---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: instruction
    dtype: string
  - name: parsed_data
    dtype: string
  splits:
  - name: train
    num_bytes: 307688397
    num_examples: 2043
  download_size: 240120615
  dataset_size: 307688397
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---