---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  - name: watermark_flag
    dtype: bool
  splits:
  - name: train
    num_bytes: 40169778.0
    num_examples: 827
  download_size: 40163155
  dataset_size: 40169778.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---