---
dataset_info:
  features:
  - name: image_file
    dtype: string
  - name: image_tensor
    sequence:
      sequence:
        sequence: float32
  splits:
  - name: train
    num_bytes: 13031138681
    num_examples: 339733
  download_size: 68936618
  dataset_size: 13031138681
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---