README.md

metadata

dataset_info:
  features:
    - name: id
      dtype: int32
    - name: seismic
      dtype: image
    - name: label
      dtype: image
  splits:
    - name: train
      num_bytes: 788711476
      num_examples: 250
    - name: valid
      num_bytes: 141944493
      num_examples: 45
  download_size: 930642408
  dataset_size: 930655969
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: valid
        path: data/valid-*