---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: labels
    sequence: int32
  - name: label_names
    sequence: string
  - name: output_text
    dtype: string
  splits:
  - name: train
    num_bytes: 2835926771.55
    num_examples: 119825
  download_size: 3140884292
  dataset_size: 2835926771.55
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---