README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: image_id
    dtype: int64
  - name: objects
    struct:
    - name: bbox
      sequence:
        sequence: float64
    - name: category
      sequence: int64
    - name: id
      sequence: int64
    - name: area
      sequence: float64
  splits:
  - name: train
    num_bytes: 24491919.0
    num_examples: 200
  - name: test
    num_bytes: 24491919.0
    num_examples: 200
  download_size: 48945938
  dataset_size: 48983838.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---