---
language:
- en
dataset_info:
  features:
  - name: id
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: predicate
    dtype: string
  - name: image
    dtype: image
  splits:
  - name: train
    num_bytes: 48169315.95619722
    num_examples: 1901
  download_size: 21388804
  dataset_size: 48169315.95619722
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---