---
dataset_info:
  features:
  - name: image1
    dtype: image
  - name: image2
    dtype: image
  - name: label
    dtype: string
  - name: variation
    dtype: string
  splits:
  - name: train
    num_bytes: 4338032
    num_examples: 400
  download_size: 909625
  dataset_size: 4338032
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---