---
dataset_info:
  features:
  - name: words
    sequence: string
  - name: ner_tags
    sequence:
      class_label:
        names:
          '0': other
          '1': header
          '2': answer
          '3': question
  - name: bboxes
    sequence:
      sequence: int64
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 13232249.0
    num_examples: 149
  download_size: 12255331
  dataset_size: 13232249.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---