README.md

metadata

dataset_info:
  features:
    - name: questionId
      dtype: int32
    - name: images
      dtype: image
    - name: plan_create
      dtype: string
    - name: plan_exec
      dtype: string
    - name: question
      dtype: string
    - name: answers
      sequence: string
    - name: query_anal
      dtype: string
    - name: final_ans
      dtype: string
  splits:
    - name: train
      num_bytes: 28102186548.359
      num_examples: 38653
  download_size: 7620067938
  dataset_size: 28102186548.359
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*