README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype: image
  splits:
    - name: train
      num_bytes: 26409443345
      num_examples: 62899
  download_size: 25400492133
  dataset_size: 26409443345
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
task_categories:
  - image-segmentation
task_ids:
  - semantic-segmentation
size_categories:
  - 10K<n<100K
license: cc