---
dataset_info:
  features:
  - name: keypoints
    sequence:
      sequence:
        sequence: float64
  - name: caption
    dtype: string
  - name: class
    dtype: string
  splits:
  - name: train
    num_bytes: 1010455745
    num_examples: 6927
  - name: validation
    num_bytes: 101423566
    num_examples: 677
  download_size: 165940111
  dataset_size: 1111879311
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---