---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  - name: extraction
    struct:
    - name: error
      dtype: string
    - name: forumala
      dtype: string
    - name: raw_output
      dtype: string
  - name: extracted_raw_output
    dtype: string
  - name: extracted_forumala
    dtype: string
  - name: extracted_error
    dtype: string
  splits:
  - name: train
    num_bytes: 47020.0
    num_examples: 10
  download_size: 55779
  dataset_size: 47020.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---