---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': duckling
          '1': rubber duck
          '2': unlabeled
  splits:
  - name: train
    num_bytes: 10929870.244
    num_examples: 1169
  download_size: 10859997
  dataset_size: 10929870.244
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---