---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    struct:
    - name: model_output
      dtype: string
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 2579486374.728
    num_examples: 12696
  - name: test
    num_bytes: 64511586.0
    num_examples: 300
  download_size: 2539877308
  dataset_size: 2643997960.728
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---