README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': cat
            '1': dog
            '2': elephant
            '3': fish
            '4': horse
            '5': lion
            '6': penguin
            '7': rabbit
            '8': tiger
  splits:
    - name: train
      num_bytes: 12734516340
      num_examples: 9000
    - name: test
      num_bytes: 6403984632
      num_examples: 4500
  download_size: 21172491982
  dataset_size: 19138500972
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*