---
dataset_info:
  features:
  - name: transform_img
    sequence:
      sequence:
        sequence: float32
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 97826520
    num_examples: 10015
  download_size: 23817407
  dataset_size: 97826520
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---