---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 993539858.2652969
    num_examples: 230000
  - name: test
    num_bytes: 18622392.734703023
    num_examples: 4311
  download_size: 1012091941
  dataset_size: 1012162251.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---