---
dataset_info:
  features:
  - name: URL
    dtype: image
  - name: conditioning
    dtype: image
  - name: caption
    dtype: string
  - name: hed
    dtype: image
  - name: thinning_hed
    dtype: image
  splits:
  - name: train
    num_bytes: 1368607116.376
    num_examples: 5984
  download_size: 1352205783
  dataset_size: 1368607116.376
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---