--- dataset_info: features: - name: image dtype: image - name: masked_image dtype: image - name: mask sequence: sequence: int64 splits: - name: train num_bytes: 3616985208.0 num_examples: 173 download_size: 472451235 dataset_size: 3616985208.0 configs: - config_name: default data_files: - split: train path: data/train-* ---