---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: bbox
    sequence:
      sequence: float32
  - name: categories
    dtype: string
  splits:
  - name: train
    num_bytes: 18180556733.55
    num_examples: 23530
  download_size: 17703783169
  dataset_size: 18180556733.55
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---