---
dataset_info:
  features:
  - name: pixel_values
    sequence:
      sequence:
        sequence: float32
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 18266081316
    num_examples: 10293
  download_size: 363550076
  dataset_size: 18266081316
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---