---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype: string
  splits:
  - name: clean
    num_bytes: 17988446.486335825
    num_examples: 736
  - name: noisy
    num_bytes: 253066270.0
    num_examples: 1560
  download_size: 265676260
  dataset_size: 271054716.4863358
configs:
- config_name: default
  data_files:
  - split: clean
    path: data/clean-*
  - split: noisy
    path: data/noisy-*
---