---
dataset_info:
  features:
  - name: questionId
    dtype: int64
  - name: question
    dtype: string
  - name: question_types
    dtype: string
  - name: image
    dtype: string
  - name: docId
    dtype: int64
  - name: ucs_document_id
    dtype: string
  - name: ucs_document_page_no
    dtype: int64
  - name: answers
    dtype: string
  splits:
  - name: train
    num_bytes: 8473652
    num_examples: 8
  download_size: 1996216
  dataset_size: 8473652
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---