---
dataset_info:
  features:
  - name: images_1
    sequence: image
  - name: images_2
    sequence: image
  - name: task
    dtype: string
  - name: success
    dtype: string
  splits:
  - name: train
    num_bytes: 5483012325.112
    num_examples: 10248
  - name: validation
    num_bytes: 552935143.0
    num_examples: 2022
  download_size: 5762427124
  dataset_size: 6035947468.112
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---