---
dataset_info:
  features:
  - name: en
    dtype: string
  - name: br
    dtype: string
  - name: en_image_clean
    dtype: image
  - name: br_image_clean
    dtype: image
  - name: en_image_aug
    dtype: image
  - name: br_image_aug
    dtype: image
  - name: metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 8805011749.0
    num_examples: 100000
  download_size: 8562511938
  dataset_size: 8805011749.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---