---
dataset_info:
  features:
  - name: image_id
    dtype: string
  - name: sentence_id
    dtype: int64
  - name: text
    dtype: string
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 57705952912.112
    num_examples: 360216
  download_size: 19887197751
  dataset_size: 57705952912.112
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---