---
dataset_info:
  features:
  - name: tokens
    sequence: string
  - name: ner_tags
    sequence: int64
  splits:
  - name: train
    num_bytes: 127019
    num_examples: 915
  - name: validation
    num_bytes: 121393
    num_examples: 946
  - name: test
    num_bytes: 130972
    num_examples: 952
  download_size: 120493
  dataset_size: 379384
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
language: br
task_categories:
- token-classification
---

### Description
Version nettoyée de [WikiAnn](https://huggingface.co/datasets/tner/wikiann).  
En effet, la version originale contenait des leaks et des duplications.
De 1000 effectifs par split, la nouvelle répartition devient alors la suivante :

```
DatasetDict({
    train: Dataset({
        features: ['tokens', 'ner_tags'],
        num_rows: 915
    })
    validation: Dataset({
        features: ['tokens', 'ner_tags'],
        num_rows: 946
    })
    test: Dataset({
        features: ['tokens', 'ner_tags'],
        num_rows: 952
    })
})
```


### Label ID
Le dictionnaire label2id est disponible [ici](https://huggingface.co/datasets/tner/wikiann/raw/main/dataset/label.json).
```python
{
    "B-LOC": 0,
    "B-ORG": 1,
    "B-PER": 2,
    "I-LOC": 3,
    "I-ORG": 4,
    "I-PER": 5,
    "O": 6
}
```