---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: language
    dtype: string
  - name: content_type
    dtype: string
  - name: quality_score
    dtype: float32
  - name: source_pdf
    dtype: string
  - name: page_number
    dtype: int32
  - name: company
    dtype: string
  splits:
  - name: train
    num_bytes: 722981639.0
    num_examples: 1000
  download_size: 245127089
  dataset_size: 722981639.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---