---
dataset_info:
  features:
  - name: tokens
    list: string
  - name: tags
    list:
      class_label:
        names:
          '0': O
          '1': B-PER
          '2': I-PER
          '3': B-ORG
          '4': I-ORG
          '5': B-LOC
          '6': I-LOC
  - name: langs
    list: string
  - name: spans
    list: string
  splits:
  - name: validation
    num_bytes: 2280489
    num_examples: 10000
  - name: test
    num_bytes: 2276892
    num_examples: 10000
  - name: train
    num_bytes: 4501856
    num_examples: 20000
  download_size: 2176280
  dataset_size: 9059237
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
  - split: train
    path: data/train-*
---

# Dataset Card for WikiANN TR

## Table of Contents
- [Dataset Description](#dataset-description)
  - [Dataset Structure](#dataset-structure)
  - [Data Fields](#data-fields)
- [Source Data](#source-data)

## Dataset Description

WikiANN is originally [released](https://huggingface.co/datasets/unimelb-nlp/wikiann) for 176 langauages. We used the `tr` subset of the dataset.

### Dataset Structure

We kept the original data structure.

### Data Fields

- **tokens** (list): a `list` of `string` features.
- **tags** (list): a `list` of classification labels, with possible values including `O` (0), `B-PER` (1), `I-PER` (2), `B-ORG` (3), `I-ORG` (4), `B-LOC` (5), `I-LOC` (6).

## Source Dataset

[hf.co/datasets/unimelb-nlp/wikiann](https://huggingface.co/datasets/unimelb-nlp/wikiann)