---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: caption
    dtype: string
  - name: image_bw
    dtype: image
  splits:
  - name: train
    num_bytes: 2222163393.0
    num_examples: 10000
  download_size: 2183756839
  dataset_size: 2222163393.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---