---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: images
    sequence: binary
  splits:
  - name: train
    num_bytes: 1263439358765
    num_examples: 3610797
  download_size: 1242613397611
  dataset_size: 1263439358765
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---