README.md

---
dataset_info:
  features:
  - name: image_id
    dtype: string
  - name: image
    dtype: image
  - name: objects
    struct:
    - name: bbox
      sequence:
        sequence: float32
    - name: category_id
      sequence: int64
    - name: category_name
      sequence: string
  splits:
  - name: train
    num_bytes: 64297216.0
    num_examples: 264
  - name: test
    num_bytes: 16074304.0
    num_examples: 66
  download_size: 80350758
  dataset_size: 80371520.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---