---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: canny_edges_seg
    dtype: image
  - name: canny_edges_normal
    dtype: image
  - name: segmented_images
    dtype: image
  - name: caption
    dtype: string
  - name: captions
    sequence: string
  splits:
  - name: train
    num_bytes: 3649998086.5
    num_examples: 18159
  download_size: 3643253935
  dataset_size: 3649998086.5
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---