README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: mask
      dtype:
        image:
          mode: L
    - name: image_id
      dtype: string
    - name: width
      dtype: int32
    - name: height
      dtype: int32
    - name: num_annotations
      dtype: int32
  splits:
    - name: train
      num_bytes: 1933763
      num_examples: 129
  download_size: 1948584
  dataset_size: 1933763
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*