---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: is_panorama
    dtype: bool
  - name: prompt_context
    dtype: string
  - name: vlm_response
    dtype: string
  splits:
  - name: train
    num_bytes: 247841579
    num_examples: 478
  download_size: 244895894
  dataset_size: 247841579
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---