---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: objects
    struct:
    - name: categories
      sequence: int64
    - name: bbox
      sequence:
        sequence: string
    - name: keypoints
      sequence:
        sequence:
          sequence: string
  splits:
  - name: train
    num_bytes: 319821477.0
    num_examples: 649
  download_size: 315819168
  dataset_size: 319821477.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
viewer: false
---