---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: objects
    struct:
    - name: Ids
      sequence: string
  - name: captions
    sequence: string
  - name: quantities
    sequence: int64
  splits:
  - name: train
    num_bytes: 1221170619.216
    num_examples: 20244
  download_size: 1142033992
  dataset_size: 1221170619.216
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---