---
dataset_info:
  features:
  - name: file_name
    dtype: image
  - name: text
    dtype: string
  - name: reverse_text
    dtype: string
  splits:
  - name: train
    num_bytes: 1905381123.305
    num_examples: 55195
  download_size: 1718410544
  dataset_size: 1905381123.305
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---