
``` |
|
|
|
|
