File size: 1,006 Bytes
ac49e62 b358aa4 ac49e62 b358aa4 ac49e62 fe703e7 22b1607 ac49e62 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
#### How to use
You can use this model with Transformers *pipeline* for NER.
```python
from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline
tokenizer = AutoTokenizer.from_pretrained("Holako/NER_model_holako")
model = AutoModelForTokenClassification.from_pretrained("Holako/NER_model_holako")
nlp = pipeline("ner", model=model, tokenizer=tokenizer)
example = "اسمي احمد"
ner_results = nlp(example)
print(ner_results)
```
#### Limitations and bias
This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains.
=======
#### Limitations and bias
This model is limited by its training dataset of entity-annotated news articles from a specific span of time. This may not generalize well for all use cases in different domains.
## Training data
Language|Dataset
-|-
Arabic | [ANERcorp](https://camel.abudhabi.nyu.edu/anercorp/)
|