---
license: mit
language:
- pt
---
# bertimbau-large-ner-selective

This model card aims to simplify the use of the [portuguese Bert, a.k.a, Bertimbau](https://github.com/neuralmind-ai/portuguese-bert) for the Named Entity Recognition task. 

For this model card the we used the <mark style="background-color: grey"> **BERT-CRF (selective scenario, 5 classes)** </mark> model available in the [ner_evaluation](https://github.com/neuralmind-ai/portuguese-bert/tree/master/ner_evaluation) folder of the original Bertimbau repo.

Available classes are:

+ PESSOA
+ ORGANIZACAO
+ LOCAL
+ TEMPO
+ VALOR

## Usage

```
# Load model directly
from transformers import AutoTokenizer, AutoModelForTokenClassification

tokenizer = AutoTokenizer.from_pretrained("marquesafonso/bertimbau-large-ner-selective")
model = AutoModelForTokenClassification.from_pretrained("marquesafonso/bertimbau-large-ner-selective")

```

## Example

```
from transformers import pipeline

pipe = pipeline("ner", model="marquesafonso/bertimbau-large-ner-selective", aggregation_strategy='simple')

sentence = "Acima de Ederson, abaixo de Rúben Dias. É entre os dois jogadores do Manchester City que se vai colocar Gonçalo Ramos no ranking de vendas mais avultadas do Benfica."

result = pipe([sentence])

print(f"{sentence}\n{result}")

# Acima de Ederson, abaixo de Rúben Dias. É entre os dois jogadores do Manchester City que se vai colocar Gonçalo Ramos no ranking de vendas mais avultadas do Benfica.
# [[
#     {'entity_group': 'PESSOA', 'score': 0.99694395, 'word': 'Ederson', 'start': 9, 'end': 16},
#     {'entity_group': 'PESSOA', 'score': 0.9918462, 'word': 'Rúben Dias', 'start': 28, 'end': 38},
#     {'entity_group': 'ORGANIZACAO', 'score': 0.96376556, 'word': 'Manchester City', 'start': 69, 'end': 84},
#     {'entity_group': 'PESSOA', 'score': 0.9993823, 'word': 'Gonçalo Ramos', 'start': 104, 'end': 117},
#     {'entity_group': 'ORGANIZACAO', 'score': 0.9033079, 'word': 'Benfica', 'start': 157, 'end': 164}
# ]]
```

## Acknowledgements

This work is an adaptation of [portuguese Bert, a.k.a, Bertimbau](https://github.com/neuralmind-ai/portuguese-bert). You may check and/or cite their [work](http://arxiv.org/abs/1909.10649):

```
@InProceedings{souza2020bertimbau,
    author="Souza, F{\'a}bio and Nogueira, Rodrigo and Lotufo, Roberto",
    editor="Cerri, Ricardo and Prati, Ronaldo C.",
    title="BERTimbau: Pretrained BERT Models for Brazilian Portuguese",
    booktitle="Intelligent Systems",
    year="2020",
    publisher="Springer International Publishing",
    address="Cham",
    pages="403--417",
    isbn="978-3-030-61377-8"
}


@article{souza2019portuguese,
    title={Portuguese Named Entity Recognition using BERT-CRF},
    author={Souza, F{\'a}bio and Nogueira, Rodrigo and Lotufo, Roberto},
    journal={arXiv preprint arXiv:1909.10649},
    url={http://arxiv.org/abs/1909.10649},
    year={2019}
}
```

Note that the authors - Fabio Capuano de Souza, Rodrigo Nogueira, Roberto de Alencar Lotufo - have used an MIT LICENSE for their work.