yongsun-yoon's picture
Update README.md
a40e20f
|
raw
history blame
633 Bytes
---
language:
- en
---
# Open Named Entity Recognition (English)
You can read the details in [this post](https://medium.com/@yongsun.yoon/cross-encoder-for-open-named-entity-recognition-4a7d485c37cc).
```python
from transformers import pipeline
nlp = pipeline('token-classification', 'yongsun-yoon/deberta-v3-base-open-ner', aggregation_strategy='simple')
text = 'Heat the olive oil in a frying pan, add the onion and cook for 5 minutes until softened and starting to turn golden. Set aside.'
entity_type = 'ingredient'
input_text = f'{text}{nlp.tokenizer.sep_token}{entity_type}' # [SEP]
nlp(input_text) # olive oil, onion
```