---
dataset_info:
  features:
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 21593789882.25
    num_examples: 1415750
  download_size: 20475948216
  dataset_size: 21593789882.25
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---