README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: image_id
    dtype: int32
  - name: objects
    struct:
    - name: bbox
      sequence:
        sequence: float32
    - name: categories
      sequence: int64
  splits:
  - name: train
    num_bytes: 8694969.0
    num_examples: 100
  - name: validation
    num_bytes: 2547546.0
    num_examples: 30
  - name: test
    num_bytes: 1474594.0
    num_examples: 30
  download_size: 12491262
  dataset_size: 12717109.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---