---
dataset_info:
  features:
  - name: image
    dtype: string
  - name: id
    dtype: string
  - name: conversations
    dtype: string
  splits:
  - name: train
    num_bytes: 4683526
    num_examples: 23785
  - name: test
    num_bytes: 977129
    num_examples: 5015
  download_size: 1360403
  dataset_size: 5660655
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---