---
dataset_info:
  features:
  - name: article_id
    dtype: string
  - name: text
    dtype: string
  - name: entity_mention
    dtype: string
  - name: start_offset
    dtype: int32
  - name: end_offset
    dtype: int32
  - name: main_role
    dtype: string
  - name: fine_roles
    list: string
  splits:
  - name: train
    num_bytes: 2181250
    num_examples: 686
  - name: test
    num_bytes: 477301
    num_examples: 91
  download_size: 497984
  dataset_size: 2658551
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---